Innovative technologies for NZEBs: an energy and economic analysis tool and a case study of a non-residential building for the Mediterranean climate

ItM

A

a

b

a

ARRAA

KBtBPN

1

tsotetiaar

h0

Energy and Buildings 121 (2016) 318–343

Contents lists available at ScienceDirect

Energy and Buildings

j ourna l ho me pa g e: www.elsev ier .com/ locate /enbui ld

nnovative technologies for NZEBs: An energy and economic analysisool and a case study of a non-residential building for the

editerranean climate

nnamaria Buonomanoa,∗, Giuseppina De Lucab, Umberto Montanaroa, Adolfo Palomboa

DII – University of Naples Federico II, P.le Tecchio, 80, 80125 Naples, ItalyDI – University of Naples Parthenope, Centro Direz. Is. C4, 80143 Naples, Italy

r t i c l e i n f o

rticle history:eceived 31 March 2015eceived in revised form 27 June 2015ccepted 20 August 2015vailable online 22 August 2015

eywords:uilding integrated energy savingsechnologiesuilding performance dynamic simulationsarametric analysisZEB

a b s t r a c t

Several new technologies can be today implemented in buildings in order to achieve the NZEB goal.In this paper a novel computer model for predicting the energy demand of buildings integrating phasechange materials, photovoltaic-thermal collectors, adjacent sunspaces and innovative daylighting controlis presented. Through this tool, DETECt 2.2, written in MatLab and conceived for research purposes, theoverall energy and economic performance of multi-zone NZEBs can be assessed. Both the active andpassive effects on the energy demands of all the above mentioned technologies, even if simultaneouslyutilized, are taken into account by means of an integrated building modelling approach. In addition,parametric and sensitivity analyses, with a single simulation run, can be carried out for design purposes.

A novel relevant case study referred to the energy design of a non-residential NZEB for Mediterraneanclimates is developed. For this building a suitable energy optimization analysis was also carried out.For each use of the indoor space, the optimal value of the pivotal design and operating parameters iscalculated. Details about the optimal position of building PCMs and thermal insulation layers, also coupled
to BIPV and/or BIPV/T systems, are provided. For the obtained best configuration very low heating andcooling demands are achieved (0.9 and 1.5 kWh/m3 y, respectively). Results about a simplified economicanalysis carried out on the investigated energy saving technologies are also reported. At last, new NZEBdefinition details and criteria are provided for non-residential buildings located in the southern Europeanzones (Mediterranean climates).
© 2015 Elsevier B.V. All rights reserved.

. Introduction

The growing concern about energy efficiency in the building sec-or has led the research interest towards the use of energy-efficienttrategies to be incorporated into the design, construction, andperation of new buildings and undertaking retrofits to improvehe efficiency of existing buildings. For this purpose, an impressiveffort has been done worldwide by the research community and byhe governments in order to promote the use of an efficient build-ng design and of a suitable building integration of energy-efficient
nd renewable energy technologies (considered as key points forchieving the goal of Nearly Zero Energy Buildings, NZEBs). Theecast of the European Directive on the Energy Performance of
∗ Corresponding author.E-mail address: [email protected] (A. Buonomano).

ttp://dx.doi.org/10.1016/j.enbuild.2015.08.037378-7788/© 2015 Elsevier B.V. All rights reserved.

Building (EPBD) imposes the adoption of several measures forimproving the energy efficiency in buildings in order to achieveNZEB for all new buildings beyond 2020 [1].

Today, a standard definition of NZEBs (e.g. concerning therelated minimum energy requirements) and/or reference NZEBs(for different building categories) is still unavailable. For this reasonthe research community is called to help for setting suitable NZEBdefinitions, to be shaped on global scale. Such definitions will bevery useful for improving the development of NZEBs. One exampleis the EPBD fulfilment. In particular, Member States are requiredto draw up specifically designed national plans for increasing inthe next years the number of NZEBs [2,3]. In order to shape effec-tive NZEB definitions, energy balance analyses, indoor comfort level
assessments, life cycle investigations and economic analyses mustbe carried out [4–6]. Among them, the energy balance analysis out-lines the primary NZEB purpose and it is considered as the basicrequest for its achievement [7–9]. Such analysis must be performed
dx.doi.org/10.1016/j.enbuild.2015.08.037

http://www.sciencedirect.com/science/journal/03787788

http://www.elsevier.com/locate/enbuild

http://crossmark.crossref.org/dialog/?doi=10.1016/j.enbuild.2015.08.037&domain=pdf

mailto:[email protected]

dx.doi.org/10.1016/j.enbuild.2015.08.037

A. Buonomano et al. / Energy and Buildings 121 (2016) 318–343 319

Nomenclature

A heat exchange surface area (m2)C thermal capacitance (J/K)Cu room utilization coefficientc specific heat (J/kg/K)L illuminance (lux)E energy (kW h)f external surface view factorG Gebhart’s matrixh convective heat transfer coefficient (W/m2 K)K glazing extinction coefficient (m−1)K� Incident Angle ModifiedI solar radiation flux (W/m2)mCO2 CO2 emission or saving (kg)m mass flow rate (kg/s)n index of refractionP electrical power (W)Q thermal load (W)R thermal resistance (K/W)T temperature (K)t time (s)x PV/T flow length (m)w width (m)

Greeks letters absorption factor

ˇref temperature coefficientı thickness (m)ε emissivityςv equivalent visible transmission coefficient� efficiencyϑi incident angle of the sun (◦)ϑr angle of refraction (◦)� Stefan-Boltzmann constant (5.67·10−8 W/m2/K4)� transmittance

SubscriptsAC referred to the HVCAC systemair PVT airc photovoltaic cellel electricityg internal gaingas natural gasgl glassin indoor airins insulationis indoor horizontal surfaceinv inverterlm lightingmod photovoltaic moduleout outdoor airPCM phase change materialref referencesky sky vaultsp set points PVT air supplied to the buildingtd tedlarv ventilation

Superscripts

conv convectioneq equivalentext external

cond conduction

int internal

also by investigating the building integration of different innovativeenergy efficient measures [4].

The NZEB target can be approached by taking into account: (i)passive design of the building envelope, for the reduction of theheating and cooling demands and loads (e.g. orientation, geome-try, high-performance thermal-isolation materials, well-designedwindows shadings, etc. [10–12]); (ii) innovative energy savingtechnological plants (e.g. HVAC, DHW, lighting systems, etc.), (iii)renewable energy applications, necessary for achieving a suitablebalance between building energy needs and renewable energiessupply [13]. Usually, the achievement of the NZEB target is strictlylinked to the implementation of passive design criteria whoseattainment also depends of different occurring local variables (cli-matic data, building use, available materials, etc. [14]). The buildingimplementation and combination of all the above mentioned meas-ures can be suitably investigated and optimized by the use ofdynamic energy simulation models and optimization procedures[15–17]. They play a crucial role in the NZEB accomplishment[18]. In particular, the main challenge of the building simulationis to provide robust insights into the NZEB principles [19]. For thispurpose, in the past years, numerous simulation tools were devel-oped, also in combination with optimization methods [20–22].Nowadays, commercial or in-house developed simulation tools arewidely applied for NZEBs analysis in order to: (i) define refer-ence buildings [6,19,23]; (ii) achieve and calculate the minimumenergy demand (which must be assessed at a global scale [24–26]);(iii) support the design procedure also for cost effective solutions[12,21,22,27].

Recently, many analyses, mostly carried out by means of NZEBssimulations, were developed in order to get the NZEB definition as afunction of the building category. Most of them are focused on res-idential buildings [23,28–33], mainly located in heating dominatedclimate zones [6,34,35]. Concerning the Mediterranean climates,the effects due to the implementation of energy efficiency measuresapplied to the residential building stock of Cyprus were examinedin a recent work [25]. Very few research works concern non-residential NZEBs. In a recent paper, energy efficiency and costoptimality analyses, based on building dynamic simulation, werecarried out on a NZEB located in the Estonian climate [34]. Theanalysis, carried out by means of dynamic simulations, focusedon a number of possible office building fenestration design solu-tions, proposed to reach the zero energy goal. A study related toan office building, located in Sweden, demonstrated the feasibil-ity of the NZEB concept when overall design ideas, constructions,installations, and energy balance of the building are taken intoaccount [36]. The main findings of an office building located inPortugal (and currently underway to reach net zero energy buildingperformance) were recently provided in [37]. Conversely, for non-residential buildings in temperate Mediterranean weather areas, alack of investigation is still observed [7].

In general, the NZEB performance analysis should be carried outby means of numerical and/or experimental data. On the otherhand, the experimental tests can be only performed after thebuilding construction, which must follow several design schemes
[4,38]. Nowadays, the design of new buildings requires the suit-able integration of different innovative elements, whose energyeffectiveness must be preventively assessed in order to avoid unex-pected and undesired performances and therefore a design failure

3 and Bu

[Sraeotct

ts(fccneNicpubaba(iNisrpusnrehaaobtep(mm

ci[Bneippewnf[f

20 A. Buonomano et al. / Energy

22]. For this purpose, the use of Building Energy Performanceimulation (BEPS) codes in the building design process, as aboveeported, is presently always recommended and it will be widelydopted particularly for NZEBs [15]. By the use of BEPS tools highnergy efficiency buildings designs can be obtained by defining theptimal set of building materials and energy saving strategies andechnologies. The energy performance and the potential savingsan be assessed even for innovative and/or not commercializedools (where experimental data are today not yet available) [38].

In this framework, recent advances in analysis and computa-ional methods, as well as computer calculation power, providedignificant opportunities for developing new in-house BEPS codesand/or improving existing ones) [17]. Such tools, mainly conceivedor research scopes, are developed despite of the availability ofommercial BEPS codes (today with high level of flexibility andomplete user interfaces/data libraries). One reason of this phe-omenon is the possibility of implementing new unreleased energyfficiency measures into such in-house BEPS codes also for suitableZEBs computer based analyses [20,27]. In general, these numer-

cal studies are well recognized to support the understanding ofomplex systems, such as NZEBs, by providing feedback on theerformance implications of all the investigated design scenarios,sage and operating conditions [38]. In particular, the effect on theuilding energy performance of focal design and operating vari-bles may be suitably detected (the eventual shortcomings cane removed) [4,38]. As a result, the use of dynamic simulationsnd sensitivity and/or parametric analysis is remarkably useful to:i) assess the energy effectiveness and the economic and financialmplications of innovative measures and technologies on the finalZEB design; (ii) provide additional insight on specific parameters

nfluencing the building performance, especially at early designteps and in case of non-standard solutions [12,22,23]. Paramet-ic analyses, based on dynamic simulation for energy optimizationurposes, while often applied to existing buildings, are still rarelytilized in case of NZEBs because of the difficulties in modellingome innovative building features (e.g. new energy saving tech-iques, novel materials, new control strategies, etc.). An exceptionegards two recent papers where, the investigated standard param-ters set concern only the building envelope of a single-familyouse [33] and the plants of an office building [11], located in alpinend subtropical climates, respectively. Note that, the recent liter-ture underlines the need of BEPS analyses to be carried out forptimal system designs. In this regard, the design process muste developed by taking into account different criteria for achievinghe building energy efficiency (not only a single technology or strat-gy) [39]. Similarly, the available literature shows that most of theroposed NZEB analyses involve common energy saving measuresdriven by local standards or economic criteria), while the imple-

entation of diverse and innovative techniques is encouraged forore sustainable development possibilities [4].Among the above mentioned novel energy saving measures spe-

ial attention has been recently paid to several innovative buildingntegrated technologies such as Phase Change Materials (PCMs)40,41], Building Integrated PhotoVoltaic (BIPV) panels and hybridIPV/Thermal (BIPV/T) plants [42–44]. For such technologies manyew simulation models were developed for assessing their influ-nce on the buildings energy demand. For the analysis of PCMsntegrated in building wallboards and in glazing systems severalurposely developed performance simulation codes were recentlyresented in [45,46]. Nevertheless, a lack of knowledge about theffects of PCMs on the whole building is still observed especiallyhen such materials are combined with other energy saving tech-
ologies [41,47]. Similarly, several simulation tools were developed
or the energy performance analysis of BIPV and BIPV/T systems48,49]. Also in this case, additional work is still required mainlyor obtaining detailed dynamic simulation models for the energy

ildings 121 (2016) 318–343

analysis of the whole building-plant system [50,51]. Note that,very few studies deal with building integrated PV systems cou-pled to underlying PCM panels. The relevance to assess the passiveand active effects on the overall building energy performance ofthe PCMs wallboards coupled to solar technologies (BIPV/T, etc.) isstrongly highlighted in [52,53]. Here, the need to develop for suchscope specific reliable simulation tools is also encouraged. Fromthis point of view few studies were carried out in [54,55]. Here,detailed building dynamic simulation models taking into accountall the related physical phenomena and including the whole energyinteraction of such integrated technologies are presented.

The above described research effort is nowadays necessary alsoin order to promote the adoption of novel high efficiency measuresand to design effective NZEBs [38]. To this purpose, although thecurrent availability of several BEPS tools, the development of newadvanced ones (and/or the enhancement of the existing codes) isstrongly recommended also in order to overcome the barrier to theNZEBs growth [16,27,56].

1.1. Aim of the study

This paper is focused on the above described framework. Inparticular, it tries to fill the above mentioned lack of knowl-edge regarding the need of comprehensive whole building energyperformance analyses in case of innovative building integratedmeasures for energy saving [27]. In this regard, the paper focuseson the description of some relevant enhancements applied to a pre-viously in-house developed dynamic BEPS code, called DETECt 2.2and written in MatLab environment [57]. The code was validatedthrough the BESTEST (Building Energy Simulation TEST) proce-dure and by means of a purposely performed benchmark analysis(also including non-residential buildings) [57,58]. Note that thisstandard code-to-code validation process is frequently adopted bythe scientific community working on BEPS codes [38]. In fact, manynew models were recently validated by such approach [20,59,60].

The tool DETECt 2.2, conceived for research purposes, was devel-oped for investigating new building envelope technologies. In thispaper, the code is utilized for developing a suitable case studyregarding the first non-residential NZEB for Mediterranean tem-perate climates. By means of the code, a parametric analysis isalso carried out in order to assess the influence of the main designand operating parameters variation on the NZEB energy behaviour.This NZEB will be built up for demonstrative purposes, in order topromote innovative energy saving techniques and strategies to beintegrated in future buildings.

Although the presented results are referred to a single casestudy, they can be useful to all the stakeholders focused on non-residential NZEBs located in temperate climates. The obtainedresults can be also taken into consideration for: (i) the NZEB defini-tion (as encouraged by the research community and governments[4,7]); (ii) comparison vs. other future NZEBs in similar operatingconditions.

The novelty of this paper regards the implementation of severalnew models within DETECt [57]. Such code is a reliable simula-tor able to dynamically predict the buildings thermo-hygrometricbehaviour and to assess the benefits of different and advancedbuilding envelope techniques, solar gain controls and daylightingsolutions in case of different weather locations, envelope materials,building shapes, orientations and geometries. The code is also ableto carry out detailed energy analyses of innovative building-plantsystems [61,62].

The main model novelties presented in this paper are referred
to:
• PCM. Here, the developed simulation tool is conceived for inves-tigating the building integration of such materials in: (i) opaque

and Bu

•

•

oaontsBds

•

•

•

to

A. Buonomano et al. / Energy

structures for walls and roofs (PCM is embedded in matrixes oftraditional building construction materials); (ii) windows glaz-ing (PCM is encapsulated in the gap of double or triple glazingsystems). Note that through such model, the optimal position ofPCM layer among the ones of the investigated building elementscan be also assessed (e.g. referred to the massive structure andthermal insulation layers). Here, the target is also to fill the abovementioned gap of knowledge about the PCM building integrationand about the combination of PCMs with other energy savingstechnologies (building sunspaces, smart daylighting control andbuilding integrated solar devices);BIPV. Here, the simulation model takes into account the mutualenergy interaction between PV systems and indoor buildingspaces. This feature is useful to calculate the heating and coolingdemand variation (passive effects) due to the building archi-tectural integration of such systems. The tool also enables thesimulation of BIPV coupled to PCM sub layers;BIPV/T. In addition to the above mentioned features, through thissimulation tool also the recovered or exhausted heat obtained byair or water utilized for cooling the PV panels is calculated. Notethat the increasing of electricity efficiency and the difference ofheating and cooling demands vs. BIPV systems are also dynam-ically assessed. In the developed model the capacitance of suchdevices is taken into account in the heat transfer calculation. Thetarget is also to follow the above mentioned literature encour-agement to develop models and tools for the PCMs-BIPV systemdesign and energy performance analysis. It is worth noting thatthe mathematical model enables the building integration of BIPVand BIPV/T modules into the building envelope, such as the roofor fac ade, and/or in parts of them (often, through the commer-cial codes, BIPV and BIPV/T cannot be simulated as only partlyintegrated in the simulated surfaces).

All such models and scripts are implemented in DETECt 2.2 forbtaining comprehensive and whole building energy performancenalyses, by properly taking into account the building integrationf such technologies. Similarly, the energy, environmental and eco-omic behaviours are calculated by assuming all the investigatedechnologies as integrated into the building. In this case, all theecondary energy (passive) effects due to the adoption of PCM,IPV and BIBV/T (heating and cooling energy saving and additionalemand) are also assessed. Several additional novelties of the pre-ented BEPS tool are:

Multi-zone modelling. Here, the heat transfer phenomena occur-ring among the different building thermal zones, and thus themutual energy interactions of such zones, are taken in to accountin the simulation model;Sunspaces modelling. Exploitation of solar greenhouse effectsdetected into the modelled building attached sunspaces is calcu-lated. The direct and diffuse solar radiation transmission throughbuilding interior windows is also assessed (in other models it isoften calculated through the perimeter openings, only);Smart daylighting modelling. Shading of windows and artificiallights can be modulated for optimizing the building heating andcooling demands, without missing the indoor visual comfort.Obviously, this feature can be also useful for assessing the incom-ing sun radiation effects on the building integrated PCMs andattached sunspaces. It is worth noting that the optimization of theenvelope performance, also by incorporating natural lighting, isconsidered as one key method for achieving very high efficiencybuildings [34].

Note that, the presented computer simulation tool allows oneo carry out parametric, as well as multi-criteria or even multi-bjective analyses (from energy, economic and environmental

ildings 121 (2016) 318–343 321

points of view), through a single simulation run. In fact, for all theinvestigated parameters (even those related to the shape and ori-entation of the building envelope), the selected ranges and stepscan be set only once at the beginning of the simulation procedure.Therefore, suitable comparisons among each investigated energyefficiency solutions can be carried out without modelling a newbuilding to be simulated for each attempt. In general, this is a veryuseful feature in the research field concerning the energy analy-sis of both new and retrofitted buildings [21,63]. In particular, itbecomes essential in case of sensitivity or parametric optimizationprocedures where the effect of the considered design ad operat-ing parameters on a selected objective function (energy efficiency,thermal comfort, economic convenience, etc.) has to be assessed[21,39,64,65]. As before mentioned, the adoption of such analysesis crucial in order to support the design and development of NZEBs.In fact, the identification of new energy efficiency solutions andthe optimization of the related features is a challenge in the NZEBresearch field [66,67].

As above mentioned, in order to show the potentiality of thepresented models, a suitable case study, referred to a new non-residential NZEB, was developed. The building, to be located inNaples (South Italy), is purposely conceived for Mediterraneantemperate climates. In particular, a multi-zone NZEB, with differ-ent innovative building integrated energy saving techniques (PCM,sunspace, smart daylighting control, etc.) is analyzed. Electricityis produced through a BIPV (or BIPV/T) system. The results of asuitable parametric analysis, developed to identify the optimal setof design and operating parameters for the best building heatingand cooling energy performance, are also discussed. A comparisonbetween BIPV and BIPV/T in terms of heating saving and extra-cooling demands and electricity efficiency is performed. Detailsabout the economic effectiveness of all the considered energy sav-ing techniques are also provided.

At the best of authors’ knowledge, this is the first analysisfocused on a non-residential NZEB conceived for Mediterraneanclimates. For such design and operating conditions new NZEB def-inition details and criteria are provided. In particular, for offices,conference rooms and expo spaces novel recommended ranges ofenergy demands for heating, cooling, lighting, appliances, ventila-tion and DHW preparation are reported. All the obtained resultscan be also useful also for comparison purposes.

2. Simulation model

2.1. General description of the multi-zone building simulationmodel

In this section, the features of the new simulation models, devel-oped in MatLab for research purposes and implemented in DETECt2.1 [57], are described. Through such initial code release the sim-ulation of temperature fields and energy fluxes for multi-zonebuildings can be carried out varying the hourly available weatherdata. Any building orientation, zoning strategy, construction mate-rials, envelope components, internal sensible and latent heat gains,operating schedules can be modelled. In particular, the buildingmodel is based on: (i) a nodal one-dimensional description ofthe thermal phenomena occurring into envelope and zones; (ii) ageometrical system description, necessary to perform a detailedcalculation of the solar radiation distribution in the simulatedthermal zones (even between adjacent ones). The code has beenvalidated through the Building Energy Simulation Test (BESTEST)
procedure and by comparing the obtained results vs. those providedby several commercial codes [57]. Details regarding the DETECt 2.1mathematical models and results of the related validation proce-dures are fully provided in [57,58].

322 A. Buonomano et al. / Energy and Buildings 121 (2016) 318–343

schem

c(t(ebaaaofwmbztitpcfp

2

hdtddbiflTddbtaMr

a

Fig. 1. Simulator

Basically, the main novelties implemented in the presented newode release (DETECt 2.2) regard the modelling and simulation of:i) PCMs for opaque and glazing systems; (ii) BIPV and BIPV/T sys-ems (also coupled to PCM elements); (iii) multi-zone buildings;iv) sunspaces. The resulting simulation code is subdivided in sev-ral sub-models, as schematized in Fig. 1. Here, the pre-calculationlock tool includes several models which are required to run onlyt the beginning of each simulation. In particular, it concerns thessessment of: (i) incident solar radiation on any arbitrarily tiltednd orientated external or internal building surface (as a functionf surfaces and sun positions and starting by direct normal and dif-use or global radiations on a horizontal surface, provided by theeather data file or calculated through the Hottel and Liu-Jordanethods [68]); (ii) shadows cast from surrounding objects and/or

uilding fixed shadings; (iii) view factors of the modelled thermalones. The core of the simulation code regards the resolution ofhe heat transfer process occurring among and within the build-ng thermal zones. This block is coupled to several scripts wherehe main simulation parameters and variable (e.g. system thermalroperties, occupancy and internal gains schedules, etc. [69]) arealculated or defined by the user, which can also choose among dif-erent mathematical models for taking into account the occurringhysical phenomena.

.2. Heat transfer through generic building elements

The building code is based on the discretization of the transienteat conduction equation, assuming a one-dimensional thermalomain, which is transformed into a suitable equivalent Resis-ance Capacitance (RC) thermal network, Fig. 2 [70]. Through suchiscretization, by adopting the finite difference method, a nodalescription of multi-zone buildings is obtained [57,58]. Thus, theuilding is subdivided in Z different thermal zones. Each one

ncludes M multi-layer elements of the building envelope (wall,oor, roof, horizontal and vertical internal partition and window).he heat transfer equation is written for each mth element (sub-ivided into N nodes). The result is a suitable set of algebraic andifferential equations (which describe the whole building thermalehaviour) to be simultaneous and iteratively solved throughouthe whole number of the integration sample times. Such equationsre numerically solved through a built-in ODE solver included in
atLab, employing variable step size Runge–Kutta and trapezoidal
ule integration methods [71].In order to calculate the building heating and cooling demands,

nd in general the heat transfer within the building and among its

e and concept.

thermal zones, the following assumptions are taken into accountfor each thermal zone: (i) indoor air is considered as uniform andmodelled as a single indoor air temperature node; (ii) thermalmasses and conductivities of the envelope elements are uniformlydiscretised as a function of the N sub-layers of different thicknesses(thus, a high order RC thermal network is obtained); (iii) for eachmth envelope component, N + 2 capacitive and surface nodes areaccounted; (iv) the indoor air mass is lumped in a single node.Details about the adopted thermal network and the numerical inte-gration methods are reported in [57].

For each zth thermal zone (z = 1, . . . , Z), and in each time step, thedifferential equation describing the energy rate of change of eachnth capacitive node (j = 1, . . . , N) of the mth element (m = 1, . . ., M)is:

Cm,ndTm,n

dt=

n+1∑j=n−1

Tm,j − Tm,n

Reqm,j

(1)

where C and T are the thermal capacitance and temperature of thenode, respectively. Req

m,jis the sum of the halves sub-layers thermal

resistances (that links the nth node to their neighbours, n − 1 andn + 1). For non-capacitive outer (n = 0) and inner (n = N + 1) surfacenodes, the algebraic equation describing the occurring heat transferis:

Tm,j − Tm,n

Rcondm,j

+ Tin,z − Tm,n

Rconvm,n

+ X · Qm,n = 0 (2)

where Rcondm,j

is the half sub-layer conductive thermal resistancethat links the outer (n = 0) and inner (n = N + 1) surface nodes totheir neighbours, (j = 1) and (j = N), respectively); Rconv

m,n is a con-vective (external or internal) thermal resistance; Rconv

m,0 connectsthe boundary non capacitive nth node to that one related to: (i)the outdoor environment air temperature, in case of perimeterenvelope components, (ii) the indoor air temperature node of theadjacent thermal zone at the previous time step; Rconv

m,N+1 connectsthe boundary non capacitive nth node to that one related to theindoor air temperature of the considered thermal zone. In case offloor elements, Rconv

m,0 links the surface node with the ground temper-ature one, becoming an equivalent thermal conductive resistance.In order to calculate the external convective thermal resistance,
the outdoor surface unitary convection heat transfer coefficientscan be either set as constant (depending on the surface condi-tion, such as vertical or horizontal wall; ascendant or descendantflow) or calculated by empirical relationships as a function of the


he RC

otd

blnf

Q

Q

wsItsm

ctestrrtar(ot

tcaRt(tcaTf

Fig. 2. Sketch of t

utdoor wind speed. The indoor surface unitary convection heatransfer coefficients depends on the surfaces slope and the heat fluxirection.

The modelled forcing function Qm,n varies as a function of theuilding element position and includes the incident solar and the

ong-wave radiation exchange acting on outer and inner surfacesodes. In particular, for each mth building element, such radiation

unctions are calculated as:

˙ m,0 = [εextm · � · f · (T4

sky − T4m,1) + ˛ext

m · Iextm ] · Am (3)

˙ m,N+1 = [εintm · � · G + ˛int

m · Iintm ] · Am (4)

here, f is the external surface view factor; A is the heat exchangeurface area; ε and are the emissivity and absorption factors;

is the total incident solar radiation flux and Tsky is the skyemperature. The long-wave radiation exchange on the internalurfaces within the zone is taken into account through the Gebhart’satrix, G =

∑Mi=1Gm,i(T4

i,N− T4

m,N), where Gm,i is a matrix coeffi-

ient. Qm,0 and Qm,N+1 act on the external and internal surfaces ofhe considered thermal zone, respectively. Such adopted method,xperimentally validated and described in [72], enables the thermalimulation of attached sunspaces or highly glazed surfaces. In par-icular, Qm,N+1 takes into account the total solar radiation flux (Iint

m )eceived by an internal mth surface including: (i) the incident solaradiation entering through windows, absorbed, reflected and dis-ributed within the internal space by selected absorption, reflectionnd view factors; (ii) the net long-wave radiation received by theemaining ith (i = 1, . . ., M and i /= m) internal surfaces. Note that XEq. (2)) is a control function, which is set to zero for non-capacitiveuter (n = 0) surface nodes of building elements separating adjacenthermal zones, and equal to 1 elsewhere.

In the general case of multi-layered gas filled glazing systems,he calculation of the occurring heat transfer within the cavity isarried out by means of two different procedures, which may belternatively selected by the user. They regard the assessment ofconvm,n that may be: (i) considered as a constant global thermal resis-ance, taking into account both the convective and radiative effectsthus Qm,n is set to zero); (ii) suitably calculated as a function ofhe occurring gas temperature and properties, thickness of the
avity between glass surfaces and window height (thus the radi-tion effects are accounted in the source term, Qm,n, of Eq. (2)).he correlation adopted in such second procedure is not reportedor sake of brevity, but thoroughly presented and summarized in
thermal network.

[73]. Note that, for the calculation of Qm,n, the optical properties ofthe glass panes (varying as a function of the incidence solar angleand the radiation absorbed by the glazed surfaces) are considereduniformly distributed within the glazing material. More details arereported in [57,58].

In each time step and for each zth thermal zone, the differentialequation describing the sensible energy rate of change of the indoorair mass, to be solved simultaneously with the system of Eqs. (1)and (2), is calculated as:

Cin,zdTin,z

dt=

M∑m=1

Tm,N − Tin,z

Rconvm,N+1

+Z∑

k=1

Tin,k − Tin,z

Rv,zns

+ (Tout − Tin,z)Rv

+ Qg,z ± QAC,z (5)

where Cin,z is the thermal capacitance of the zone indoor air, whosetemperature, Tin,z, is considered homogeneous in the space (perfectindoor air mixing). The first term on the right-hand side of Eq. (5)describes the heat exchange between the internal surfaces nodesof the building elements enclosing the zth zone and the indoor air.The thermal resistances Rv and Rv,zns take into account both theair ventilation and infiltration thermal loads. In particular, to theindoor air node (at Tin,z temperature) of the considered zth zone,Rv links the external node (outdoor air at Tout temperature), whileRv,zns links the indoor air node related to any adjacent thermal zone(at Tin,k temperature, calculated at the previous time step and withk = 1, . . ., Z and k /= z). The infiltration thermal load is modelledby means of the model presented in [74,75]. Qg,z represents thelumped heat source term, consisting of convective sensible internalgains due to occupants (where the metabolic rate is calculated as afunction of the indoor air temperature), lights and equipment. Notethat, the internal gain due to occupancy depends on the relatedselected scheduling. A direct and indirect influence on the build-ing energy demands is caused by such thermal solicitation (e.g.through the control of: windows solar shadings for daylighting,ventilation rates, indoor air temperature set points, etc.). QAC,z isthe sensible heat to be supplied to (or removed from) the ther-mal zone by an ideal HVAC system, aiming at maintaining theindoor air at the desired set point temperature [57]. Note that, in
Eqs. (1)–(4), the subscript z is not included for sake of simplicity.QAC,z of Eq. (5) is calculated according to a Proportional Integral (PI)control strategy, widely used in buildings for heating and coolingsystems control for its easy design and implementation [76]. In the

3 and Bu

pfdtttob[

edbilshc

2

e

2

niadPafBmmtinetieiuastuttbrttir

C

g

I


resented simulation code, an accurate, stable and robust methodor the tuning of the PI control gains, for the different operating con-itions, was implemented. Through the adopted tuning strategy,he proportional and the integral gains of the controller are thosehat minimize a quadratic function weighting the temperatureracking error and the sensible heat load [77]. Authors are workingn addressing the feasibility of more innovative control strategiesased on the online adaptation of the control gains (according to78]), as reported in [79,80].

Finally, the system including all building thermal zones is mod-lled through a thermal network of Z × ((N + 2) × M + 1) nodes; theynamics of the whole set of node temperatures are describedy the differential and algebraic Eqs. (1), (2) and (5), by taking

nto account the internal and external solicitations (consisting ofinear and nonlinear input terms). As a consequence of such state-pace formulation, during each simulation time step, convectioneat transfer coefficients and building envelope features (such asapacities, conductivities, etc.) are assumed as invariant.

.3. Enhancements applied to the building simulation code

In the following, the mathematical models related to the mainnhancements included in DETECt 2.2 are discussed.

.3.1. Building integrated PhotoVoltaic/Thermal systemsSince the DETECt simulation code is based on RC thermal

etworks the implementation in the original model of build-ng integrated solar systems (such as solar thermal collector, PVnd PV/T panels) can be suitably achieved [81]. The first modelescription regards Building Integrated PhotoVoltaic (BIPV) andhotoVoltaic/Thermal (BIPV/T) systems [82]. Here, the followingssumptions are taken into account: (i) one dimensional heat trans-er, (ii) isothermal surfaces of the PV module and channel (forIPV/T), (iii) neglected system edge heat losses [83]. In order toodel a BIPV (or BIPV/T) system, the interaction between suchodules and the underlying building envelope must be accurately

aken into account. In particular, the back system temperaturenfluences the building heating and cooling loads and, simulta-eously, the module temperature strongly affects its electricityfficiency [84,85]. The system conductive, convective and radiativehermal exchanges are suitably accounted through the mathemat-cal model described in Section 2.2. The interaction between thelectrical and thermal efficiencies is properly modelled by takingnto account the thermal and optical properties of the PV mod-les and the related amount of incident radiation. The PV modulesre modelled by several capacitive nodes, each one is related to aingle module layer [83,86]. Note that for sake of simplicity, thehermal inertia of the covering glass is neglected. The PV mod-les are modelled as integrated in the building envelope. Therefore,heir thermal behaviour can be assessed by accurately modifyinghe previous Eqs. (1) and (2) (as reported in Section 2.2, the energyalance model of each capacitive node is obtained by dividing theelated layer in two half thickness sub-layers). The operating cellemperature (TPV) is accurately calculated by taking into accounthe solar radiation absorbed by the PV cell. The resulting energys added on the right side of the above reported Eq. (1). This one,eferred to the PV cell capacitive node, becomes:

PVdTPV

dt=

n+1∑j=n−1

Tm,j − TPV

Rcondm,j·n

+ IPV · APV (6)

IPV is the effective absorbed solar radiation per unit of PV cellross area (APV), calculated as:

PV = (K� · �gl · ˛PV − �PV ) · Iextm (7)

ildings 121 (2016) 318–343

where �gl is the glazing cover transmittance at normal incidenceangle, ˛PV is the PV cell absorptance, �PV is the PV module efficiency,Iextm is the global incident solar radiation (transmitted through

the glass cover if applied to the PV layer) and K� is the IncidentAngle Modifier (IAM). K� is defined as the ratio of the radiationabsorbed by the cell at some incidence angle divided by the radi-ation absorbed by the cell at normal incidence (K� = �(ϑi)/�(0)),as reported in [87]. For the case of single glass cover, K� can becalculated by following different procedures implemented in thecode [87,88]. Nevertheless, the default procedure is based on thecalculation of the refraction angle (ϑr) and of the cover glass trans-mittance at any incidence angle and at the normal one (�(ϑi) and�(0), respectively) [87]. Such parameters are calculated as:

ϑr = arcsin

(nout

ngl· sin(ϑi)

)

�(ϑi) = e−(Kıgl/ cos(ϑr )) ·[

1 − 12

(sin2(ϑr − ϑi)

sin2(ϑr + ϑi)+ tan2(ϑr − ϑi)

tan2(ϑr + ϑi)

)]

�(0) = limϑi→0

�(ϑi) = e−(Kıgl) ·[

1 −(

1 − ngl

1 + ngl

)2]

(8)

where, ngl and nout are the cover glass and air refraction indexes,respectively; K is the glazing extinction coefficient and ıgl is theglazing thickness. Typical input parameters for PV modules areadopted by following the study reported in [87]. Note that, K� takesinto account sky diffuse, ground reflected and beam componentsof the solar radiation.

The photovoltaic cell efficiency, �c, is assessed by taking intoaccount the reference cell efficiency in standard conditions (�ref,provided by the manufacturers), and the cell temperature [84]:

�c = �ref [1 − ˇref · (TPV − Tref ) + log10 · Iextm ] (9)

where �ref is the cell electrical efficiency at the reference temper-ature (Tref at 25 ◦C) and solar radiation flux (1000 W/m2), ˇref isthe temperature coefficient and is the solar radiation coefficient(equal to 0.12 for crystalline silicon modules). ˇref depends on thetemperature at which the PV efficiency drops to zero (it is equal to0.0045 ◦C−1 for crystalline silicon, 0.0035 ◦C−1 for CIS, 0.0025 ◦C−1

for CdTe and 0.0020 ◦C−1 for a-Si) [89]. A linear expression for thePV electrical efficiency can be also adopted, by imposing = 0 [84].

The photovoltaic module efficiency is lower than the cell effi-ciency (�c) because of the electricity amount lost in the moduleconnections. The module efficiency is calculated as: �PV = �c · �mod,where �mod is the module conversion efficiency. The net power(PPV) produced by the system is obtained by the gross electricalpower produced by the PV module reduced by the electricity lossdue to the inverter efficiency �inv. PPV is calculated as:

PPV = (�PV · Iextm ) · �inv · APV (10)

Note that, the calculation of the produced electricity is relatedto the effective PV surface. The modelled BIPV system configurationis a fully integrated PV module, directly integrated in the buildingroof. As above reported, the higher the working temperature ofthe solar cells (it occurs particularly after a continuous operation),the higher the cells efficiency loss. Therefore, the electricity yieldcan be improved by cooling the PV module with a fluid stream,as it occurs in Building Integration PhotoVoltaic/Thermal system(BIPV/T) systems [90].

The second model implemented in the building simulation code
relates to a BIPV/T system. Here, through either air or liquids, thethermal energy generated by the PV system can be suitably recov-ered (or exhausted). In this case, electricity and thermal energy aresimultaneously produced. The modelled BIPV/T configuration is the


d BIPV

oTtmmrseesdgBeaddmo(

m

watte

momnsc

Fig. 3. Cross section of the modelle

ne reported in Fig. 3. Here, a BIPV ventilated system is sketched.he considered BIPV/T system is designed and modelled in ordero cover a continuous roof surface, where several rows of BIPV/T

odules are connected in parallel. The implemented BIPV/T ther-al model is based on the approach proposed in [90]. Here, each

ow of the roof is subdivided in several finite elements, each otherubsequent in the cooling stream length direction. For each finitelement, the thermal energy and electricity balances on the PV lay-rs and cooling fluid channel as well as on the top and bottomurfaces of the roof back material are assessed through the aboveiscussed Eqs. (1), (2), (6) and (9). The heat recovered in each sin-le finite element is obtained by solving the energy balance on theIPV/T cooling fluid duct. Here, the outlet fluid temperature of onelement (Tair out) is assumed as input (Tair in) in the subsequent one,s shown in Fig. 3. In each finite element, the differential equationescribing the sensible energy rate of change of the flowing fluidepends on: (i) the rate of heat received from the back of the PVodule (tedlar layer at temperature Ttd) at the air side, (ii) the rate

f heat exchanged through the cooling fluid and the roof materialat temperature Tins). It is calculated as it follows:

˙ air · cfluiddTair

dx= w · hup(Ttd − Tfluid) + w · hlow(Tins − Tfluid) (11)

here dx is the finite element length in the cooling stream directionnd w is the related width; hup and hlow are the convective heatransfer coefficients for the upper and lower surfaces, calculatedhrough the correlations reported in [90]. The solution of Eq. (11)nables the adoption of different heating recovery strategies.

It must be noted that in case of a BIPV/T system, although theodelled roof rows are subdivided in many finite elements, only

ne internal node facing the indoor air of the considered ther-
al zone (see Eq. (2) with n = N + 1) is taken into account. Such
ode is linked to the neighbour kth nodes in which the roof isubdivided through the conductive thermal resistances (Rcond

m,j,k), cal-

ulated according to the modelled finite element heat exchange

/T system (summer configuration).

surface area, Fig. 3. This implies that Eq. (2) must be rewritten as itfollows:∑

k

Tm,j,k − Tm,n

Rcondm,j,k

+ Tin,z − Tm,n

Rconvm,n

+ X · Qm,n = 0 (12)

In case of a BIPV system, the roof is subdivided in a maximumof two sections, whose surface areas depend on the ratio betweenthe surface area covered by PV cells and the total roof one. Conse-quently, only two resistances (Rcond

m,j,k) link the related nodes to the

surface internal one facing the indoor air. This assumption allowsthe authors to suitably couple the roof BIPV/T and BIPV modelsto the DETECt building dynamic simulation code. Similar consider-ations are taken into account for dynamically modelling the energyperformance of BIPV or BIPV/T integrated in building vertical walls.

2.3.2. Building integrated phase change materialsThe developed model for the dynamic simulation of the building

integrated Phase Change Materials (PCMs) is based on an effec-tive heat capacity approach [41,91]. Such method, implemented inDETECt 2.2 for the whole building analysis, enables the numericalmodelling of PCMs by taking into account an equivalent heat capac-ity (assessing the latent heat as an increased amount of sensibleheat in the transition phase). This approach simplifies the phase-change heat transfer phenomenon which is dealt with as a singlephase non-linear conduction problem [41]. Here, a larger numberof degrees of freedom with respect to the real melting process isobtained. On the other hand, it presents some limitations due tothe difficulties in replicating the real system behaviour. In the sim-ulation model a high number of nodes in the modelled RC thermalnetworks has to be taken into account because of the assumed con-stant properties of materials during each simulation time step [92].In the developed model, the behaviour of PCMs undergoing phase
change is taken into account in Eq. (1) by adopting a temperaturedependent thermal capacitance (Cm,n). In each modelled materiallayer, the specific heat of PCMs is considered as constant only incase of a single phase; in all the other cases, it is assumed variable


ar rad

ws(mcp

c

w(cpIalwihiptm

eegeig(t(riuTadlp

Fig. 4. Indoor sol

ith the occurring temperature (cPCM(T)). According to Eq. (13),uch temperature affects the state of aggregation of the PCM layerliquid state, undergoing the phase change, solid state). The above

entioned equivalent heat capacity usually fits a Gaussian shapedurve, where the maximum corresponds to the peak melting tem-erature [93]. In particular cPCM is calculated as:

PCM(T) = a · e−((T−b)/c)2(13)

here a is the maximum increment of cPCM due to the latent heatheight of the curve peak), b is the average temperature of the phasehange for melting and solidification (which determines the curveeak position) and c is the range of the phase change (curve width).

n the developed building simulation code, a database of materi-ls properties, to be implemented in Eq. (13) (obtained throughaboratory tests or provided by manufacturers), is included. It is

orth noting that the profiles of the equivalent heat capacity dur-ng the melting and solidification processes are slightly different (aysteresis phenomenon is typical for paraffin materials). Accord-

ng to this phenomenon, different parameters in Eq. (13) for suchrocesses must be set. Note that also PCMs embedded in tradi-ional building materials matrixes (gypsum, concrete, etc.) can be

odelled and simulated by the presented model.The above described simulation method is adopted also for mod-

lling the thermal behaviour of PCMs encapsulated in transparentlements [46,94]. Note that, in order to suitably model complexlazing systems, (such as double or triple glazing gaps filled withither PCM or air/gas), the thermal inertia of glasses is always takennto account. Conversely, for the nodes at the interface between thelass panes and their gap the heat capacity is neglected (Eqs. (1) and2)) [95,96]. In the presented model the PCMs optical behaviour isaken into account and a PCM layer is modelled as: (i) translucentit behaves as a diffusive medium, thus the direct incident solaradiation is converted into transmitted/reflected diffuse radiation)f the phase is solid, non-completely solid and non-completely liq-id; (ii) transparent (as a conventional glass) if the phase is liquid.he assessment of the PCM solar and light transmittance in liquid
nd solid phase follows the assumptions reported in [46]. In theeveloped building simulation code, a database of the solar and
ight transmittances of different glazing systems is also included. Inarticular, for double glasses with liquid phase PCMs, the nominal

iation modelling.

solar and light transmittances are set as 0.75 and 0.85, respectively.The same parameters are assumed equal to 0.46 and 0.55, respec-tively, for solid phase and mushy state. Note that the reduction ofthe solar and light transmittances is due to the increase of both theabsorptance and reflectance [46,97,98].

Several simplifications are assumed in the presented calcula-tion procedures, such as: (i) the convection within the phase changematerial layer is always neglected (since in the opaque building ele-ments PCMs are often encapsulated in mini or micro cells and, morein general, since in most of the time the PCM is not completely liq-uid); (ii) the radiative exchange between the glasses surfaces facingthe cavity/cavities is always neglected (because of the PCM non-transparency to the long-wave radiation [46]). As a result, in caseof PCM encapsulated in glasses, in Eq. (2) the thermal resistance,Rconv

m,n , becomes a conductive resistance and Qm,n is set to zero.

2.3.3. Building integrated sunspaceIn the presented simulation code, a detailed model for the cal-

culation of the solar radiation distribution and reflection insidethermal zones is implemented [57]. Such model, is suitable to sim-ulate the thermal behaviour of building sunspaces and buildingindoor spaces with highly glazed surfaces [72]. The effects of theincident solar and the long-wave radiation in the building spacesare modelled through the forcing function term Qm,n of Eqs. (3) and(4).

In order to determine the energy saving achieved through build-ing sunspaces, a suitable calculation of the related thermal andoptical behaviour is carried out. In particular, the assessment ofthe sunspace solar heat gains is performed by taking into accountthe energy transferred through the sunspace external glazing sur-faces, absorbed by the sunspace internal surfaces and transmittedby conduction and radiation to the adjacent building thermal zones(through separation walls and glazing surfaces). Note that, by thedeveloped model the indoor spaces are heated also by the solarradiation directly transmitted through both the external sunspacewindows and the sunspace to indoor space ones, Fig. 4. Here, the
calculation of the incident direct solar radiation transmitted tothe adjacent thermal zones is obtained by taking into account thesunspace ceiling and lateral walls as overhang (Fig. 4) and fins,respectively. Note that, all the optic features of the sunspace glazing


pe in N

sa

2

tstossitaaig[tc

L

wCsamsfoab

ssd

Fig. 5. NZEB prototy

ystem are calculated as a function of the solar radiation incidencengle [82].

.3.4. Building integrated lightingIn order to design highly energy efficient buildings by reducing

he need of artificial lighting and the heat gains due to the naturalunlight, an integrated approach incorporating both lighting andhermal simulations is adopted. In this regard, because of the lackf local measured daylight data and in order to assess for all theky conditions the luminous efficacy of global, diffuse and directolar radiation on vertical building surfaces, a simplified method ismplemented [99]. Here, although the luminous efficacy (given byhe ratio between the illuminance (Lext) and the irradiance (I)) isssumed as constant, reliable predictions of the global illuminancere still obtained [99–101]. In addition, the average illuminance onndoor horizontal surfaces, Lis, is assumed as isotropic and homo-eneous and is calculated by adopting a simplified lumen method102]. By this approach the same calculation procedure for bothhe side and top lighting is taken into account. In particular, Lis isalculated as:

is = Lext · ςv · Cu · Agl

Ais(14)

here Lext is the illuminance on transparent external surfaces,u is the room utilization coefficient [102], ςv is the transmis-ion coefficient of the visible radiation, Agl and Ais are the glazingnd horizontal surface areas, respectively. Note that in the imple-ented model the angular distributions of the indoor illuminance,

urface reflectance and glare are disregarded. Suitable relationshipsor assessing the sky conditions (clear, partially overcast and totallyvercast) are selected as a function of the Perez’s index calculateds a function of the solar zenith angle, the diffuse horizontal andeam normal irradiances [101].

As a result, the optimal tilt of modulating windows external solarhadings is automatically calculated for minimizing the summerolar heat gains, by simultaneously obtaining the desired indooraylighting [103,104]. For insufficient daylighting, the artificial

aples (South Italy).

lights are switched on during any simulation time interval. In par-ticular, the control of the solar shadings devices is obtained bymodelling the glass transmittance as a function of: (i) the beamsolar radiation incident on the windows; (ii) the desired totalilluminance (average standard indoor horizontal lighting require-ment) on the considered building zone (typically set from 200to 1000 lux) [105]. Continuous and step dimming lighting con-trols can be carried out (aiming at assuring the set point, Lsp, foreach indoor space use). In case of continuous dimming control, theartificial light output (Llm) and thermal/power input (Plux) contin-uously and linearly increase as the daylight illuminance decreases.Through the step dimming strategy, Plux and Llm vary in discrete andequally spaced steps. Note that the lighting power ranges from 0(when daylighting is sufficient to reach Lsp) to the maximum power(always provided at zero daylight illuminance), depending on theilluminance efficiency (�lm) of the adopted lighting technology.

In each simulation time step, the total amount of Llm to beprovided by the lighting system is calculated as the differencebetween the lighting set point and the calculated indoor illumi-nance (Llm = Lsp − Lis), while the related Plux is calculated as:

Plux =

⎧⎨⎩

Llm

�lmAis if Lsp > 0

0 if Lsp ≤ 0(15)

It is worth noting that, due to the very low visible efficiency oflighting devices, almost all the consumed electricity is convertedto heat.

3. Case study

Aim of this study is the energy design and performance anal-ysis of a non-residential NZEB to be located in Naples (Southern
Italy, Mediterranean temperate climate 40◦20′ N–14◦15′ E). The ini-tiative concerning the construction of this NZEB stems from theaction ED6 of the Sustainable Energy Action Plan (SEAP, Covenant ofMajors of the European Community, August 3rd 2012) and from an

328 A. Buonomano et al. / Energy and Bu

eo

gfroM

3

hwEzmT

are modulated as a function of the occurring minimum daylight-

Fig. 6. Simulated thermal zones.

xplicit Resolution (n. 517 on April 21st 2011) of the Municipalityf Naples [106].

The building will be built up on three floors (two above theround level). It will host offices (at the ground floor) and con-erence and exposition spaces (at the first floor). Some equipmentooms and stories will be located at the basement. To the knowledgef the authors, it is the first Italian non-residential NZEB project inediterranean areas [106].

.1. Building geometry and envelope features

The building concept is a result of bioclimatic criteria for passiveeating and cooling techniques. The building (Fig. 5) is conceivedith a rectangular shape (15.0 × 24.5 m), its longitudinal axis is

ast-West oriented and it is subdivided in ten different thermal
ones (Fig. 6). The surface area of the conditioned spaces at the base-ent, ground floor and first floor are 91, 238, 253 m2, respectively.
he S/V (surface to volume) ratio is 0.38.

ildings 121 (2016) 318–343

A high thermal capacitive and insulating envelope is taken intoaccount. In the preliminary design, the adopted U values for all theopaque building elements and windows were selected equal to themaximum ones allowed by the present Italian regulation concern-ing the energy efficiency in public buildings [1]. Initial designedsuperficial masses of the opaque building elements attend suchrules too. The solar absorptance of all external surfaces is set equalto 0.3 (reflective cool paint) for opaque surfaces (0.9 for the solarfield surfaces), while their thermal emissivity is set equal to 0.9.Concerning interior surfaces, the solar absorptance of walls, ceil-ings and floors are set to 0.2, 0.1 and from 0.3 to 0.9 (as a functionof the considered indoor space), respectively. The emissivity of suchsurfaces and the related long wave absorptances are assumed equalto 0.9 and 0.1, respectively. The building shape and the interiordesign are conceived in order to maximize the natural ventilationby stack effect. Details about the building elements stratigraphy arereported in Table 1.

For minimizing the winter heat loss and the summer solar cool-ing loads, no windows are designed on the eastern, northern andwestern fac ades. The building window to wall surface area ratio isquite low (about 15%), while it becomes very high (about 70%) whenreferred to the southern fac ade only (because of the wide windowsconceived also for maximizing the winter solar heat gain).

The first floor terrace windows are equipped by external hori-zontal variable tilt solar shadings, while horizontal overhangs aremodelled on the top of the roof windows. At the southern side of theground floor, a sunspace is designed in order to maximize the win-ter passive heating, Fig. 5. During the summer season such spacebecomes (by completely opening the external sliding windows) ashaded open porch. Note that due to such building metamorpho-sis, the simulated thermal zones become nine and the S/V ratiodecreases to 0.36. The porch ceiling width and height are designedin order to avoid the indoor space superheating enhancing the sys-tem energy efficiency. In fact, the direct solar radiation strikes onthe sunspace south vertical opaque wall and interior windows inwinter and on the outdoor floor of the open porch in summer. Allthe analyzed windows typologies are reported in Table 2.

At last, in order to maximize the yearly energy performanceof the roof building integrated solar field, made of PV panels andthermal solar collectors, a roof slope of 30◦ is taken into account,Fig. 5. The BIPV panels occupy the 70% of the roof area and aremade of mono-crystalline silicon cells (156 × 156 mm) with a nom-inal efficiency and a peak power of 0.147 and 205 W (standard testconditions), respectively. Each PV module is 1694 mm length and998 mm width. Note that the refraction index of the cover glass, theglazing extinction coefficient and thickness and the typical inputparameters for PV modules are taken into account [87]. The solarthermal collectors field (30% of the roof surface area) consists ofinnovative building integrated flat-plate evacuated solar thermalpanels [107]. Note that such large installed solar field capacitiesare conceived and designed to also supply energy to an existingpublic building adjacent to the investigated one.

3.2. Building operating conditions

The occupancy in each zone is scheduled as a function of theweek day and day hours. The sensible heat gain and the vapouremission due to people are modelled by taking into account therelated activity rate (e.g. [108]) and the indoor air temperature. Theheat gain due to machineries depends on the building spaces useand on an hourly profile. As above described the artificial lightsand the solar radiation transmitted through windows shadings

ing (visual comfort) required in the building. For all the indoorspaces, suitable outdoor air flow rates (ventilation) are modelledas a function of the building use, time and occupancy (note that


Table 1Features of the opaque building envelope (CASE 0).

Building opaque element WeightMs [kg/m2]

TransmittanceU [W/m2K]

Layers fromoutside to inside

Thicknessd [mm]

Partition walls vs. notheated/cooled spaces

79

0.56

Interior plasterboard 20Thermal insulation 30Hollow bricks 100Interior plasterboard 20

Wall vs. ground 820 0.33

Cast concrete 300Thermal insulation 50Aerated bricks 30Interior plasterboard 20

Vertical walls 242 0.35

Exterior plasterboard 30Thermal insulation 50Semi-hollow bricks 300Interior plasterboard 20

Interior wall 78 0.74

Interior plasterboard 20Hollow bricks 100Thermal insulation 20Interior plasterboard 20

Sunspace to indoorwall (South) 482 1.82

Stonework 30Brickwork 250Interior plasterboard 20

Ground floor slab 489 0.75

Concrete ground 160Thermal insulation 20Concrete slab 150Cement-based mortar 10Ceramic tiles 20

1st floor slab vs. notheated/cooled spaces) 187 0.74

Interior plasterboard 20Thermal insulation 20Concrete slab 150Cement-based mortar 10Ceramic tiles 20

2nd and 1st floor slab (vs.heated/cooled spaces) 187 0.74

Interior plasterboard 20Concrete slab 150Thermal insulation 20Cement-based mortar 10Ceramic tiles 20

Terrace slab to sunspace 266 0.33

Ceramic tiles 20Cement-based mortar 50Thermal insulation 60Concrete slab 160Interior plasterboard 20

Horizontal roof 233 0.34

Tiles (roofing) 20Thermal insulation 70Concrete slab 50Interior plasterboard 20

Tilted roof 233 0.33

PV panels/solar collectors 45Thermal insulation 80Concrete slab 230Interior plasterboard 20

Table 2Investigated glazing types.

Glazing type Gap TransmittanceU [W/m2 K]

Solar heat gain coefficientg [%]

Solar transmittance�sol [%]

Visible Transmittance�vis [%]

Emissivityε

1 6/13/6 Air 2.7 0.70 0.61 0.78 0.842 6/13/6 Argon 2.5 0.70 0.60 0.78 0.843 6/13/6 (low-ε) Argon 1.6 0.58 0.51 0.75 0.104 6/8/6 (low-ε) Krypton 1.3 0.54 0.47 0.74 0.105 6/13/6/13/6 Air 1.7 0.61 0.48 0.71 0.846 6/8/6/8/6 (low-ε) Krypton 0.9 0.46 0.29 0.63 0.10

N able th

tcei

ote that all the investigated windows are equipped by metallic frames with a suit

he ventilation flow rates are modulated emulating a suitableontrol of indoor CO2). A 50% efficiency air-to-air sensible heatxchanger is taken into account for the energy recovery of the build-ng exhausted air. The air infiltration rate is always equal to 0.4 vol/h

ermal cut.

(even during the night hours). Details about design features andnominal simulation assumptions are reported in Table 3. Here, itcan be observed that daily variable profiles are adopted in order tosimulate the occupants’ behaviour.


Table 3Simulation assumptions.

Buildingspace

Hours Occupancydays per week

People Artificial lighting[W/m2]

Machineries[W/m2]

Outdoor air flowrate [l/s p]

[p/m2] [Wsen/p] [gv/h] Winter Summer

at 20 and 26 ◦C

Office9:00–12.00

5 (Mon. to Fri.)0.12 77, 45 60, 105 Modulated from 0 to

10 (min. 500 lux onoffice desk level)

9 1113:00–14:00 0.03 77, 45 60, 10515:00–18:00 0.10 77, 45 60, 105

Expo

9:00–11:003(Mon., Wed., Fri.)

0.28 82, 47 67, 117Modulated from 0 to 8(min. 400 lux at 1.5 mheight from floor)

– 6.012:00–13:00 0.35 82, 47 67, 11714:00–15:00 0.15 82, 47 67, 11716:00–18:00 0.25 82, 47 67, 117

Conference9:00–12:00

1(Fri.)

0.6 60, 37 50, 63 Modulated from 0 to 6(min. 200 lux on officedesk level)

5 5.513:00–14:00 0.1 60, 37 50, 6315:00–18:00 0.4 60, 37 50, 63

N ated aa

doaaettJv3sfmr

3

bpaswtt

st

TI

N

ote that: the sensible thermal loads and vapour emission due to people are moduls a function of the hourly occupancy.

Simulations are carried out by using the IWEC hourly weatherata (air dry bulb temperature and humidity, solar radiation, etc.)f Naples. Here, the heating and cooling degree days are 1163nd 185 kd, respectively. During winter, outdoor air temperaturesre not excessively low (design and average outdoor temperaturequal to 2 and 10 ◦C, respectively), while quite high outdoor airemperatures and humidity occur in several summer days (designemperature and humidity: 32 ◦C, 60%). The simulation starts atanuary 1st and ends at December 31st. The HVAC system is acti-ated from 09:00 to 18:00, in winter from November 15th to March1st and in summer form June 1st to September 30th (in August it iswitched off) [109]. The modelled indoor air set-point temperatureor the heating season is 20 ◦C (sensible heating only). The sum-

er set-points for temperature and humidity are 26 ◦C and 60%,espectively (sensible cooling and dehumidification).

.3. Parametric analysis

In order to further optimize the design of the investigateduilding, several innovative energy efficiency measures (e.g. high-erformance envelopes, sun control and shading devices, windowsnd glazing, passive solar heating and natural ventilation) wereelected. Their combined effects on the building energy demandsere assessed by means of a suitable parametric analysis, obtaining

he optimal set of the main energy design and operating parameters
hat minimize the heating and cooling demands.
In the carried out optimization procedure of the developed casetudy, the investigated design parameters concerning the perime-er vertical walls (Table 1) are:

able 4nvestigated combinations of the external walls and roof layers.

Building element Wall and roof configurations

1

2

3

4

5

6

7

8

ote that o corresponds to walls exterior plasterboards and roof covers, while i correspon

s a function of the indoor air temperature; the ventilation flow rates are modulated

1. external walls insulation thickness – InsThwall;2. insulation position in the external walls stratification –

InsPswall. The investigated combinations of the external wallsand roof layers are reported in Table 4. Here, the insulationlayer can be positioned either externally (configuration 1) orinternally (configuration 5) to the massive building element(semi-hollow brick) with respect to the conditioned space;

3. external walls semi-hollow bricks density – Wghwall;4. external walls PCM thickness – PcmThwall;5. PCM position in the external walls stratification – PcmPswall. In

this case six different combinations of the PCM layer position,the insulation layer and the semi-hollow brick are taken intoaccount (configurations 2, 3, 4 and 6, 7, 8 of Table 4).

The investigated design parameters concerning the roof are:6. roof insulation thickness – InsThroof;7. insulation position in the tilted roof stratification – InsPsroof. In

this case, the insulation layer can be positioned either above(configuration 1, Table 4) or below (configuration 5, Table 4)the massive building element (concrete slab);

8. roof concrete slab density – Wghroof;9. roof PCM thickness – PcmThroof;

10. PCM position in the roof stratification – PcmPsroof. In this case,under the PV panel six different combinations related to theposition of PCM layer, the insulation layer and the concreteslab are taken into account (configurations 2, 3, 4 and 6, 7, 8of Table 4).

The investigated design parameters concerning the buildingwindows are:

11. window to wall surface area ratio – W/W%. This parameter refersto the windows glazing surface of the southern building fac ade

Layers

A = thermal insulation; B = hollow brick/concrete slabA = PCM; B = thermal insulation; C = hollow brick/concrete slabA = thermal insulation; B = PCM; C = hollow brick/concrete slabA = thermal insulation; B = hollow brick/concrete slab; C = PCMA = hollow brick/concrete slab; B = thermal insulationA = PCM; B = Hollow brick/concrete slab; C = thermal insulationA = hollow brick/concrete slab; B = PCM; C = thermal insulationA = hollow brick/concrete slab; B = thermal insulation; C = PCM

ds to walls and roof interior plasterboards.


Table 5Results of the parametric analysis.

Variable Range Step Starting value (CASE 0) Optimal value (CASE OPT)

InsThwall [mm] 40–90 10 50 90 (wall: UOPT = 0.23 W/m2 K)InsPswall Table 4 1 Configuration 1 Configurations 4 (offices) and 8 (elsewhere)Wghwall [kg/m3] 800–1300 250 800 1050PcmThwall [mm] 10–30 10 0 30PcmPswall Table 4 1 – Configurations 4 (offices) and 8 (elsewhere)InsThroof [mm] 60–110 10 80 110 (roof: UOPT = 0.23 W/m2 K)InsPsroof Table 4 1 Configuration 1 Configuration 2Wghroof [kg/m3] 800–1300 250 1050 1300PcmThroof [mm] 10–30 10 0 30PcmPsroof Table 4 1 – Configuration 2W/W% [%] 30–70 20 70 70Typgl1 Table 2 1 Glazing type 1 Glazing type 3Typgl2 Table 2 1 Glazing type 3 Glazing type 3 (East office conference room) glazing type 6 (elsewhere)Wdsg [m] 1–5 1 4 3SWAbssg 0.3–0.6 0.15 0.6 0.6

11

1

111

1

atittto0

cm(cct

E

wsoia

FAbssg 0.3–0.6 0.15 0.6

SWThsg [mm] 0.15–0.30 0.05 0.25

NFCac [vol/h] 1–3 0.5 0.5

(except to the exterior one of the sunspace that is not variedin this analysis). It is also worth noting that the increase of thewindow surface on the walls is obtained by widen downwardsthe windows;

2. glazing type of the sunspace exterior windows – Typgl1, Table 2;3. glazing type of the remaining building windows – Typgl2,

Table 2;The investigated design parameters concerning the sunspace,

are:4. zone width – Wdsg (distance from front windows to back wall).

Note that this parameter in summer corresponds to the porchoverhang width;

5. solar absorption coefficient of the interior wall – SWAbssg;6. solar absorption coefficient of the floor – FAbssg;7. thickness of the wall separating the sunspace and the interior

zones at the ground floor – SWThsg.Another investigated parameter is:

8. night free cooling air ventilation flow rate – NFCac.

The parametric analysis is carried out by varying each one of thebove reported parameters in a suitable range by a discrete varia-ion step while the remaining parameters are kept constant to theirnitial value. Initial value, ranges and variation steps of each inves-igated parameter, are reported in Table 5. They were selected byaking into account the Italian regulation requirements and someechnical feasibility and market criteria. In the same table, the setf parameters taken into account for the reference scenario (CASE) are also included.

In order to find out the optimal set of parameters, the selectedomprehensive objective function is the maximum annual PES (pri-ary energy saving). It corresponds to the minimum yearly EP

primary energy consumption) for the indoor space heating andooling. For each thermal zone and for the whole building, EP is cal-ulated by taking into account the electricity consumption of an airo air heat pump/chiller. In particular, it is assessed as:

P =∑

�

QHVAC (�)/COP(�)�el,conv

(16)

here QHVAC is the calculated heating/cooling demand, COP is the
ystem coefficient of performance (variable as a function of theperating conditions as defined by the manufacturers) and �el,convs the conventional Italian average electricity production efficiencyt power plant (46%).
0.450.303

The PES (primary energy saving) obtained through a suitable setof parameters vs. the reference configuration ones, is calculated as:

PES = 1 − EP

ERSP

(17)

where ERSP is the primary energy consumption obtained for the

reference CASE 0.The presented simulation model also includes the assessment

of the greenhouse gases emissions and savings, using the CO2standard and LCA (life cycle analysis) emission factors, taken fromEU official data [110]. Such factors take into account the CO2 emit-ted during system operation and, in case of the LCA approach, theCO2 produced during the system manufacturing process. Accordingto the EU data [110], the adopted emission factors for the calcula-tion of 1 kg of CO2 per 1 kWh of consumed energy are: (i) 1 = 0.480(standard) and 0.708 (LCA), for the energy consumption due to elec-tricity (Eel); (ii) 2 = 0.205 (standard) and 0.237 (LCA) for that onedue to natural gas (Egas). Obviously, the LCA CO2 emissions factorsof photovoltaic and solar thermal collectors are equal to zero [111].Therefore, the CO2 emission is calculated as:

mCO2 = (1 · Eel + 2 · Egas) (18)

4. Results and discussion

In the following, the results of the carried out analyses arereported. For sake of brevity, due to the huge amount of resultingdata, only the most important findings are described.

4.1. Dynamic energy performance simulation: optimizationprocedure

The results of such analysis are reported as a function of: (i) heat-ing, cooling and total primary energy consumptions EPH, EPC and EPT,respectively; (ii) relative primary energy savings, PES. Note that, inthe following figures a PES equal to zero corresponds to the set ofparameters selected for the reference CASE 0. Obviously, a posi-tive (negative) PES corresponds to an enhancement (worsening) ofthe energy performance. Notice that such investigation should bealso carried out through a life cycle and embodied energy analy-sis approach of all the innovative energy saving technologies takeninto account. Although such investigation is out of the scopes of this
paper, the following remarks can be considered: (i) many buildingelements will be obtained by recycled materials (mortars, plaster-boards, concretes); (ii) for all the building elements (glasses, metals,PCM, thermal insulation, etc.) specific studies are available in


ing an

lmeir

4

iocmtaaWtnw

Fig. 7. West office: (a) heating, (b) cool

iterature on their Life Cycle Assessment, Life Cycle Energy Assess-ent and Life Cycle Carbon Emissions Assessment, also for low

nergy buildings [111,112]. In the following the influence of thenvestigated parameters of the carried out parametric analysisesults is discussed.

.1.1. Thermal insulation and massThe increase of the insulation levels is a recognized useful aspect

n the design of building components. At the same time the effectf dynamic thermal properties on the energy performance must beorrectly estimated when optimizing the overall building perfor-ance, as reported in [39]. Therefore, the influence on the PES of the

hickness of the external wall thermal insulation layer (InsThwall)nd of the weight of the external wall semi-hollow bricks (Wghwall)re here investigated. Such effects are reported in Fig. 7 for the
est ground office (thermal zone 3, Fig. 6). Such figure shows
hat for any Wghwall by increasing the thermal insulation thick-ess an increase of the heating PES is always achieved (Fig. 7a),hile the opposite occurs for the cooling one (Fig. 7b), as expected.

Fig. 8. Expo room: (a) heating, (b) cooling an

d (c) total PES vs. InsThwall and Wghwall .

By varying Wghwall two opposite effects are obtained: (i) for lowInsThwall, a higher Wghwall is always preferred; (ii) for high InsThwall,the lower the Wghwall the higher the PES. This behaviour is basicallydue to the non-residential (high internal gains, etc.) and non-continuous use of the building. Note also that the optimal Wghwalldepends on other investigated parameters, as explained in the fol-lowing. In Fig. 7b a slight variation of the cooling PES vs. InsThwallfor any Wghwall is observed. As a consequence, the trend of the totalPES basically follows the heating one (Fig. 7c). Note that the totalPES ranges from −3 to 5.6% for office spaces (thermal zones 3 and5, Fig. 6), from −3.9 to 6.6% for the conference room (thermal zone1, Fig. 6) and from −4.5 to 6.7% for the expo space (thermal zone2, Fig. 6). The calculated optimal InsThwall and Wghwall are about8.5 cm and 800 kg/m3. Additional advantages are not expected foran increase of thermal insulation thickness and wall bricks density.
Note that similar PES trends are obtained also for the other buildingthermal zones.
Fig. 8 shows PES vs. InsThroof and Wghroof for the expo space (ther-mal zone 2, Fig. 6). Such profiles are also representative of the PESs

d (c) total PES vs. InsThroof and Wghroof .

and Buildings 121 (2016) 318–343 333

ototbAndi(hblhCst8t

ImIfeItpT

4

iicbah(npsittimcwa

tfiealcpap

ibPu


btained for the conference room (thermal zone 1, Fig. 6) varia-ion of InsThroof and Wghroof mostly affects the thermal behaviourf these thermal zones. Through this figure it is possible to observehat rising InsThroof, an increase of the PES is always obtained foroth the heating and cooling seasons (Figs. 8a and b, respectively).

similar result is obtained for the investigated Wghroof. It is worthoting that the trend of the cooling PES related to the roof (Fig. 8b)iffers from the one detected for vertical walls (Fig. 7b). Such result

s mainly due to the high solar absorption coefficient of the roofdue to the roof integrated solar field) which causes indoor over-eating during summer. Such effect can be successfully reducedy increasing the thermal insulation thickness InsThroof. The calcu-

ated optimal InsThroof is about 11 cm. It must be also noted that theigher Wghroof, the higher the heating and cooling PES. Note that forASE O, the conductivity, density and specific heat of the concretelab are equal to 0.640 W/mK, 1050 kg/m3 and 840 J/kgK, respec-ively; for the semi-hollow brick, they are equal to 0.600 W/mK,00 kg/m3 and 840 J/kgK; for the thermal insulation, they are equalo 0.023 W/mK, 40.0 kg/m3 and 1290 J/kgK.

The best position of the thermal insulations, InsPswall andnsPsroof, within the stratigraphy was also investigated in terms of

aximum PES. In particular, for the heating demands the optimalnsPswall resulted to be the external one (configuration 1, Table 4)or the office zones, while the internal one for the conference andxpo rooms (configuration 5, Table 4). For the roof, the optimalnsPsroof is always the external one (configuration 1, Table 4). Fromhe maximum cooling PES, InsPswall and InsPsroof must be alwaysositioned at the external side of the stratigraphy (configuration 1,able 4), as better explained in the following.

.1.2. PCM position and thickness in opaque elementsThe adoption of phase change materials for improving the build-

ng energy efficiency was analyzed. For this purpose, a PCM layers embedded into the building perimeter envelope (within verti-al walls and roof stratifications). Through the parametric analysis,oth the PCM thickness (PcmThwall) and position (PcmPswall) werenalyzed. It must be noted that the PCM layer effects on both theeating and cooling demands depends on the material activationas a function of the charging and discharging cycle). Such phe-omenon is governed by the PCM melting temperature range andeak. In this analysis, the PCM features were selected for the coolingeason applications. This choice is due to the higher building cool-ng demands vs. the heating ones. Therefore, the selected meltingemperature ranges between 19 and 28 ◦C while the peak meltingemperature is 26 ◦C. The simulation data are referred to compos-te material panels obtained by mixing gypsum with PCM paraffin

icrocapsules (Micronal® – BASF). Here, the content of PCM micro-apsules is about 42% of the whole mass fraction of the panel,hose thermal properties (specific heat, density and conductivity)

re accounted by the correlations reported in [113].The PCM panel temperature is computed for each simulation

ime step. The phase change growth of the specific heat of the paraf-n microcapsules is calculated through Eq. (9). Here, a and c are setqual to 28.9 kJ/kgK and 2.8 K for T < Tp and equal to 27.7 kJ/kgKnd 1.0 K for T ≥ Tp, respectively. The specific heat of solid andiquid phase, is set equal to 2.5 and 2.0 kJ/kgK, respectively. Theooling/solidification curve is shifted of 1 ◦C towards lower tem-eratures vs. the melting one [114]. The microcapsules densitynd conductivity are set equal to 980 kg/m3 and 0.14 W/mK (liquidhase [113]).

Note that for the Mediterranean climate the optimal PCM melt-
ng temperatures range between 18 and 22 ◦C for winter andetween 24 and 32 ◦C for summer [115]. Therefore, an optimalCM application for the whole year should be attained through these of two different PCM panels, to be separately designed for the
Fig. 9. July 3rd–July 6th: inside surface temperatures (West-wall of the West office).

heating and the cooling seasons. On the other hand, this solution ispresently unfeasible because of the high cost of such materials.

The effects of PCM adoption in wallboard can be observed inFig. 9. Here, for the West ground office, the free floating (switchedoff HVAC system) temperature of the interior surface of the Westperimeter wall, with and without PCM, is depicted for threesummer sample days. In this figure, the temperature amplitudereduction and the time delay of the PCM composite wall vs. the tra-ditional one are clearly visible. Note that during the cooling season,the maximum difference of the indoor air temperatures, calculatedwith and without PCM, is 2.4 ◦C, in accordance with [45]. At theleft top side of Fig. 9, the specific heat of the simulated PCM vs.the related temperature is reported. Figs. 10 and 11 show the dis-tribution of wall temperature in free floating regime, across Eastperimeter wall of the East ground office for June 13th, without andwith the simulated PCM panel, respectively (wall configurations 1and 4 in Table 4). On the left side of these figures, the simulatedhourly temperature profiles for the internal and external wall sur-faces and the indoor and outdoor air are depicted. Correspondingly,the temperature field across the wall is depicted for every hourof the day. In Fig. 10 a slight reduction of the temperature fromthe external wall surface to the internal one is observed. On thecontrary, in Fig. 11, the adoption of 3 cm of a PCM layer implies:(i) a significant reduction of the temperature across the wall, (ii)a very small fluctuation of the internal surface temperature thatalways approaches 26 ◦C during the day. In particular, by compar-ing Fig. 10 vs. Fig. 11 it can be observed that the additional heatcapacity obtained by PCM reduces the internal surface tempera-ture fluctuation up to 3 ◦C. It must be also noted that, due to thehot summers (with also high night temperatures), charging anddischarging processes are not always completed.

The parametric analysis carried out for assessing the optimalPCM position in the external walls and roof stratification (throughthe investigated parameters: PcmPswall and PcmPsroof, respectively)shows that such position (minimizing the total heating and cool-ing demand) is the interior and the exterior one for all the verticalwalls and roof, respectively. As a consequence, the optimal positionof the roof thermal insulation (InsPsroof) corresponds to configura-tion 2 (Table 4). Here, the PCM is between the external coveringand the thermal insulation layer. For the conference room and theexpo space, in the perimeter walls the thermal insulation (InsPswall)must be positioned as in configuration 8, while for both the Eastand West ground offices, as shown in configuration 4 (Table 4). All
such results are summarized in Table 5. Basically, the differencesamong the thermal zones behaviours are due to the diverse simu-lated occupancy patterns and expositions [116]. In particular, the


Fig. 10. East-wall without PCM, July 13th: wall surfaces, indoor and outdoor temperature time history and wall temperature spatial distribution.

door t

cto

fi

F

Fig. 11. East-wall with PCM, July 13th: wall surfaces, indoor and out

onvenience to suitably locate the PCM depends on which building
hermal solicitation is higher between the internal and the externalne.
In Fig. 12, the obtained cooling PESs for all the investigated con-gurations of the perimeter vertical walls stratigraphy and for all

ig. 12. PES vs. InsPswall and PcmPswall (for the wall stratigraphy see Table 4).

emperature time history and wall temperature spatial distribution.

building thermal zones are reported. By such results a comparisonamong thermal zones and wall stratigraphy typology can be carriedout.

The calculated optimal thickness of the PCM panels includedin the walls (PcmThwall) and roof (PcmThroof) providing the low-est energy demand is 3.0 cm. This result is also due to the highthermal conductivity of the selected PCM. Notice that higher PCMthicknesses decrease the system aptitude to charge/discharge thebuilding heat.

For the adopted summer PCM melting point range (between 19and 28 ◦C), according to [117], a positive effect during the heat-ing season is also obtained. In fact, the solidification point is aboveand close to the required indoor air temperature set point (set to20 ◦C). Simulations show that, by comparing the results obtainedthrough the use of PCM with those of the CASE 0 (without PCM),a small difference is achieved. In particular, the calculated heat-
ing PES ranges from 2.3 to 4.1% (as a function of the investigatedthermal zones) and basically it is obtained at the beginning and atthe end of the heating season. In fact, in those periods the build-ing indoor air temperature often falls within the comfort range


F rd

P

(ototaatstdtts

chfl1t�ebt

4

srfibivosUfmgp(a

ig. 13. May 3 : average temperature and efficiency of the PV cell with and withoutCM.

between 20 and 26 ◦C) without the use of the HVAC system. Thisccurrence is due to: (i) the temperate winter climate (mild outdooremperatures and quite high solar radiation); (ii) the low influencef the outdoor temperature (due to the adopted thermal insula-ion); (iii) the remarkable internal gains (due to equipment, lightsnd occupants). Note that, the peaks of the winter daytime indoorir temperatures are averagely lower than those obtained withouthe PCM adoption, because of the latent energy stored in the con-idered PCM panels. Obviously, such latent heat is then transferredo the indoor air balancing part of the heating demand occurringuring the late afternoon hours. The obtained yearly PES is due tohe cooling season too. In fact also during this time the indoor airemperature often falls within the comfort range without the HVACystem need (as it is detectable in Fig. 11).

An interesting result concerns the behaviour of BIPV devicesoupled to PCM panels. In this regard, Fig. 13 shows the hourly timeistories of the PV cell temperature (TPV) and electric efficiency (�c)

or a spring sample day. The reported results are obtained by simu-ating (without and with PCM) the roof stratigraphy configurations

and 2 (Table 4). In this figure it is clearly shown that the lowerhe average PV cell temperature Tc the higher the electric efficiencyc. The adoption of the PCM layer increases the PV yearly electricfficiency up to 4.6%. With configuration 2 the electricity producedy the BIPV system (PPV) resulted equal to 19.3 MWh/y, 4.2% higherhan the one achieved without PCM (standard configuration 1).

.1.3. Glazing type of the sunspace exterior windowsThe glazing type of the sunspace exterior windows, Typgl1, are

elected in order to optimize the transmission of the incident solaradiation (e.g. through glazing with a high Solar Heat Gain Coef-cient, SHGC) and to minimize the related heat loss. To resolveoth the above mentioned problems, for all the investigated glaz-

ng types (Typgl1 reported in Table 2) similar SHGCs and solar andisible transmittances are taken into account (with the exceptionf the glazing type 6). As a consequence, the differences amongimulation results can be basically ascribed to the variation of the-value and infrared radiation reflectance, respectively. In Fig. 14,

or the two office zones adjacent to the sunspace, the heating pri-ary energy consumption (EpH) is plotted for all the investigated

lazing typologies (Typgl1). For the considered case study, the besterformances are obtained with the low-� glazing types 3 and 4Table 2). Here, the higher SHGC of type 3 is almost counterbal-nced by the lower type 4 U-value. From this point of view, the

Fig. 14. EPH heating primary energy consumption vs. Typgl1 .

EpH of type 6 (although the related very low U-value = 0.9 W/m2K)resulted higher than those of types 3 and 4. Obviously, the heatingconsumption of the conference room and expo space are slightlyinfluenced by the Typgl1 variation. As an example, for the expospace, EpH shifts only from 1.74 kWh/m3y for glazing type 3 to1.83 kWh/m3 y for glazing type 1.

4.1.4. Sun space and windows sizesIn winter, by reducing the sunspace width (Wdsg) with glazing

type 1 a maximum heating PES of 8.5% (related to the zone attachedto the sunspace) is obtained (CASE OPT). Conversely, adopting glaz-ing types with lower U-values even higher energy consumptionsvs. CASE 0 can be obtained. These different behaviours depend onthe combination of two opposite phenomena such as solar gainsand heat transfer, which are strongly influenced by the sunspaceshape (surface to volume ratio). Note that, lowering Wdsg a solargain growth is obtained for the increased incident solar radiationon the wall separating the sunspace and the adjacent offices (namedsunspace to indoor wall (South) in Table 1). In summer, the lowerthe width of the open porch (obtained by the windows openingof the sunspace), the lower the solar shading due to the relatedoverhang. Therefore, an increase of the cooling demand is alwaysachieved.

By increasing the window to wall surface area ratio (W/W%),higher building transmission heat transfer and solar heat gains aresimultaneously obtained. Obviously, the optimal W/W% is strictlyconnected to the above discussed Wdsg. The results of the carriedout parametric analysis are reported in Fig. 15. Here, the primaryenergy consumptions vs. Wdsg and W/W%, for the west office areshown. During winter (Fig. 15a and b), by increasing Wdsg, two dif-ferent results are obtained: (i) an increasing EpH for low W/W%; (ii)a EpH minimum for high W/W%. In summer, by increasing W/W% andby decreasing Wdsg, a growth (almost linear) of the cooling demand(EpC) is always observed (Fig. 15c). In Fig. 15d, the total (yearly)energy requirement (EpT) vs. Wdsg for different W/W% is shown.Note that the optimal W/W% and Wdsg are 70% and 3.0 m, respec-tively. In addition, by the carried out analysis resulted that theoptimal detected W/W% of 70% is sufficient to obtain the standarddaylighting requirements of 500 lux on the offices desks for morethan 75% of the yearly working hours. A remarkable increase of day-lighting was obtained by lower Wdsg. Nevertheless, the calculatedlighting energy reduction resulted to be not counterbalanced bythe increase of the heating and cooling demands obtained for lowerWdsg. Such results are supported by those obtained by a detaileddynamic daylighting analysis carried out with a commercial sim-
ulation software (DIVA 2.0), which will be presented in a futurepaper. Note that with DIVA 2.0 the interaction with the buildingthermal behaviour cannot be assessed.


ions vs. W/W% and Wdsg . (a and b) EPH; (c) EPC; (d) EPT .

4

aaAliaya(tbgifoSCwrhrw

4

rgiiprtT

Fig. 15. West office: primary energy consumpt

.1.5. Sunspace interior wall and floorThe thermal behaviour of the winter sunspace (summer porch)

lso depends on the solar absorption coefficients of the floor (FAbsg)nd wall separating the sunspace and the adjacent offices (SWAbsg).s expected, simulation results show that the higher SWAbsg the

ower the heating requirement. The cooling demand resulted quitendependent of such parameter because of the high solar radiationngles and the porch overhang action. Note also that, the overallearly PES, obtained modulating SWAbsg and FAbsg, ranges from 0.7nd 1.6%. The resulted optimal SWAbsg and FAbsg are 0.6 and 0.45red painted wall and light grey floor, respectively). The effect onhe building energy performance of the thickness of the indoor walletween the sunspace and the offices (SWThsg) was also investi-ated. The calculated optimal thickness corresponds the maximumnvestigated one of 30 cm (higher thicknesses were not investigatedor space reasons). A light impact on the yearly energy demand isbtained through the SWThsg modulation. In particular, by varyingWThsg in the investigated range (Table 5) the yearly PES (respect toASE 0) resulted ±1.1%. Note that, the correspondent U-values andall weights vary between 2.2 to 1.6 W/m2K and 482 to 587 kg/m2,

espectively. Here, the higher the U-value the higher the wintereat transfer from the sunspace to the offices and the higher theelated summer internal gains discharge. At the same time a heavy-eight wall strongly decreases the summer cooling demand.

.1.6. Glazing type of the other building windowsAs for the sunspace windows, also for the remaining ones are

equired to optimize the ratio between heat transfer and solar heatains. The results of this carried out parametric analysis are shownn Figs. 16 and 17. In Fig. 16 the yearly primary energy consumptions reported for all the investigated glazing types (Typgl2). Here, it is
ossible to observe that for the East office and for the conferenceoom the best energy performance is achieved through the glazingype 3 (best ratio between SHGF (0.58) and U-value (1.6 W/m2K),able 2). This result is clearly visible also in Fig. 17 where for such
Fig. 16. Yearly primary energy consumption (EPT) vs. Typgl2 .

building zones the heating, cooling and total energy demand arereported for all the investigated glazing types. For the remainingbuilding indoor spaces the best solution is achieved with the glaz-ing type 6 (best ratio between SHGF (0.46) and U-value (0.9 W/m2K),Table 2). Note that for the West office a different behaviour in termssolar heat gain is achieved vs. the East one. This result is due to theshading vertical fin located at the West side of the sunspace, Fig. 5.During summer higher solar angles occur vs. winter ones and therequired solar radiation control is obtained at the ground floor bythe above discussed porch overhang. At the first floor such effect is
achieved by the modelled windows shadings and the related 1 mwidth overhang. Here, the cooling energy consumption is slightlyinfluenced by Typgl2. This result is clearly visible in Fig. 17 where


Fe

dmiboNrz

waTmTpNrtmiittfhbti

4

(nhiccipNahc1tn

ig. 17. East office (a) and conference room (b): heating, cooling and total primarynergy consumption vs. Typgl2 .

ifferent scales on the graphs vertical axes were adopted. The maxi-um total yearly PES (CASE OPT) resulted equal to 1.3%. Such result

s calculated respect CASE 0 (where Typgl2 = 3 is adopted in all theuilding thermal zones) and it is obtained with Typgl2 = 3 in Eastffice and conference room and Typgl2 = 6 in the remaining zones.ote that, PES ranges from −7.2 to 0.6%, for Typgl2 equal to 6 and 1,

espectively, if the same glazing type is adopted in all the thermalones.

At last, windows with double glasses filled with PCM paraffinax were also taken into account for the conference room. All the

ccounted system thermophysical properties are referred to [97].he adoption of such devices was hypothesized only for this ther-al zone because of the related low required level of lighting (see

able 3) and the unnecessary windows transparency. Although theromising potential of such innovative technology (especially forZEBs [46]) the performed sensitivity analysis showed that the

eduction of the incoming solar radiation vs. traditional glazingypes implies a decrease of the overall obtained energy perfor-

ance. This effect is due to the growth of the heating demand thatn the considered case study resulted always higher than the cool-ng one, Fig. 17. Note that such growth is not counterbalanced byhe cooling requirement reduction. This occurrence is also due tohe effect induced by the 1 m width overhang modelled on the con-erence windows. Nevertheless, in case of residential use or eveningours building utilization an improvement of the system thermalehaviour is expected (e.g. winter stored solar energy transferredhe indoor space during evening hours, summer vantages of thencreased building thermal inertia).

.1.7. Night free cooling ventilationIn this section, the influence on the night free cooling ventilation

NFCac) on the building energy performance is discussed. Naturalight ventilation is modelled in summer during the non-occupancyours in order to discharge the energy stored in the envelope dur-

ng daytime (also enhancing the PCM heat release). In Fig. 18, theooling PES vs. the night ventilation flow rate (NFCac, volume airhanges per hour) and the density of the external walls (Wghwall)s shown. As it is possible to observe, for low volume air changeser hour a PES maximum is obtained for Wghwall = 1050 kg/m3. ForFCac = 1.0 h−1 the maximum reduction of the cooling demand isbout 3.7%. For higher NFCac the heavier the external walls, theigher the PES. For NFCac = 3.0 h−1 the maximum reduction of the
ooling demand reaches 4.1%. Remind that an optimal Wghwall of050 kg/m3 resulted from the optimization analysis carried out onhe whole building for CASE 0. Summarizing, with high summeright ventilation rates the higher the external walls and roof masses
Fig. 18. Conference room: cooling PES vs. NFCac and Wghwall .

the lower the cooling demands. Without summer night ventilationthe best performance is obtained through lower wall masses. It isworth noting that, during Mediterranean climate summer months,night outdoor dry bulb temperatures are relatively high. Thus, somedifficulties occur to completely discharge the stored thermal energyof the building envelope (with and without PCM). Because of this,in the considered case study, only a slight summer energy saving isobtained by adopting PCM building elements and night ventilation.

4.1.8. Energy and economic savingsThe main results of the carried out parametric analysis are sum-

marized in Tables 5 and 6. In particular, in Table 5 the obtainedset of parameters that minimize the heating and cooling require-ments, (CASE OPT) are shown. In Table 6, for both the CASE 0and OPT the heating and cooling demands (not primary) and theelectricity demands for lighting, appliances and ventilation, forthe whole building and for the main investigated thermal zones,are reported. As it is possible to observe, very low heating andcooling demands are obtained. In particular, for CASE 0 the cal-culated heating and cooling yearly demands of the whole buildingare 4.8 and 8.1 kWh/m2 y, respectively. For CASE OPT they are 3.8and 6.6 kWh/m2 y. For the heating and cooling demands the yearlysavings correspond to 15 and 18%, respectively (obviously, suchresult is influenced by the non-residential building use). Noticethat, for CASE OPT the energy demand for the artificial lightingresulted 46.9% of the primary energy demand due to space heat-ing and cooling, while about 17.6% of the total primary energydemand (also including appliances and ventilation). In Table 6the energy demands (reported also in kWh/m3 y because of thenon-standard floors heights and building shape) can be taken intoaccount as NZEB definitions for southern Europe zones (Mediter-ranean climates). Although the obtained results cannot yet beverified through measurements, they were compared with thoseprovided by BPIE (Buildings Performance Institute Europe). Notethat, these results were achieved through suitable numerical sim-ulations carried out for one reference multi-storey office buildinglocated in three different European climate zones (Copenhagen,Stuttgart and Madrid). Hence, a comparison is possible just for theoffice spaces. Despite of some differences due to the adopted oper-ating conditions and climates, a satisfactory agreement is observedby comparing the obtained results (Table 6) vs. the BPIE ones. Inparticular, for heating, DHW and appliances the agreement is verygood. The higher cooling demand is caused by the hotter sum-mer climate occurring in Naples vs. the weather zones selected by
BPIE (even Madrid). The lower lighting and ventilation demandsare instead due to the developed smart daylighting control andto the too high ventilation requirement set by BPIE for NZEBs[19,118], respectively. Nevertheless, it is rather difficult to compare


Table 6Calculated specific energy demands for the NZEB in Naples.

Mode Unit Whole building West office East office Conference room Expo space

Case

0 OPT 0 OPT 0 OPT 0 OPT 0 OPT

HeatingkWh/m3y 1.1 0.9 1.5 1.3 1.6 1.4 1.3 1.1 1.2 0.9kWh/m2y 4.8 3.8 5.0 4.1 5.2 4.5 7.3 5.8 6.2 4.8

CoolingkWh/m3y 1.8 1.5 3.2 2.6 3.3 2.7 1.1 0.9 2.3 1.9kWh/m2y 8.1 6.6 10.7 8.6 11.0 8.9 5.9 4.9 12.5 10.3

LightingkWh/m3y 0.9 0.8 1.9 1.7 1.5 1.4 0.2 0.2 0.4 0.4kWh/m2y 3.9 3.5 6.3 5.7 5.1 4.7 1.2 1.2 2.3 2.3

ApplianceskWh/m3y 1.4 5.9 5.9 0.4 –kWh/m2y 6.0 19.4 19.4 2.2 –

VentilationkWh/m3y 0.6 1.1 1.1 0.3 0.6kWh/m2y 2.6 3.7 3.7 1.7 3.3

aisN

ptm(eitIiwpfoan((w

DHWkWh/m3y 0.3 0.5

kWh/m2y 1.2 1.7

nd overlap specific energy demands of different NZEBs locatedn different climates, as well as of similar NZEBs which easily canhow different behaviours [7]. Note that, data about non-residentialZEBs in Mediterranean climates are still unavailable in literature.

The weight of each investigated measure on the building energyerformance, varying as a function of the investigated parame-ers ranges, is shown in Fig. 19. Here, for each single investigated

easure and for the CASE OPT solutions, the energy performanceranging from the maximum PES to the maximum primary energyxtra-demand, both vs. CASE 0) is shown. As it is clearly shownn such figure, a significant impact on the PES is obtained throughhe thermal insulation of perimeter walls and roof (InsPswall andnsPsroof) and the sunspace exterior glazing type (Typgl1). A mediummpact is instead achieved by implementing PCM in the perimeter

alls and roof and by varying its position in the related stratigra-hies (PcmPswall). A slight effect is also obtained through the nightree cooling ventilation (NFCac). A lower PES sensitivity impact isbserved vs. the weight of the external walls and roof (Wghwallnd Wghroof). At last, a weak influence is resulted vs. the thick-
ess of the indoor wall between the sunspace and the officesSWThsg) and the solar absorption coefficients of the related floorFAbsg). Respect to the CASE 0 of the considered case study aorsening effect (higher energy demand) is observed varying the
Fig. 19. PES, extra costs and economic savings of

0.5 0.1 0.21.7 0.3 1.0

initial window to wall surface area ratio (W/W%) and the glazingtypes of the remaining building windows (Typgl2) that in this anal-ysis are considered the same for all the windows. Note that respectto the CASE 0 an increase of the energy demand can be also obtainedby varying the investigated parameters in the considered ranges(SWAbsg and W/W%). In the same figure the influence of all theconsidered solutions for the CASE OPT (Tot) is also shown. Finally,thanks to the parametric analysis, an additional overall PES of about16.9% vs. that one of CASE 0 (which is an already sustainable design)is detected. Note that, obviously, due to the complexity and connec-tion of the investigated phenomena, the superposition of the singleeffects cannot be applied. In the same Fig. 19, a simplified systemeconomic analysis is also reported. In particular, for all the potentialenhancement measures the related extra-cost vs. the investmentone of the CASE 0 are depicted, together with the yearly economicsavings (obtained for the maximum achievable PES).

The net primary energy target is achieved also through a build-ing integrated PV (BIPV) solar system. By such plant an electricityproduction of 18.5 MWh/y is obtained. In addition, the thermal
energy produced through the solar thermal collectors (to be usedfor space heating, cooling and DHW preparation) is 2.92 MWh/y.Note that, due to the large solar field, such energies amount sur-passes the overall building demand (the surplus of electricity and
the considered energy efficiency solutions.


ction

toOaarbBaagsptrasdwsttBgaeaisdmcoiBtwoe

Fig. 20. Primary energy demand as a fun

hermal energy is exported to a close non-residential buildingwned by the Municipality of Naples). For the CASE 0 and CASEPT the influence on the heating and cooling demands of both BIPVnd integrated solar thermal systems was also analyzed. The resultsre reported in Fig. 20. Here, the energy requirements obtained byeplacing the BIPV system with a BIPV/T one are also depicted foroth the CASE 0 and CASE OPT. In summer, through the modelledIPV/T configuration (Fig. 3) the PV collectors are cooled by an aver-ge outdoor air flow rate of 0.5 kg/s. Such air stream flows in a rect-ngular channel (1.0 m width, 4.5 cm height) underlying and inte-rated to the PV panels. In winter such channel is closed throughuitable air dampers and therefore a closed air gap between the PVanels and the remaining roof stratigraphy is obtained. In Fig. 20he same results are reported for both the CASE 0 and CASE OPTeplacing the PV collectors with tiles (roofing type in Table 1, solarbsorbance = 0.4). In this case, the solar field is modelled througholar thermal collectors only. In the same figure the primary energyemands are reported also by replacing the solar thermal systemith tiles (a traditional roof was taken in to account without solar

ystem technologies). The adoption of PV and solar thermal collec-ors implies a reduction of the heating demands with an increase ofhe cooling one. As an example, by comparing CASE OPT referred toIPV/T + BIST configuration vs. CASE OPT without solar technolo-ies (traditional roof), a 18.6% reduction of the heating demandnd a 8.4% increase of the cooling one are obtained. These differentffects almost counterbalance each other. For the previous example

negligible reduction of the overall primary energy consumptions detected (2.2%), Fig. 20. Obviously, an additional primary energyaving is achieved through the electricity and thermal energy pro-uctions. In particular, the PES due to the PV collectors resulteduch higher than the correspondent heating and cooling energy

onsumptions obtained with or without the building integrationf solar system technologies. For the selected case study, the min-mum yearly energy demand is obtained for the CASE OPT withIPV/T (EpT = 2.41 kWh/m3y). At last, it is interesting to observe that
he cooling energy saving achieved by replacing the BIPV systemith the BIPV/T one (about 6.5%). In this case, the resultant increase
f the electricity production is about +3.7%, obtained due to the PVfficiency growth caused by the BIPV/T air stream cooling.

of the implemented solar technologies.

A comparison analysis is performed in order to check the reli-ability of the developed models. In particular, for a similar BIPV/Tsystem the results of the carried out simulation resulted similar tothose achieved through TRN.SYS 17 (Type 567-6). In addition, forthe case study multi zone building, modelled without solar systemtechnologies, the heating and cooling demands were also calcu-lated with EnergyPlus. The obtained output differences range from2 to 6% for the offices and from 5 to 7% for remaining investigatedthermal zones, for both the heating and cooling demands.

At last it is worth noting that, the proposed work is mostly basedon the minimization of the yearly energy demand, which has aremarkable weight on the overall costs. Therefore, a complete eco-nomic analysis (including a life cycle cost analysis) is postponed toa future investigation, being out of the scope of this paper. Nev-ertheless, only some information about the extra-costs of CASEOPT only (with BIPV/T) vs. CASE 0 are reported in the following.Note that the design of the initial building configuration (CASE 0)follows all the requirements provided by the Italian rule on thebuilding energy efficiency [1]. In fact, all the solutions adoptedfor CASE 0 have to be taken into account as mandatory. There-fore, in this case the energy performance analysis is prior withrespect to the economic assessment, due to the necessity to find outall the measures for the building energy efficiency enhancement(CASE OPT). The results of this comparison analysis are reportedin Fig. 19. Here, it is clearly visible that, the extra-costs of theconsidered energy saving solutions are very low. The only excep-tion is the extra cost related to the PCM implementation, equal to46.4 kD . The total extra-cost for all the accounted enhancementsreaches 48.2 kD , but without PCM it decreases at 1.6 kD . At last,in the same figure the yearly economic savings obtained for eachCASE OPT investigated parameter are also shown. Here, it is clearlyshown that the high PCM extra-costs are not counterbalanced bythe correspondent economic saving (18.6 D /year). Notice that, aneconomic saving of 64.2 D /year is obtained adopting all the consid-ered CASE OPT solutions except the PCM one (not shown in figure
for sake of brevity). Note that, the above mentioned CASE 0 electric-ity production of 18.5 MWh/y is partially utilized in the building,while the rest, equal to 14.7 MWh/y, is sold to the national grid(at 0.08 D /kWh following the present Italian rule). The electricity

3 and Bu

pitTBidod1

brNticeagaaopfIcbliN(tsocaoadmf

4

brataao

•


roduction for final CASE OPT (with BIPV/T) is 19.9 MWh/y, becom-ng 19.2 MWh/y without PCM. In this case, the remaining part ofhe electricity production, equal to 15.8 MWh/y is sold to the grid.he reduction of heating and cooling demands for CASE OPT (withIPV/T and without PCM) vs. CASE 0 (in terms of electricity demand)

s equal to 0.510 MWh/y, while the one also due to the lightingemand is about 0.580 MWh/y. Similarly, the net extra-productionf electricity (PV electricity production minus overall electricityemand) obtained through BIPV/T vs. BIPV (without PCM) is about.27 MWh/y.

It is worth noting that, although the presented work is mostlyased on the energy performance analysis, the above discussedesults can be useful to assess the economic feasibility of the finalZEB configuration (CASE OPT). Nevertheless, as it is well known,

he initial NZEBs investment grows up towards traditional build-ngs. Therefore, specifics policy instruments must be taken intoonsideration by governments to overcome such occurrence. As anxample, incentives are nowadays essential for the cost effectivedoption of innovative energy saving measures (e.g. solar technolo-ies, PCM wallboards, etc.). On the other hand, in the last years,

growth of the market of such technologies has been detectednd, thanks to their high potential use in buildings, a reductionf production and installation costs is expected. Nevertheless, theroposed NZEB is a pilot project and will be a demonstrative toolor very efficient buildings in the temperate Mediterranean climate.n addition, the priliminary NZEB configuration (CASE 0) was con-eived through several design and operating parameters imposedy the present Italian rules for energy efficiency in buildings. Simi-

arly, concerning the use of the life cycle assessment tool, althoughts importance in defining the NZEB concept, such approach forZEBs is still outside the current intention of the European EPBD

e.g. [1,2]), as also reported by several authors (e.g. [4,5]). For allhese reasons, a comprehensive cost optimal analysis of the pre-ented NZEB, as well as its life cycle assessment, are out of the scopef this paper. Nevertheless, the CO2 emission/savings (mCO2 ) werealculated by taking into account the standard and the life cyclenalysis emission factors, above reported. Note that in the devel-ped case study, only the electricity consumption was taken intoccount. Since, as above mentioned, the amount of energies pro-uced by renewable sources surpasses the overall building demand,CO2 is a saving. In particular, it resulted equal to 14.8 and 10.0 t/y

or the standard and LCA approach, respectively.

.2. Semi-stationary energy performance analysis

The heating and cooling energy requirements of the case studyuilding were calculated by following the semi-stationary methodeported in EN 13790 (Italian release) [119,120]. Through suchpproach, the assessment of the heating primary energy consump-ion and the cooling demand (not primary energy), as well as thechieved energy performance class, are carried out by taking intoccount a 24 h HVAC system running (standard procedure). Thebtained results are:

in winter, the calculated heating primary energy demands are2.43 and 1.92 kWh/m3y for CASE 0 and CASE OPT, respectively.Presently, in Italy such results correspond to the so-called energyperformance classes “A” and “A+′′

for non-residential buildings,respectively. Note that, for the investigated case study, class“A+′′

(the best class) is achieved if the obtained primary energydemand is lower than the 25% of the reference threshold equal
to 7.45 kWh/m3y (calculated as a function of the building shapeand the occurring weather zone). Note that, an average COP ofthe reference HVAC system equal to 2.5 was taken into accountin the calculation;
ildings 121 (2016) 318–343

• in summer, the calculated cooling energy demands for CASE 0 andCASE OPT are 10.11 and 9.39 kWh/m2y, respectively. Such resultscorrespond to the so-called energy performance classes, “II” and“I”. The best performance class, “I”, is achieved if the obtainedcooling demand is lower than the reference threshold equal to10 kWh/m2y.

Obviously, through this simplified method (even by adoptingthe tailored procedure in order to simulate the effective energydemands) many remarkable thermophysical phenomena (e.g. ther-mal inertia, PCM behaviour, etc.) and real operating conditionscannot be taken into account. Only by means of dynamic simu-lations, as above discussed, detailed energy performance analysescan be carried out.

5. Future analyses

Other novel analyses, models and results will be presented ina successive paper. In particular, in order to suitably evaluate theperformance of the presented building from the environmental andeconomic point of view, life cycle assessment (LCA) and life cyclecost analysis (LCCA) will be carried out. In addition, new measuresand systems as well as models will be investigated, such as: (i) inno-vative solutions for envelope (aerogel material in substitution ofthe glazing surfaces [109], etc.); (ii) advanced HVAC systems (heat-ing HVAC plant exploiting the air pre-heated by the BIPV/T system,solar heating and cooling with double stage absorption chiller, lowenthalpy geothermal heat pump/chiller, energy efficient ground toair heat exchange system and summer evaporative cooling equip-ment [109]); (iii) dynamic thermo-hygrometric and visible comfortanalysis of the multi-zone building; (iv) advanced control strategiesfor indoor air temperature and humidity.

6. Conclusions

In this paper a new simulation model for the dynamic energyperformance analysis of multi-zone buildings with integrated inno-vative energy saving techniques is presented. In particular, newmodels are provided for: (i) PCM embedded in opaque build-ing elements and encapsulated in windows glazing; (ii) BuildingIntegrated PhotoVoltaic (BIPV) and PhotoVoltaic/Thermal (BIPV/T)devices; (iii) building adjacent sunspaces; (iv) optimized build-ing daylighting control. The dynamic building energy performancesimulator, called DETECt 2.2 and written in MatLab, is developedfor research purposes. The developed tool enables the energy per-formance analysis of NZEBs equipped with new (or not common)energy saving technologies, strategies and/or materials. A nov-elty of the presented dynamic simulation tool is the possibility toassess on the whole examined building level both the active andpassive influences of all the above mentioned technologies (evenwhen simultaneously adopted). This goal is achieved since all suchenergy saving technologies are modelled in the code as buildingintegrated. The tool also includes a feature for suitable parametricanalyses, which can be easily performed through a single simula-tion run. This feature provides greater support for the identificationof optimal energy efficiency solutions and building design scenar-ios, playing a crucial role for the design of the next generation ofbuildings, such as NZEBs. As a result of such analyses, the optimalset of operating and design parameters (which minimise the heat-ing and cooling energy requirements of the simulated building) isobtained. Such approach allows one to properly analyze and inter-
pret the simulation results while taking into account the influenceof all the analyzed measures on the overall building heating andcooling demands, as a function of: (i) the building use (internal heatgains and ventilation of offices, dwellings, etc.); (ii) the design and

and Bu

of(e

maMaioahS

•

•

•

•

ecm


perating system conditions (wall/roof layers combination, nightree cooling ventilation rate, detailed occupancy scheduling, etc.);iii) the weather zone (hourly data of temperature, solar radiation,tc.).

In order to show the potentiality of the presented simulationodel, a suitable case study was developed. At the best of the

uthor’s knowledge it is referred to the first non-residential NZEB inediterranean climates. A sensitivity analysis related to the design

nd operating parameters with a heavy impact on the overall build-ng energy performance is carried out. Details regarding the energyptimization procedure and the system economic performance arelso provided. All the obtained results will be useful for stake-olders working on non-residential NZEBs in temperate climates.everal interesting and innovative findings can be highlighted:

new NZEB definition details are provided for non-residentialbuildings located in the southern European countries (Mediter-ranean climates). In particular, for offices, conference rooms andexpo spaces novel recommended ranges of energy demandsfor heating, cooling, lighting, appliances, ventilation and DHWpreparation are reported (such ranges can be obtained betweenthe energy demands referred to CASE 0 and CASE OPT in Table 6).Such demands can be suitably balanced by the on-site producedrenewable energies. Details about several design and operatingparameters that could be taken into account in similar NZEBs arereported in Table 3.the adoption of PCM wallboards in opaque elements reducesthe building energy demand. In particular, the minimum energyrequirement is obtained by applying the PCM panels as interiorlayer in the perimeter walls (vs. massive and insulation ones)and as exterior layer into the roof. The best position of the ther-mal insulation depends on the space use. By coupling PCM to theroof BIPV (BIPV/T) panels, a 4.2% (8.1%) increase of PV electricityproduction is obtained. For the developed case study, no energysavings are obtained through the PCM adoption in windows glaz-ing. Currently, the initial cost of such materials, for both buildingwalls and roof applications, is still excessively high for achievingacceptable payback periods;for the presented NZEB the results of the comparison analysisamong the investigated building roof configurations recommendto adopt a BIPV/T systems (70% of the roof surface) with buildingintegrated solar thermal collectors (30% of the roof surface). Thebetter BIPV/T system performance vs. the BIPV one is due to thelower extra-cooling demands and to a slightly higher electricityproduction. In addition to the electricity production a buildingpassive effect is obtained through such systems (higher coolingdemands and lower heating ones). Nevertheless, the yearly pri-mary energy demand of the HVAC system resulted almost equalto the one necessary in case of traditional medium reflectanceroofs. In particular, for the examined NZEB such overall HVACenergy requirement is completely covered by the produced onsite renewable energy;the optimal windows typology obviously depends on the consid-ered building space use. According to the investigated glazingtypes, for sunspace exterior windows, sunspace to East officewindows and conference room windows, the best energy perfor-mance is obtained by adopting low-� 6/13/6 double glasses filledwith Argon (U = 1.6 W/m2 K and SHGF = 0.58). For the remainingwindows, the minimum energy demand is obtained with low-�6/8/6/8/6 triple glasses filled with Krypton (U = 0.9 W/m2 K andSHGF = 0.46).

Very low energy requirements are achieved by the optimal mod-lled NZEB configuration (0.9 and 1.5 kWh/m3y for heating andooling, respectively). The primary energy saving achieved by opti-izing all the considered parameters is 16.9% (13.3% without PCM).

ildings 121 (2016) 318–343 341

For the developed case study, such energy demand can be totallycovered by the electricity produced by the BIPV/T system. It is worthnoting that, the simulated starting building configuration (CASE 0)is a net zero energy building as well. Therefore, in case of buildingsrefurbishments, higher energy savings can be obviously obtainedthrough the considered energy efficiency measures.

Acknowledgments

Authors wish to gratefully acknowledge Action TU1205 (Build-ing Integration of Solar Thermal Systems, BISTS) of the EuropeanCOST (Cooperation in Science and Technology), Transport andUrban Development (TUD), for the sponsorship and the preciousscientific support.

References

[1] European Parliament, Directive 2010/31/EU of the European Parliament andof the Council on the energy performance of buildings (recast) (9 May 2012),2010.

[2] European Parliament, Commission delegated regulation N. 244/2012.Supplementing directive 2010/31/EU of the European Parliament and of theCouncil on the energy performance of buildings (16 January 2012), 2012.

[3] REHVA Federation of European Heating Ventilation and Air-ConditioningAssociations, NZEB technical definition and system boundaries for nearlyzero energy buildings. 2013 revision for uniformed national implementationof EPBD recast prepared in cooperation with European standardizationorganization CEN. Report No. 4, 2013.

[4] S. Deng, R.Z. Wang, Y.J. Dai, How to evaluate performance of net zero energybuilding – a literature research, Energy 71 (2014) 1–16.

[5] M.M. Sesana, G. Salvalai, Overview on life cycle methodologies andeconomic feasibility for nZEBs, Build. Environ. 67 (2013) 211–216.

[6] C. Becchio, P. Dabbene, E. Fabrizio, V. Monetti, M. Filippi, Cost optimalityassessment of a single family house: building and technical systemssolutions for the nZEB target, Energy Build. 90 (2015) 173–187.

[7] J. Kurnitski, F. Allard, D. Braham, D. van Dijk, C. Feldmann, J. Fox, K. Graslund,P. Heiselberg, F. Hovorka, R. Kosonen, J. Lebrun, Z. Magyar, L. Mazzarella, I.Mar-tinac, V. Novakovic, J. Railio, O. Seppanen, I. Sartori, J. Zirngibl, M.Schmidt, M. Virta, K. Voss, A. Wahlstrom, NZEB technical definition andsystem boundaries for nearly zero energy buildings. 2013 revision foruniformed national implementation of EPBD recast prepared in cooperationwith European standardization organization CEN. Report No. 4, REHVA –Federation of European Heating Ventilation and Air-ConditioningAssociations, 2013.

[8] F. Garde, A. Lenoir, A. Scognamiglio, D. Aelenei, D. Waldren, H.N. Rostvik, J.Ayoub, L. Aelenei, M. Donn, M. Tardif, S. Cory, Design of net zero energybuildings: feedback from international projects, Energy Procedia 61 (2014)995–998.

[9] J.S. Bourrelle, I. Andresen, A. Gustavsen, Energy payback: an attributionaland environmentally focused approach to energy balance in net zero energybuildings, Energy Build. 65 (2013) 84–92.

[10] M. Premrov, V. Zegarac Leskovar, K. Mihalic, Influence of the building shapeon the energy performance of timber-glass buildings in different climaticconditions, Energy (2015), http://dx.doi.org/10.1016/j.energy.2015.05.027.

[11] Y. Sun, Sensitivity analysis of macro-parameters in the system design of netzero energy building, Energy Build. 86 (2015) 464–477.

[12] K. Negendahl, Building performance simulation in the early design stage: anintroduction to integrated dynamic models, Autom. Constr. 54 (2015) 39–53.

[13] J. Kurnitski, A. Saari, T. Kalamees, M. Vuolle, J. Niemelä, T. Tark, Cost optimaland nearly zero (nZEB) energy performance calculations for residentialbuildings with REHVA definition for nZEB national implementation, EnergyBuild. 43 (11) (2011) 3279–3288.

[14] A. Athienitis, W. O’Brien, Modeling, Design, and Optimization of Net-ZeroEnergy Buildings, Wiley, 2015.

[15] M.S. Todorovic, BPS, energy efficiency and renewable energy sources forbuildings greening and zero energy cities planning: harmony and ethics ofsustainability, Energy Build. 48 (2012) 180–189.

[16] S. Attia, State of the Art of Existing Early Design Simulation Tools for NetZero Energy Buildings: A Comparison of Ten Tools, Université catholique deLouvain, 2011.

[17] N. Fumo, A review on the basics of building energy estimation, Renew.Sustain. Energy Rev. 31 (2014) 53–60.

[18] D. Kolokotsa, D. Rovas, E. Kosmatopoulos, K. Kalaitzakis, A roadmap towardsintelligent net zero- and positive-energy buildings, Solar Energy 85 (12)
(2011) 3067–3084.
[19] T. Boermans, A. Hermelink, S. Schimschar, J. Grözinger, M. Offermann, BPIE –Principles for nearly-zero energy buildings. Paving the way for EffectiveImplementation of Policy Requirements, B. Building Performance InstituteEurope, 2011.

http://refhub.elsevier.com/S0378-7788(15)30224-3/sbref0590









































































































































































































































































































































dx.doi.org/10.1016/j.energy.2015.05.027




















































































































































































































































3 and Bu
[20] P. Fonseca i Casas, A. Fonseca i Casas, N. Garrido-Soriano, J. Casanovas,Formal simulation model to optimize building sustainability, Adv. Eng.Softw. 69 (2014) 62–74.

[21] S. Attia, M. Hamdy, W. O’Brien, S. Carlucci, Assessing gaps and needs forintegrating building performance optimization tools in net zero energybuildings design, Energy Build. 60 (2013) 110–124.

[22] J. Nembrini, S. Samberger, G. Labelle, Parametric scripting for early designperformance simulation, Energy Build. 68 (Part C) (2014) 786–798.

[23] Z. Szalay, A. Zöld, Definition of nearly zero-energy building requirementsbased on a large building sample, Energy Policy 74 (2014) 510–521.

[24] M. Kapsalaki, V. Leal, M. Santamouris, A methodology for economic efficientdesign of net zero energy buildings, Energy Build. 55 (2012) 765–778.

[25] P.A. Fokaides, E.A. Christoforou, S.A. Kalogirou, Legislation driven scenariosbased on recent construction advancements towards the achievement ofnearly zero energy dwellings in the southern European country of Cyprus,Energy 66 (2014) 588–597.

[26] P. Penna, A. Prada, F. Cappelletti, A. Gasparella, Multi-objectivesoptimization of energy efficiency measures in existing buildings, EnergyBuild. 95 (2015) 57–69.

[27] S. Attia, E. Gratia, A. De Herde, J.L.M. Hensen, Simulation-based decisionsupport tool for early stages of zero-energy building design, Energy Build.49 (2012) 2–15.

[28] F. Reda, P. Tuominen, Å. Hedman, M.G.E. Ibrahim, Low-energy residentialbuildings in New Borg El Arab: simulation and survey based energyassessment, Energy Build. 93 (2015) 65–82.

[29] A. Houlihan Wiberg, L. Georges, T.H. Dokka, M. Haase, B. Time, A.G. Lien, S.Mellegård, M. Maltha, A net zero emission concept analysis of asingle-family house, Energy Build. 74 (2014) 101–110.

[30] I. Zacà, D. D’Agostino, P.M. Congedo, C. Baglivo, Assessment ofcost-optimality and technical solutions in high performancemulti-residential buildings in the Mediterranean area, Energy Build. 102(2015) 250–265.

[31] M. Ferrara, E. Fabrizio, J. Virgone, M. Filippi, A simulation-basedoptimization method for cost-optimal analysis of nearly zero energybuildings, Energy Build. 84 (2014) 442–457.

[32] M. Cellura, F. Guarino, S. Longo, M. Mistretta, Different energy balances forthe redesign of nearly net zero energy buildings: an Italian case study,Renew. Sustain. Energy Rev. 45 (2015) 100–112.

[33] M. Ferrara, J. Virgone, E. fabrizio, F. Kuznik, M. Filippi, Modelling zero energybuildings: parametric study for the technical optimization, Energy Procedia62 (2014) 200–209.

[34] E. Pikas, M. Thalfeldt, J. Kurnitski, Cost optimal and nearly zero energybuilding solutions for office buildings, Energy Build. 74 (2014) 30–42.

[35] A. Mohamed, A. Hasan, K. Sirén, Fulfillment of net-zero energy building(NZEB) with four metrics in a single family house with different heatingalternatives, Appl. Energy 114 (2014) 385–399.

[36] B. Berggren, M. Wall, K. Flodberg, E. Sandberg, Net ZEB office in Sweden – acase study, testing the Swedish Net ZEB definition, Int. J. Sustain. BuiltEnviron. 1 (2) (2012) 217–226.

[37] L. Aelenei, H. Gonc alves, From solar building design to net zero energybuildings: performance insights of an office building, Energy Procedia 48(2014) 1236–1243.

[38] J.A. Clarke, J.L.M. Hensen, Integrated building performance simulation:progress, prospects and requirements, Build. Environ. 91 (2015) 294–306.

[39] N. Aste, F. Leonforte, M. Manfren, M. Mazzon, Thermal inertia and energyefficiency – parametric simulation assessment on a calibrated case study,Appl. Energy 145 (2015) 111–123.

[40] M. Pomianowski, P. Heiselberg, Y. Zhang, Review of thermal energy storagetechnologies based on PCM application in buildings, Energy Build. 67 (2013)56–69.

[41] N. Soares, J.J. Costa, A.R. Gaspar, P. Santos, Review of passive PCM latent heatthermal energy storage systems towards buildings’ energy efficiency,Energy Build. 59 (2013) 82–103.

[42] A. Athienitis, S.A. Kalogirou, L. Candanedo, 3. 11 – Modeling and simulationof passive and active solar thermal systems, in: A. Sayigh (Ed.),Comprehensive Renewable Energy, Elsevier, Oxford, 2012, pp. 357–417.

[43] S.B. Sadineni, S. Madala, R.F. Boehm, Passive building energy savings: areview of building envelope components, Renew. Sustain. Energy Rev. 15 (8)(2011) 3617–3631.

[44] Y. Tripanagnostopoulos, New designs of building integrated solar energysystems, Energy Procedia 57 (2014) 2186–2194.

[45] S. Guichard, F. Miranville, D. Bigot, H. Boyer, A thermal model for phasechange materials in a building roof for a tropical and humid climate: modeldescription and elements of validation, Energy Build. 70 (2014) 71–80.

[46] F. Goia, M. Perino, M. Haase, A numerical model to evaluate the thermalbehaviour of PCM glazing system configurations, Energy Build. 54 (2012)141–153.

[47] J. Heier, C. Bales, V. Martin, Combining thermal energy storage withbuildings – a review, Renew. Sustain. Energy Rev. 42 (2015) 1305–1325.

[48] T. Yang, A.K. Athienitis, A study of design options for a building integratedphotovoltaic/thermal (BIPV/T) system with glazed air collector and multiple
inlets, Solar Energy 104 (2014) 82–92.
[49] B. Agrawal, G.N. Tiwari, Optimizing the energy and exergy of buildingintegrated photovoltaic thermal (BIPVT) systems under cold climaticconditions, Appl. Energy 87 (2) (2010) 417–426.

ildings 121 (2016) 318–343

[50] C. Lamnatou, J.D. Mondol, D. Chemisana, C. Maurer, Modelling andsimulation of building-integrated solar thermal systems: behaviour of thesystem, Renew. Sustain. Energy Rev. 45 (2015) 36–51.

[51] COST Action TU1205 (BISTS), Overview of BISTS state of the art, models andapplications, 2015.

[52] C.J. Ho, A.O. Tanuwijava, C.-M. Lai, Thermal and electrical performance of aBIPV integrated with a microencapsulated phase change material layer,Energy Build. 50 (2012) 331–338.

[53] T. Ma, H. Yang, Y. Zhang, L. Lu, X. Wang, Using phase change materials inphotovoltaic systems for thermal regulation and electrical efficiencyimprovement: a review and outlook, Renew. Sustain. Energy Rev. 43 (2015)1273–1284.

[54] D. Bigot, F. Miranville, A.H. Fakra, H. Boyer, A nodal thermal model forphotovoltaic systems: impact on building temperature fields and elementsof validation for tropical and humid climatic conditions, Energy Build. 41(11) (2009) 1117–1126.

[55] D. Bigot, F. Miranville, H. Boyer, M. Bojic, S. Guichard, A. Jean, Modeloptimization and validation with experimental data using the case study ofa building equipped with photovoltaic panel on roof: coupling of thebuilding thermal simulation code ISOLAB with the generic optimizationprogram GenOpt, Energy Build. 58 (2013) 333–347.

[56] K. Lam, Sustainability performance simulation tools building designsimulation tools for building design building design, in: V. Loftness, D. Haase(Eds.), Sustainable Built Environments, Springer, New York, 2013, pp.526–594.

[57] A. Buonomano, A. Palombo, Building energy performance analysis by anin-house developed dynamic simulation code: an investigation for differentcase studies, Appl. Energy 113 (2014) 788–807.

[58] A. Buonomano, A. Palombo, NZEBs design and simulation: a new tool fordynamic energy performance analyses ECOS 2014, Proceedings of the 27thInternational Conference on Efficiency, Cost, Optimization, Simulation andEnvironmental Impact of Energy systems, Vol. 3, 2014, pp. 1596–1613, ISBN978-1-63439-134-4.

[59] J. Van der Veken, D. Saelens, G. Verbeeck, H. Hens, Comparison of steadystate and dynamic building energy simulation programs, in: Proceedings ofthe International Buildings IX ASHRAE Conference on the Performance ofExterior Envelopes of Whole Buildings, 2004.

[60] A.P. Melo, D. Cóstola, R. Lamberts, J.L.M. Hensen, Assessing the accuracy of asimplified building energy simulation model using BESTEST: the case studyof Brazilian regulation, Energy Build. 45 (2012) 219–228.

[61] A. Buonomano, F. Calise, A. Palombo, Buildings dynamic simulation: Waterloop heat pump systems analysis for European climates, Appl. Energy 91 (1)(2012) 222–234.

[62] A. Buonomano, F. Calise, A. Palombo, Solar heating and cooling systems byCPVT and ET solar collectors: A novel transient simulation model, Appl.Energy 103 (2013) 588–606.

[63] J.W. Chuah, A. Raghunathan, N.K. Jha, ROBESim: a retrofit-oriented buildingenergy simulator based on EnergyPlus, Energy Build. 66 (2013) 88–103.

[64] X. Li, J. Wen, Review of building energy modeling for control and operation,Renew. Sustain. Energy Rev. 37 (2014) 517–537.

[65] A. Ioannou, L.C.M. Itard, Energy performance and comfort in residentialbuildings: sensitivity for building parameters and occupancy, Energy Build.92 (2015) 216–233.

[66] M.J.N. Oliveira Panão, M.P. Rebelo, S.M.L. Camelo, How low should be theenergy required by a nearly zero-energy building? The load/generationenergy balance of Mediterranean housing, Energy Build. 61 (2013) 161–171.

[67] U. Desideri, L. Arcioni, D. Leonardi, L. Cesaretti, P. Perugini, E. Agabitini, N.Evangelisti, Design of a multipurpose “zero energy consumption” buildingaccording to European Directive 2010/31/EU: life cycle assessment, EnergyBuild. 80 (2014) 585–597.

[68] B.Y.H. Liu, R.C. Jordan, The interrelationship and characteristic distributionof direct, diffuse and total solar radiation, Solar Energy 4 (3) (1960) 1–19.

[69] A.L. Pisello, F. Asdrubali, Human-based energy retrofits in residentialbuildings: a cost-effective alternative to traditional physical strategies, Appl.Energy 133 (2014) 224–235.

[70] C.P. Underwood, An improved lumped parameter method for buildingthermal modelling, Energy Build. 79 (2014) 191–201.

[71] MATLAB and Statistics Toolbox Release 2010b, The MathWorks, Inc., Natick,Massachusetts, United States.

[72] J.-M. Mottard, A. Fissore, Thermal simulation of an attached sunspace and itsexperimental validation, Solar Energy 81 (3) (2007) 305–315.

[73] J.L. Wright, A correlation to quantity convective heat transfer betweenvertical window glazings, ASHRAE Trans. 2 (1996) 940–946.

[74] M.H. Sherman, D.T. Grimsrud, Measurement of infiltration using fanpressurization and weather data, in: Fist Symposium of the Air InfiltrationCentre on Instrumentation and Measurement Techniques, Windsor,England, October 6–8, 1980.

[75] R.T. Muehleisen, S. Patrizi, A new parametric equation for the wind pressurecoefficient for low-rise buildings, Energy Build. 57 (2013) 245–249.

[76] J.A. Candanedo, V.R. Dehkordi, A. Saberi-Derakhtenjani, A.K. Athienitis,Near-optimal transition between temperature setpoints for peak load
reduction in small buildings, Energy Build. 87 (2015) 123–133.
[77] A. Buonomano, U. Montanaro, A. Palombo, S. Santini, Indoor air temperaturecontrol in buildings via an optimal tuned PI strategy, Int. J. Eng. Innov.Technol. 4 (4) (2014) 77–85.



























































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































and Bu
[78] M. Di Bernardo, U. Montanaro, J.M. Olm, S. Santini, Model reference adaptivecontrol of discrete-time piecewise linear systems, Int. J. Robust NonlinearControl 23 (7) (2013) 709–730.

[79] A. Buonomano, U. Montanaro, A. Palombo, S. Santini, Multi-zone buildingsthermo-hygrometric analysis: a novel dynamic simulation code based onadaptive control, in: Proceedings of Building Simulation Applications2015-2nd IBPSA-Italy Conference, Bozen, Italy, 2015.

[80] A. Buonomano, U. Montanaro, A. Palombo, S. Santini, Adaptive control forbuilding thermo-hygrometric analysis: a novel dynamic simulation code forindoor spaces with multi-enclosed thermal zones, in: Proceedings of the 6thInternational Conference on Building Physics for a Sustainable BuiltEnvironment, Turin, Italy, 2015 (to be published on Energy Procedia).

[81] F. Azadian, M.A.M. Radzi, A general approach toward building integratedphotovoltaic systems and its implementation barriers: a review, Renew.Sustain. Energy Rev. 22 (2013) 527–538.

[82] S.A. Kalogirou, Solar Energy Engineering Processes and Systems, 2nd ed.,Academic Press, Elsevier, 2014.

[83] V. Delisle, M. Kummert, A novel approach to compare building-integratedphotovoltaics/thermal air collectors to side-by-side PV modules and solarthermal collectors, Solar Energy 100 (2014) 50–65.

[84] E. Skoplaki, J.A. Palyvos, Operating temperature of photovoltaic modules: asurvey of pertinent correlations, Renew. Energy 34 (1) (2009) 23–29.

[85] P. Hoang, V. Bourdin, Q. Liu, G. Caruso, V. Archambault, Coupling optical andthermal models to accurately predict PV panel electricity production, Sol.Energy Mater. Sol. Cells 125 (2014) 325–338.

[86] A.D. Jones, C.P. Underwood, A thermal model for photovoltaic systems, SolarEnergy 70 (4) (2001) 349–359.

[87] W. De Soto, S.A. Klein, W.A. Beckman, Improvement and validation of amodel for photovoltaic array performance, Solar Energy 80 (1) (2006) 78–88.

[88] G. Barker, P. Norton, Building America System Performance Test Practices:Part 1 – Photovoltaic Systems National Renewable Energy Laboratory,Technical Report, NREL/TP-550-30301, 2003.

[89] R.K. Mishra, G.N. Tiwari, Energy and exergy analysis of hybrid photovoltaicthermal water collector for constant collection temperature mode, SolarEnergy 90 (2013) 58–67.

[90] S. Pantic, L. Candanedo, A.K. Athienitis, Modeling of energy performance of ahouse with three configurations of building-integratedphotovoltaic/thermal systems, Energy Build. 42 (10) (2010) 1779–1789.

[91] L.F. Cabeza, A. Castell, C. Barreneche, A. de Gracia, A.I. Fernández, Materialsused as PCM in thermal energy storage in buildings: a review, Renew.Sustain. Energy Rev. 15 (3) (2011) 1675–1695.

[92] P.A. Mirzaei, F. Haghighat, Modeling of phase change materials forapplications in whole building simulation, Renew. Sustain. Energy Rev. 16(7) (2012) 5355–5362.

[93] A. Bastani, F. Haghighat, J. Kozinski, Designing building envelope with PCMwallboards: design tool development, Renew. Sustain. Energy Rev. 31(2014) 554–562.

[94] P. Fokaides, A. Kylili, S. Kalogirou, Phase change materials (PCMs) integratedinto transparent building elements: a review, Mater. Renew. Sustain. Energy4 (2) (2015) 1–13.

[95] R.Z. Freire, W. Mazuroski, M.O. Abadie, N. Mendesa, Capacitive effect on theheat transfer through building glazing systems, Appl. Energy 88 (2011)4310–4319.

[96] M. Liu, K.B. Wittchen, P.K. Heiselberg, F.V. Winther, Development and
sensitivity study of a simplified and dynamic method for double glazingfacade and verified by a full-scale fac ade element, Energy Build. 68 (Part A)(2014) 432–443.
[97] F. Goia, M. Perino, V. Serra, Experimental analysis of the energy performanceof a full-scale PCM glazing prototype, Solar Energy 100 (2014) 217–233.

ildings 121 (2016) 318–343 343

[98] K.A.R. Ismail, J.R. Henrııquez, Parametric study on composite and PCM glasssystems, Energy Convers. Manage. 43 (7) (2002) 973–993.

[99] M. Cucumo, A.D. Rosa, V. Ferraro, D. Kaliakatsos, V. Marinelli, Experimentaldata of global and diffuse luminous efficacy on vertical surfaces atArcavacata di Rende and comparisons with calculation models, EnergyConvers. Manage. 50 (1) (2009) 166–173.

[100] M.S. Mayhoub, D.J. Carter, Methods to estimate global and diffused luminousefficacies based on satellite data, Solar Energy 85 (11) (2011) 2940–2952.

[101] L. Robledo, A. Soler, Modeling the luminous efficacy of diffuse solar radiationon inclined surfaces for all sky conditions, Energy Convers. Manage. 44 (1)(2003) 177–189.

[102] J.F. Kreider, A. Rabl, Heating and Cooling of Buildings, McGraw-Hill, NewYork, NY, 2009.

[103] L. Bellia, F. De Falco, F. Minichiello, Effects of solar shading devices on energyrequirements of standalone office buildings for Italian climates, Appl.Therm. Eng. 54 (1) (2013) 190–201.

[104] A. Tzempelikos, A.K. Athienitis, The impact of shading design and control onbuilding cooling and lighting demand, Solar Energy 81 (3) (2007) 369–382.

[105] CEN European Committee for Standardizations, EN 12464-1:2011. Light andlighting–lighting of work places – Part 1: indoor work places, Brussels,2011.

[106] Municipality of Naples, Resolution of the City Council of Naples n. 517, 2011,http://www.comune.napoli.it/flex/FixedPages/IT/Delibere.php/L/IT/frmSearchHaveData/OK/frmSearchText/-/frmSearchMonth/4/frmSearchYear/2011/frmTargetData/1303336800 (21.04.11).

[107] TVP Solar, http://www.tvpsolar.com/index.php, in.[108] J. Zhao, B. Lasternas, K.P. Lam, R. Yun, V. Loftness, Occupant behavior and

schedule modeling for building energy simulation through office appliancepower consumption data mining, Energy Build. 82 (2014) 341–355.

[109] P. Mazzei, A. Palombo, Economic evaluation of hybrid evaporativetechnology implementation in Italy, Build. Environ. 34 (5) (1999) 571–582.

[110] E.a.E.R., Energy and Environment Report 2008, 2008.[111] C.K. Chau, T.M. Leung, W.Y. Ng, A review on Life Cycle Assessment, Life Cycle

Energy Assessment and Life Cycle Carbon Emissions Assessment onbuildings, Appl. Energy 143 (2015) 395–413.

[112] L.F. Cabeza, L. Rincón, V. Vilarino, G. Pérez, A. Castell, Life cycle assessment(LCA) and life cycle energy analysis (LCEA) of buildings and the buildingsector: a review, Renew. Sustain. Energy Rev. 29 (2014) 394–416.

[113] T. Toppi, L. Mazzarella, Gypsum based composite materials withmicro-encapsulated PCM: experimental correlations for thermal propertiesestimation on the basis of the composition, Energy Build. 57 (2013) 227–236.

[114] G. Evola, L. Marletta, F. Sicurella, A methodology for investigating theeffectiveness of PCM wallboards for summer thermal comfort in buildings,Build. Environ. 59 (2013) 517–527.

[115] F. Ascione, N. Bianco, R.F. De Masi, F. de’ Rossi, G.P. Vanoli, Energyrefurbishment of existing buildings through the use of phase changematerials: energy savings and indoor comfort in the cooling season, Appl.Energy 113 (2014) 990–1007.

[116] B.M. Diaconu, Thermal energy savings in buildings with PCM-enhancedenvelope: influence of occupancy pattern and ventilation, Energy Build. 43(1) (2011) 101–107.

[117] D. Heim, J.A. Clarke, Numerical modelling and thermal simulation ofPCM–gypsum composites with ESP-r, Energy Build. 36 (8) (2004) 795–805.

[118] A. van de Bree, B. von Manteuffel, L. Ramaekers, M. Offermann (Eds.),
ECOFYS – Role of Building Automation related to Renewable Energy innZEB’s, E.C. Institute, Netherlands, 2014, August.
[119] EN ISO 13790:2008, Energy performance of buildings. Calculation of energyuse for space heating and cooling.

[120] UNI/TS 11300:2008, Energy performances of buildings (Part 1 – Part 4).



























































































































































































































































































































































































































































































































































































































































































































































































http://www.comune.napoli.it/flex/FixedPages/IT/Delibere.php/L/IT/frmSearchHaveData/OK/frmSearchText/-/frmSearchMonth/4/frmSearchYear/2011/frmTargetData/1303336800





















http://www.tvpsolar.com/index.php
























































































































































































































































































































Innovative technologies for NZEBs: an energy and economic analysis tool and a case study of a non-residential building for the Mediterranean climate

Documents