SAE2013 IntermediateHeatedTank Corrected (06!07!2013 10-02-48)

Answers To Reviewers

The answers are in RED

Reviewer #: 88632

Interesting paper, and methodology, though the gain obtained by the proposed

system is limited. Hereafter some comments and needed clarifications :

End of P1 : add a space between bythis . DONE

Fig 2 : Axis legend and values too small . All figures have been re sized.

P2, right col parag 3 'as a function of '?? Corrected

Fig 4 : Text too small and should be defined All figures have been re sized.

P4 eq (8) V_prin should be defined Mistake corrected => V_main

Fig 7 bottom : Texh should be Tech ? Corrected

Tabl 1 : Left column difficult to read (black vs dark blue) All Tables have been

changed accordingly to reviewer remark

K_34 should be Forced Conv. (conv. missing ?) Corrected

P5, right col parag 4 'as a function of ' ?? Corrected

Tabl 2 : Id Tabl1 for colors All Tables have been changed accordingly to reviewer

remark

Tabl 3 : The ICE is capable of 100 kW for 1.2 L, that's quite impressive for a

passenger car ? Sorry for the mistake : 70 kW

P7, right col parag 4 'a Heated Intermediate…' ? (2 times) Corrected

Table 6 : Conv consumption is 4,67 L/100, exactly the same value for NEDC and WLTP ? Then an other cycle could have been chosen ? WLTP has been chosen because it will become the new reference cycle in replacement to NEDC. Effectively there is no significant difference in consumption, but the WLTP has a

higher demand in energy.

Could you precise if the engine is cold or hot for all the start of the simulations (NEDC and WLTP). The simulations has been conducted with data maps

established from ‘hot’ steady state experimental conditions .

Fig 11 to 16 : Difficult to exploit, text is too small and grid too visible Figures have

been re sized

P11 : 2nd

paragraph, it should be fig 17 ? Corrected

In conclusion : The effect of the proposed device seem to be small on fuel consumption gain, but validation of the simulations on a test bench would be

interesting.

Reviewer #: 88632

Overall, very well written paper. Clearly explained concept, backed by a good theoretical model and simulation results. Nice contribution

n. Would like to see it verified with Experimental Results. That will be interesting.The comments below are mostly editorial in nature.

1. Page one, “bythis” should be written as by this” Corrected

2. Page 2, after first equation. “Tank” should be “tank” corrected

3. Figure 2 shows clearly the benefits of heating the tank. Nice introduction and problem set up discussion. Thank you

4. Page 3 “Once the engine switch to the …” should be “Once the engine switches to the …” Corrected

5. Page 6 “For confidential reason…” should read “For confidentiality reasons…”Corrected

W What is the WLTP drive cycle ?Worldwide harmonized Light vehicles Test Procedures . Every acronyms have been correctly defined before use.

Graphs are not possible to read in black and white. Consider enlarging the key graphs and making them more understandable. . All figures have been re sized.

Reviewer #: 89166

The paper presents analysis of improving the efficiency of pneumatic hybrid engine with addition of an intermediate heated tank. The proposed idea involves placing an intermediate tank between engine combustion chamber and main tank. The intermediate tank is heated using exhaust gases. Additionally, a bypass valve is added to prevent cooling of intermediate tank when the engine is running in

pneumatic motor mode.

Authors should address following concerns:

1. The citations in the paper are not ordered in the order of their appearance. Please reorder citations appropriately. Corrected

2. From practical standpoint, having an bypass valve would be prone to fouling and sticking. Authors should comment how they plan to address that? Exactly as it is done with a EGR Valve … Reviewer is right, problems will occurs, exactly as those encountered with an EGR Valve.

3. 4. What is m in equation 2? ‘M’, ‘Cp’, ‘Cv’ and ‘r’ have been clearly definedin

the text 5. In thermodynamic modeling section, none of the equation variables are

described and not all variables are part of figure. Like Cp, Cv, r, Vprin? Vprin was a mistake (V main) . Corrected.

6. For simulation results, what kind of fidelity of models was used? What reviewer means ?

7. What is a WLTP cycle? Please use full name first time before

abbreviation.Worldwide harmonized Light vehicles Test Procedures . Every acronyms have been correctly defined before use.

8. In terms of fuel economy, there isn’t much difference between single mainisothermal tank and intermediate tank for both cycles, 0.01l/km? Effectively there is no significant difference…

9. The “hot” simulations don’t make any sense. What was the thinking behind such an experiment, as it doesn’t represent any real world scenario? This experiment unfavorably adds additional energy in the intermediate

tank. Reviewer is right. The thinking was to show that the hotter is the initial state of the tank, the higher are the efficiencies. Effectively those conditions would not be encountered on a Test Procedure, but could be encountered in real life..

10. Did simulations comparing intermediate tank and single main isothermal

tank considered additional weight for the intermediate tank? Also what is the size of main tank used in simulation? Did authors adjust the size of main tank for simulation with no intermediate tank to account for additional storage? The total volume for the pressurized air has been maintained constant : if intermediate tank grows up , main tank decreases.

11. 12. There are few grammatical mistakes like on Page 5, first line, “A thermal

resistance model, as depicted in fig7 is used” for what? Please proof read final draft. We have done the best as we can..

Page 1 of 12

Type Paper Number Here

Improving the Overall Efficiency of a Pneumatic-Combustion Hybrid

Engine by adding an intermediate Heated Tank

Pascal BREJAUD(*), Pascal HIGELIN, Alain CHARLET, Kristan

GILLET, Julien LEMAIRE (Do NOT enter this information. It will

be pulled from participant tab in MyTechZone)

Affiliation (Do NOT enter this information. It will be pulled from participant tab in MyTechZone)

Copyright © 2012 SAE International

ABSTRACT

Several works have previously shown that the concept of

pneumatic-combustion hybrid engine is an interesting alternative to the Electric Hybrid Vehicle, by leading to

equivalent fuel savings for a probable lower cost. However,

these studies have shown that the thermal insulation of the

tank is a problem. Indeed, due to its size and its location, the

adiabaticity of the pneumatic tank cannot be guaranteed.

During a regenerative braking (pneumatic pump mode) the hot

and pressurized air that is send to the tank cools, pressure

drops and density increases. When reusing the air in

pneumatic motor mode, the mass necessary to fill the cylinder

is greater than the one that would have been necessary if the

air was not cool at its stay in the tank. This phenomenon is the major cause to the quite low regenerative efficiency that has

been observed on a prototype engine. This paper proposes and

evaluates a solution to this problem by using an intermediate

air tank heated by the exhaust gases while the engine operates

in the conventional combustion mode. The proposed concept

is developed in detail and 0D thermal and thermodynamic

modeling is proposed. The fuel savings, for the NEDC and

WLTP driving cycles under several thermal initial states of the

intermediate tank, are estimated by numerical simulations.

Finally, the influence of the intermediate heated tank size is

investigated.

INTRODUCTION

Pneumatic Hybrid Engine

Figure 1: The pneumatic-combustion hybrid engine

Nowadays, it is commonly accepted that the hybridization of

Internal Combustion Engine (ICE) is a smart way to reduce

pollutant and greenhouse gas emission. The single technology

encountered in the street is the Electric Hybrid Vehicle

(EHV), but other reversible energies may be considered:

mechanical (flywheels), hydraulic and pneumatic. The concept

of the pneumatic hybrid engine, which has been investigated

by several research groups [1,2,3,4,5,6] is based on a

conventional internal combustion engine in which one additional valve, called the charging valve, connects the

combustion chamber to an air tank as shown Figure 1. Intake

and exhaust valves can remain camshaft-driven, and only the

charging valve is moved by a fully variable actuator. As a

result, the new major modes allowed by this additional valve

Page 2 of 12

(pneumatic pump, pneumatic motor and supercharged)

remains 4-Strokes cycles [7].

Pneumatic Pump: During a regenerative braking, this mode

transforms vehicle kinetic energy into a re-usable pneumatic

energy. The cycle begins with a conventional intake stroke.

The charging valve opens during the compression stroke as soon as the in-cylinder pressure reaches that of the charging

port. The valve closes shortly after TDC, and the cycle

finishes with conventional expansion and exhaust strokes.

Pneumatic Motor: This mode is used to recover mechanical

work from the pneumatic energy, without any combustion.

The cycle begins with classical intake and compression

strokes. The charging valve opens when approaching TDC,

and remains open during part of the expansion stroke in order

to maintain pressure at a high level and generate the desired

indicated work. The cycle finishes with a conventional exhaust

stroke.

Pneumatic Supercharged mode: This mode is used to compensate for the "turbo lag" effect, while temporarily boost

the engine during a strong acceleration transient. The charging

valve opens shortly after intake valve closure, and closes when

the desired additional air mass has flowed into the in-cylinder.

The end of the cycle remains the same as a conventional four

stroke one.

Several previous publications [3, 7, 8, 9] have shown that the

concept of pneumatic hybrid engine could provide benefits in

terms of fuel consumption and greenhouse gas emissions that

are comparable with those of an electric hybrid vehicle [10]

but with lower weight and presumably lower cost.

Thermal Insulation of the Air Tank

One major drawback encountered on the pneumatic hybrid engine concept is due to the non-adiabaticity of the Air Tank.

Indeed, while using the pneumatic pump mode during a

regenerative braking, the pressurized air sent to the tank is also

at high temperature level. Assuming an homentropic

compression from the ambient (Pext, T ext ) the air temperature

level can be estimated:

1

,homair

air entropic ext

ext

PT T

P

(1)

Ideally, if both the tank and the charging pipe were perfectly

adiabatic, the air temperature level could be maintained over

time. In practice, this is not the case. Due to its volume

(around 30liters) the air tank cannot be localized near the

engine, and a quite long charging pipe is needed. Even with a

thermal insulation of both air tank and charging pipe, the heat

transfer remains significant and the pressurized air quickly

cools to reach the ambient temperature. This phenomenon has

been experimentally observed on real pneumatic hybrid engine

by Trajkovic[11],Brejaud [12] and Donitz [3] . These authors

show that, without any additional heating system, the air tank

has to be considered isothermal.

Figure 2: Relative air mass consumed In pneumatic Motor Mode

As Function As Temperature [Wi=310 J/cycle- Ptank=12 bar-1200 RPM]

Finally, when reusing the air in pneumatic motor mode, the

mass necessary to fill the cylinder is greater than the one that

would have been necessary if the air was not cool at its stay in

the tank.

Figure 2 gives the calculated relative air mass needed in

pneumatic motor mode as a function of the incoming air

temperature, in order to produce an indicated work of 120

J/cycle, 1200 RPM, for an air tank pressure level of 12 bars.

This figure clearly shows that a great benefit exists in using

heated air instead of "cold" one coming from an isotherm air

tank.

Single Main Heated Tank

Figure 3: Single Main Air Tank Concept heated with exhaust gases. Source: [5]

293 393 493 593 693 7930

20

40

60

80

100

Relative air mass consumed in pneumatic motor mode [%]

Air Temperature [K]

Page 3 of 12

This observation has leaded several authors to propose to heat

the air tank with the exhaust gases produced while the engine

is used in conventional mode. Tai and al. [5] was the first

research team to propose the concept of a single Air Tank

heated with the exhaust gases in a technical paper [5].The

proposed concept is also described in a US Patent [13]. See Figure 3.

Figure 4: Energy Domains

This proposal for heating the air tank allows a new pathway in

the management of energy domains, as depicted in dashed line

by Figure 4. It is then possible, due to the heat exchanger

created by the double-walled air tank, to recover and to

transfer a part of the exhaust gas heat to the compressed air.

Proposed Concept

Figure 5: Schematic representation of the proposed concept

From the authors’ point of view, the idea for heating the air

tank is a smart way to improve the overall efficiency of the

hybrid engine, but its implementation by a single tank is unrealistic. As mentioned before and due to its volume, it is

not possible to embed the air tank near the exhaust pipe of the

engine. Consequently two large length air circuits are needed

for both exhaust gas and the pressurized air circuits. The heat

transfer phenomenon along these pipes finally leads to an

important reduction of the energy efficiency of the concept.

First, while the engine operates in conventional mode, a

significant part of the exhaust gas heat is dissipated along the pipe. Second, while the engine operates in pneumatic motor

mode, the hot pressurized air issued by the air tank is cooled

before entering into the cylinder.

For these reasons, authors propose an alternative charging air

path. See Figure 5. The proposed system consists of three

main components connected by various pipes: an intermediate

tank, a main tank and a optional three-way valve called "By-

pass valve". The main tank is a reservoir of compressed air

that presents a large capacity, up to 50 liters. The latter is

single-walled and has no special thermal insulation, so it has

to be considered as isothermal. The intermediate tank is

double-walled, low-capacity, and acts as a heat exchanger between the pressurized air issued from the main tank and the

exhaust gas of the engine.

The main and the intermediate tank may be respectively

considered as a mass and enthalpy reservoir. The main

function for the intermediate tank is to guarantee the

availability of a sufficient volume of “hot” and pressurized air

that will be typically consumed during a vehicle start from a

null speed while using the pneumatic motor mode. Once the

engine switches to the conventional combustion mode, the

new function for the intermediate tank is then to warm-up the

cold air that has came from the main tank in previous state. Next, when the engine is stopped, the double-wall acts as

insulation and limits the heat losses for the hot pressurized air.

The three-way “By-pass” valve is used to derive the flow of

exhaust gas when it is not desired that it passes through the

intermediate tank. Its presence is motivated by two reasons:

1. In pneumatic motor mode, because of the expansion

performed in the engine cylinder, the air discharged from

the exhaust valve reaches extremely low temperatures.

Thus, in order not to cool the intermediate tank, the

addition of a three way "By-Pass" valve is desirable. The

presence of the latter is not mandatory, but is

recommended in order to maximize the energy potential of the concept.

2. The heat transfer process while the exhaust gas crosses

the intermediate tank lowers the temperature of the latter.

As a consequence, the efficiency of the catalytic converter

could be negatively impacted. The presence of the three-

way "By-Pass" valve allows, if desired, not to cool the

exhaust and get a rapid rise in temperature of the catalytic

converter.

Moreover, it is appropriate that the pipe connecting the

charging valve to the intermediate tank is the most thermally insulated as possible. It is important to minimize the heat

Page 4 of 12

losses of the air between the outlet of the intermediate tank

and its entry into the engine cylinder.

Proposed Study

This paper is organized as follows. First, a 0D thermodynamic

model of the coupling of the intermediate and main tank is

proposed. Second, a 0D quasi-static thermal modeling of the

intermediate tank is developed. Third, the fuel savings, for the

NEDC and WLTP driving cycles under several thermal initial states of the intermediate tank, are estimated by numerical

simulations. Finally, the influence of the intermediate heated

tank size is investigated.

THERMODYNAMIC MODELLING

Pneumatic Pump Mode: Air is entering

the Intermediate Tank

Figure 6: Thermodynamic model

The first law of thermodynamics, written for the intermediate

tank, leads to:

. . . ( )

. . . .

v air v air in out

p air in p ext out

m c dT c T dm dm

Q W c T dm c T dm

(2)

Where m, cp and cv respectively represents the air mass inside

intermediate tank, the specific heat at constant pressure and

the specific heat at constant volume. Using the perfect gas

law, the elementary rise of pressure inside the intermediate tank can be expressed as:

( ) .int air in out air

int

rdp T dm dm m dT

V (3)

Where r represents the perfect gas constant (r=cp-cv). A similar

method, applied to the main tank gives:

. .ext outmain

main

r T dmdp

V (4)

Assuming there is no pressure losses ( ntmain idp dp dp )

and re-arranging equations (2),(3) and (5), the transferred

mass from the intermediate tank to the main tank can be

determined:

. .

.

p in

outint

v ext p

ma

n

in

i

air

Q c T dmdm

Vc T c T

V

(5)

Finally, the elementary rises of pressure and temperature, in

the intermediate tank, can be respectively determined using

equations (6) and (7):

extout

main

rTdp dm

V (6)

1( )int

air ext out air out in

main

VdT T dm T dm dm

m V

(7)

Pneumatic Motor Mode: Air is leaving

the Intermediate Tank

In this case, dmin and dmout both have to be considered as

negative values. Moreover, it is assumed (isotherm assumption

of the main tank) that the air mass transferred from the main

tank to the intermediate tank (dmout) is at the exterior

temperature level. Accordingly, Equation (5) becomes:

. .

.

p in

outint

v

air

ext

m

ext p

ain

Q c T dmdm

Vc T c T

V

(8)

The elementary rises of pressure and temperature remains

governed by equations (6) and (7).

QUASI-STATIC HEAT TRANSFER

MODEL

Page 5 of 12

Figure 7: Resistor Model

A thermal resistance model, as depicted in Figure 7 has been

used for estimating the heat fluxes. Fives temperature nodes

are considered: exterior (Text) air inside the intermediate tank

(Tair) exhaust gas (Tech) and the inside wall temperature (Twi) and the outside wall temperature (Two) of intermediate tank.

The nature of the convective heat exchange process (Natural

Convection, Forced Convection and Conduction) between two

nodes of the model depends on the operating mode of the

engine. Table 1 presents the different possible cases. It is

assumed that the 3 ways "By-Pass" valve derives the "cold"

exhaust gas flow while the engine operates in both pneumatic

motor and pump modes.

Table 1: Heat Exchange Type as a function of the Engine Operating Mode.

Engine Operating Mode Off Conventional Pneumatic

Motor

k12

Natural Conv. Natural Conv. Natural Conv.

k23

Natural Conv. Forced Conv. Natural Conv.

k24

Conduction Conduction Conduction

k34

Natural Conv. Forced Conv. Natural Conv.

k45

Natural Conv. Natural Conv. Forced Conv.

Natural Convection

In the case of a natural convection, the following correlation,

valid for a heat exchange between a stainless steel cylinder

and air, and was used:

0.25

1.32e

Tk

D

(9)

De and ΔT are respectively the cylinder diameter and the

temperature gradient (SI Units).

Forced Convection

In the case of forced convection, the Colburn correlation has

been used:

0.8 0.330.023( ) ( )Nu Re Pr (10)

The Prandtl number is fixed to a value adapted for an ideal

gas: Pr=0.75.

The Reynolds number is calculated assuming a flow inside a

tube, with following expression:

. .mv DRe

(11)

Where D and vm are respectively the tube diameter and the

mean gas velocity of the pulsed flow inside the tube. vm has

been estimated using following expression :

120.m

m Nv

S

(12)

Where Δm, ρ, N and S are respectively the mass transferred

per cycle, the density, the engine revolution speed and the

cross section offered to the gas flow.

In combustion mode, Δm and ρ are determined using some

two dimension data maps as a function of the engine

revolution speed and the effective torque. These 2D data maps

have been experimentally measured on a real engine under

stationary conditions on a test bench.

In pneumatic mode, Δm is determined using four dimensions

data maps as function as Pressure Tank, air temperature,

engine speed and engine effective torque. These 4D data maps

have been built using a quasi static modeling of the engine

under stationary conditions. The engine model that has been

used to build these 4D Maps is detailed in several previous

works from authors [14, 15, 12].

Conduction

The conductive heat exchange between both metallic walls of

the intermediate tank (nodes #2 and #4) is evaluated by using

Page 6 of 12

the Fourier’s Law, assuming a steady state Conduction

Phenomena.

SIMULATION BACKGROUND:

Intermediate Heated Tank

Figure 8: Design of the Intermediate Heated Tank.

In a first approach, in order to work with numerical data as

representative as possible, a simple CAD design of the

intermediate tank has been conducted. Figure 8 gives a

perspective view of the proposed design. The intermediate

tank is entirely made from stainless steel. It is primary

constituted with 49 tubes (Diameter: 10 mm, thickness: 0.2

mm) as thin as possible in order to reduce thermal inertia but

sized to withstand a 20 bar pressure at 750 K. The outer wall

is a standard DN 250 tube, 3 mm thickness. The intermediate

tank is designed as a cross-flow heat exchanger: the exhaust

gases, entering and leaving via the below ports, are routed through baffles in order to maximize the heat transfer

phenomena. One notes that a diffuser / deflector, not shown in

Figure 8, is intended to better distribute the incoming air flow

into the 49 tubes.

Table 2: Numerical data for the 3 intermediate heated tanks considered.

Size #1 Size #2 Size #3

Vint

[liter] 2.5 5 7.5

Vech

[liter] 5.3 10.70 16

mext

[kg] 5.953 8.749 11.543

mint

[kg] 0.973 1.946 2.919

Sext

[m²] 0.1275 0.2551 0.3825

Sint

[m²] 0.5329 1.0658 1.5987

Sair

[m²] 0.0038

Sech

[m²] 0.0088

By varying the primary tubes length, three different volumes

for the hot pressurized air contained in the intermediate tank,

have been investigated (2.5, 5 and 7.5 liter). In each case, the

main tank volume was 30 liters. Table 2 gives the

characteristics for the three considered tanks. Vext and Vint are

respectively the total volume for the exhaust gases and the air contained in the tank. m ext and mint are respectively the total

mass for the outer and the inner parts of the tank. S ext and S int

are respectively the total heat exchange surface for the outer

and the inner walls. S air and S exh are respectively the cross

section offered to air and exhaust gas flows used to determine

the mean velocity, see equation (12).

Engine and Vehicle

Table 3: Combustion-Pneumatic Hybrid Engine

Cylinder 3

Fuel Type Gasoline

Displacement Around 1.2 liter

Compression Ration Around 10

Max. Power 70 kW around 6.000 RPM

Table 3 and Table 4 respectively give the characteristic of the

considered engine and vehicle used for the road cycle

simulations. For confidential reasons, more details cannot be

given.

Table 4: Vehicle characteristics

Mass Around 1050 [Kg]

Drag Around 1.5 [m²]

Cx Around 0.4

Rolling Resistance Around 0.07 [N/Kg]

Transmission 5 manual gearbox

Page 7 of 12

Energy Management Strategy

Figure 9: Causal strategy

The energy management retained for present work is a causal

strategy, which is based on the engineer’s experience

(heuristics) and consequent intuition about how to operate the

energy source in the most efficient regions. Figure 9 depicts

the organization chart for engine mode choice. This energy

management strategy has been selected because it leads to an

effective fuel savings close to the optimum obtained by a

dynamic programming [16, 17], but with a very significant

reduction of computational time. Moreover, the aim for authors is not to determine this optimum fuel savings but to

establish if a benefit exists in the use of an intermediate heated

tank compared to a single isotherm main tank.

Efficiency Indicators

In order to precisely quantify, during a entire road cycle

simulation, the efficiencies over the entire energetic chain,

several indicators have been defined. See Figure 10.

Figure 10: Efficiency Chain

For a given engine cycle performed (index j), ,j regenW is

defined as the negative work per cycle recoverable at

crankshaft in order to produce the needed deceleration of the

vehicle assuming there is no use of the braking system.

,j pumpW represents the effective work per cycle recovered

when the pneumatic pump mode is used. Because the

regenerative pneumatic pump mode is often unable to

consume all the recoverable energy that is available, a first

efficiency can be defined as the Pump Usage Efficiency. This

indicator has been evaluated over the entire road cycle as:

,

,

Pump_Usage_Efficiency

j pump

j regen

j regen

j regen

W

W

Next, assuming a homentropic compression of the pumped air

mass (Δmin), a Coefficient of Performance for the pneumatic

pump mode can be defined as:

,hom ,hom

_

, ,

. .j entropic p air entropic

j regen j regen

pneu pump

j regen j regen

in

j pump j pump

H m c T

COPW W

Because, the main tank has to be considered in practice as

isothermal, a significant part of the Enthalpy transferred by the

Pump mode is lost by heat transfer process. Δmin and Δmout

are respectively defined as the mass of pressurized air that

inflows and outflows from the intermediate tank. The

Coefficient of Performance of the isotherm tank is then

defined as:

,

_ tan

,hom ,hom

. .

.

j isotherm out p ext

j regen j regen

isotherm k

j entropic in p air entropic

j regen j regen

H m c T

COPH m c T

When this pressurized cold air is re-used while the engine

operates in pneumatic motor mode, the latter is warmed while

its stay in the intermediate heated tank. ,air outT , Δmin and

Δmout are respectively defined as temperature of the warmed

air out-flowing from the exchanger, the mass of warmed air

out-flowing from the exchanger and the mass of cold air that

out-flowing the main tank . The coefficient of Performance of

the exchanger can then be defined as:

, ,

,

. .

. .

j exchanger in p air out

j motor j motor

exchanger

j isotherm out p ext

j motor j motor

H m c T

COPH m c T

Next, the coefficient of performance for pneumatic motor is

defined as:

, ,

_

,,. .

j motor j motor

j motor j motor

pneu motor

p air out

j motorj motor

inj exchanger

W W

COPC TmH

It then possible to define an Overall Regenerative Efficiency

over the entire road cycle as:

Page 8 of 12

,

,

Overall_Regenerative_Efficiency

j motor

j motor

j regen

j regen

W

W

SIMULATION RESULTS:

Cycles and Initial Thermal State Study

Two road cycles were considered: the New European Driving

Cycle (NEDC) cycle and a probable but still unconfirmed

Worldwide harmonized Light vehicles Test Procedures

(WLTP) cycle. For both cycles, four different simulations

were performed:

1. A reference simulation in pure combustion mode, with no hybridization and no Start & Stop effect.

2. A single main isotherm tank simulation for a 20 °C

exterior temperature level.

3. A 5 liters intermediate heated tank with an initial “cold”

thermal state, defined as the exterior temperature level.

4. A 5 liters intermediate heated tank simulation with an

initial “hot” thermal state, defined as the intermediate

heated tank final thermal state obtained for simulation #3.

Tables 5 and 6 respectively give the numerical results obtained

for the NEDC and WLTP cycles.

Figure 11, Figure 12, Figure 13, Figure 15 respectively display the simulation results as function as time for the

Combustion Mode, Single Main Isothermal Tank, Cold Start

and Warm Start with an Heated Intermediate Tank.

Figure 14 and Figure 16 respectively give the temperature as

function as time of the 5 nodes of thermal resistance model,

for the Cold Start and the Warm Start with an Heated

Intermediate Tank.

Table 5: NEDC Results

Conventional

Combustion

Mode

Single

Main

Isothermal

Tank

Intermediate

Heated Tank

Cold

Start

Hot

Start

Fuel Mass

Consumed[Kg] 0,38573 0,35163 0,35136 0,34937

Consumption

(liter/100km) 4,67 4,26 4,25 4,23

Overall regenerative efficiency (%) x 6,41 6,93 8,57

Pneumatic Pump

Usage Efficiency

(%) x 29,16 27,12 28,99

COP Pneumatic

Pump(%) x 134,31 126,48 135,21

COP Isotherm

Tank (%) x 54,40 52,68 54,55

COP Exchanger

(%) x x 116,64 142,23

COP Pneumatic

Motor(%) x 17,55 20,17 17,92

Pair init (bar) x 11,10 11,70 11,10

Pair final (bar) x 11,07 11,72 11,14

Tair

init (K) x 293 293 470

Tair

final (K) x 293 470 577

Two

Init (K) x 293 293 350

Two

final (K) x 293 350 440

Relative Fuel

Saving Due to

Hybridization (%) x 21,04% 21,66% 25,94%

Relative Fuel

Saving Due to

Start Stop (%) x 78,96% 78,34% 74,06%

Page 9 of 12

Table 6: WLTP Results

Conventional

Combustion

Mode

Single Main

Isothermal

Tank

Intermediate

Heated Tank

Cold

Start

Hot

Start

Fuel Mass

Consumed[Kg] 0,79515 0,76289 0,76048 0,75393

Consumption

(liter/100km) 4,67 4,48 4,46 4,42

Overall regenerative efficiency (%) X 4,88 5,80 8,10

Pneumatic Pump

Usage Efficiency

(%) X 33,03 32,37 31,67

COP Pneumatic

Pump(%) X 153,77 148,07 153,07

COP Isotherm

Tank (%) X 53,84 53,12 53,74

COP Exchanger

(%) X x 120,20 167,76

COP Pneumatic

Motor(%) X 15,08 16,71 18,70

Pair init (bar) X 10,00 10,50 10,80

Pair final (bar) X 9,89 10,68 10,85

Tair

init (K) X 293 293 609

Tair

final (K) X 293 609 695

Two

Init (K) X 293 293 481

Two

final (K) X 293 481 594

Relative Fuel

Saving Due to

Hybridization (%) X 31,85% 36,60% 46,66%

Relative Fuel

Saving Due to

Start Stop (%) X 68,15% 63,40% 53,33%

Figure 11: WLTP Conventional Combustion Mode

Figure 12: WLTP Single Main Isothermal Tank

Figure 13:WLTP Cold Start with Heated Intermediate Tank

0 500 1000 1500 2000-20

0

20

40

60

80

100

120

140

Time[s]

CycleSpeed [km/h]

EngineSpeed [rpm]/100

WorkDemand [J]/100

Tank Pressure [bar]

Engine mode [10=Motor 20=Pump 30=Conv]

Cumulative comsumption [liter]*100

GivenWork [J]/100

0 500 1000 1500 2000-20

0

20

40

60

80

100

120

140

Time[s]

CycleSpeed [km/h]

EngineSpeed [rpm]/100

WorkDemand [J]/100

Tank Pressure [bar]

Engine mode [10=Motor 20=Pump 30=Conv]

Cumulative comsumption [liter]*100

GivenWork [J]/100

0 500 1000 1500 2000-20

0

20

40

60

80

100

120

140

Time[s]

CycleSpeed [km/h]

EngineSpeed [rpm]/100

WorkDemand [J]/100

Tank Pressure [bar]

Engine mode [10=Motor 20=Pump 30=Conv]

Cumulative comsumption [liter]*100

GivenWork [J]/100

Page 10 of 12

Figure 14: Temperature Evolution-WLTP- Cold Start with Intermediate Heated Tank. [K]

Figure 15: WARM Start with Heated Intermediate Tank

Figure 16: Temperature Evolution-WLTP-WARM Start With Heated Intermediate Tank.

Discussion

NEDC Cycle

The Single Main Isothermal Tank simulation leads to a value

of -8.78% for the relative fuel consumption in reference to the

combustion. Table 5 shows that this fuel saving is due, at first

order (78.96%) to the suppression of the engine idling (Start &

Stop Effect) and at second order (21.04%) in the use of the

pneumatic motor mode to drive the vehicle instead of the conventional mode (Hybridization).

The use of an intermediate heated tank slightly improves the

relative fuel consumption in reference to combustion mode:

respectively -8.91 % and -9.43% for the “cold” and “hot”

initial thermal state. This rather low benefit is due to the low

level of engine load required by this cycle, which does not

allow an efficient heating of the intermediate tank.

However, a significant improvement of the overall

regenerative efficiency is observed: a value of 8.57% is

obtained for the intermediate tank (warm start) in comparison

to a value of 6.41% obtained for isothermal tank. An analysis

of Table 5 clearly establishes that this overall efficiency improvement is due to the COP exchanger that reaches a value

up to 135.21%.

WLTP Cycle

First, one notes that the WLTP cycle is different than the

NEDC because of its higher energy demand and less idling

phases. In one hand, this cycle is unfavorable to every hybridization technology and, in other hand, is favorable to the

intermediate heated tank concept.

Consequently, the single main isothermal tank simulation

leads to a poor value of -4.06% for the relative fuel

consumption in reference to the pure combustion mode. The

heated intermediate tank concept slightly ameliorates the

situation and leads to relative fuel consumption values of -

4.36% and -5.18% for the “cold” and “hot” initial state. This

improvement is one more time due to COP exchanger which

reaches a value up to 167.76% for the “hot” initial state

simulation. This undeniable heating of the air finally extends

the frequency of use of pneumatic motor mode, as it is shown by Relative Fuel Saving Due to Hybridization that

respectively evolutes, for the Single Isothermal Main Tank

and for the “hot” Initial State of Intermediate Tank, from

31.85% to 46.66%. One notes that, with the use of an

intermediate heated tank, a balance is almost reached between

the contribution of the Hybridization and that of the Start &

Stop.

0 500 1000 1500 2000250

300

350

400

450

500

550

600

650

Time[s]

Two [K]

Twi [K]

Tair [K]

T ech [K]

0 500 1000 1500 2000-20

0

20

40

60

80

100

120

140

Time[s]

CycleSpeed [km/h]

EngineSpeed [rpm]/100

WorkDemand [J]/100

Tank Pressure [bar]

Engine mode [10=Motor 20=Pump 30=Conv]

Cumulative comsumption [liter]*100

GivenWork [J]/100

0 500 1000 1500 2000450

500

550

600

650

700

750

Time[s]

Two [K]

Twi [K]

Tair [K]

T ech [K]

Page 11 of 12

Heated Intermediate Tank Size Study

In this part, the effects of the internal capacity of intermediate

heated tank are investigated. Three different sizes have been

considered, as given in Table 2. The considered road cycle is

the WLTP one, for a “hot” initial thermal state of the

intermediate tank. Table 7 gives the obtained numerical

results, and Figure 17 displays the evaluation of the air inside

the intermediate tank (Tair) as function as time, for the three considered capacities.

Table 7: Numerical Results for the Intermediate Tank Size Study.

2.5 liter 5 liter 7.5 liter

Fuel Mass Consumed[Kg] 0,75439 0,75393 0,75339

Consumption (liter/100km) 4,43 4,42 4,42

Overall regenerative efficiency (%) 7.91 8,10 8.4

Pneumatic Pump Usage Efficiency

(%) 32,17 31,67

31,72

COP Pneumatic Pump(%) 152,36 153,07 151,93

COP Isotherm Tank (%) 53,67 53,74 53,60

COP Exchanger (%) 149,80 167,76 178,58

COP Pneumatic Motor(%) 18,61 18,70 19,10

Pair init (bar) 10,80 10,80 10,80

Pair final (bar) 10,83 10,85 10.92

Tair

init (K) 609 609 609

Tair

final (K) 744 695 661

Two

Init (K) 481 481 481

Two

final (K) 628 594 571

Relative Fuel Saving Due to

Hybridization (%) 46,07 46,66

47,36

Relative Fuel Saving Due to Start

Stop (%) 53,93 53,33

52,64

Figure 17: Air Temperature as Function as Time for the 3 different size of the intermediate tank

Discussion

An analysis of Table 7 shows that the Intermediate Tank

volume has a slight impact on the fuel saving and the overall

Regenerative Efficiency. Compared to the reference size of 5

liters, a reduction or increase of 2.5 liters, respectively leads to

relative change for the overall regenerative efficiency of -

2.35% and +3.70%. The relative variation values for the COP

exchanger are respectively of -10.71 % and +6.45% for a

reduction and an increase of 2.5 liters of the reference size.

An observation of Figure 1 highlights two opposite effects . First, the more the tank is large, the less is the air temperature

drop during a pneumatic start of the vehicle. Second the more

the tank is large, and less is the final temperature level reached

by the air at the cycle end. First effect is beneficial, but the

second one is unfavorable.

Previous remarks show that the reference size of 5 liters,

chosen to be a typical consumed air volume during a

pneumatic start of the engine, was an appropriate compromise.

Of course, a slight benefit exists in increasing the intermediate

tank size, but as mentioned before, the tank has to keep a

reasonable size in order to be located nearest the exhaust ports

of the cylinder head.

0 500 1000 1500 2000400

600

800

0 500 1000 1500 2000400

600

800

0 500 1000 1500 2000400

600

800

Time[s]

2.5 liters

5 liters

7.5 liters

Page 12 of 12

SUMMARY/CONCLUSIONS

In this paper, an alternative to the single main heated tank

concept, that authors judge unrealistic, has been proposed.

Both thermodynamic and steady state heat exchange models

have been proposed. The use of the numerical model, for road

cycle simulations, has shown that the proposed concept leads

to a slight but undeniable increase of the overall regenerative

efficiency. The reason for this improvement is the COP exchanger that reaches values up to +167% for a “hot” initial

state of the intermediate tank. It has been shown that the

Internal Tank volume has an impact on the fuel saving and

other efficiency indicators, but that a volume around 5 liters is

sufficient and probably makes possible its implantation nearest

the exhaust ports of the engine.

Future works have to experimentally validate the theoretical

benefits highlighted. A validation of the heat exchange

correlations that have been used is needed. An intermediate

heated tank prototype has to be built and implemented on a

real engine on a test bench in order to validate the thermal

model for engine operating mode switches.

REFERENCES

[1] M. SCHECHTER. New cycles for automobile

engines. SAE Technical Paper, 1999-01-0623, 1999.

[2] P. HIGELIN and A. CHARLET. Thermodynamic

cycles for a new hybrid pneumatic-combustion engine

concept. SAE Paper, 2001-09-23, 2001.

[3] C. DONITZ, I. VASILE, C. ONDER, and

L. GUZZELLA. Realizing a concept for high efficiency and

excellent drivability: The downsized and supercharged hybrid

pneumatic engine. SAE paper, 2009-01-1326, 2009.

[4] M. ANDERSSON, B. JOHANSSON, and

A. HULTQVIST. An air hybrid for high power absorption and

discharge. SAE Paper, 2005-01-2137, 2005. [5] C. TAI, T.TSAO, M.B. LEVIN, and

M. SCHECHTER. Using camless valvetrain for air hybrid

optimization. SAE Paper, 2003-01-0038, 2003.

[6] S. TRAJKOVIC, P. TUNESTAL, and

B. JOHANSSON. Investigation of different valve geometries

and valve timing strategies and their effect on regenerative

efficiency for a pneumatic hybrid with variable valve

actuation. SAE Paper, 2008-01-1715, 2008.

[7] C. DONITZ, I. VASILE, C. ONDER, and

L. GUZZELLA. Modeling and optimizing two- and four-

stroke hybrid pneumatic engines. IMechE Part D - Journal of

Automobile Engineering, Vol. 223, No. 2, pp. 255-280, 2009. [8] A. IVANCO. Evaluation of energy management

strategies for a hybrid pneumatic engine. PhD thesis,

Université d’Orléans, 2009.

[9] S. TRAJKOVIC, P. TUNESTAL, and

B. JOHANSON. Vehicle driving cycle simulation of a

pneumatic hybrid bus based on experimental engine

measurements. SAE Technical paper, 2010-01-0825(2010-01-

0825), 2010.

[10] A. ELGOWAINY, A. BURNHAM, M. WANG,

J. MOLBURG, and A. ROUSSEAU. Well-to-wheels use and

greenhouse gas emissions of plug-in-hybrid electric vehicles.

SAE paper, 2009-01-1309, 2009. [11] S. TRAJKOVIC. The Pneumatic Hybrid Vehicle. A

new Concept For Fuel Consumption Reduction. PhD thesis,

Lund University, 2010.

[12] P. BREJAUD. Etude théorique et expérimentale d’un

nouveau concept de moteur hybride thermique pneumatique.

PhD thesis, Université d’Orléans., 2011.

[13] SCHECHTER. Operating an air-hybrid vehicle with

two-stages compression and expansion. us patent us 7,543,668

bi, 2009.

[14] P. BREJAUD, A. CHARLET .,

Y. CHAMAILLARD, A. IVANCO, and P. HIGELIN. Pneumatic-combustion hybrid engine: A study of the effect of

the valve train sophistication on pneumatic modes. Oil and

Gas Science and Technology-Rev. IFP, DOI:

10.2516/ogst/2009054, 2009.

[15] P. BREJAUD, P. HIGELIN, A. CHARLET, and

Y. CHAMAILLARD. One dimensional modeling and

experimental validation of single cylinder pneumatic

combustion hybrid engine. SAE paper, 2011-24-0074, 2011.

[16] A. IVANCO, A. CHARLET, Y. CHAMAILLARD,

and P. HIGELIN. Energy management strategies for hybrid-

pneumatic engine studied on an Markov chain type generated

driving cycle. SAE Paper, 09PFL-1130, 2009. [17] C. DONITZ, I. VASILE, C. ONDER, and

L. GUZZELA. Dynamic programming for hybrid pneumatic

vehicles. In Proceeding ACC’09 proceedings of the 2009

conference on American control conference, 2009.