OPTEEMAL data requirements experiences

@[email protected] project has received funding from the European Union’s Horizon 2020

research and innovation programme under Grant Agreement No 680676

Workshop on Interoperable data models forBuilding's Life Cycle Energy Management Processes

OptEEmAL-SWIMing Vocamp13–14 October 2016, University College London, England

OPTEEMAL Data requirement experiences

Gonçal Costa, [email protected], FUNITEC (ARC Engineering and Architecture La Salle)

Álvaro Sicilia, [email protected], FUNITEC (ARC Engineering and Architecture La Salle)

Leandro Madrazo, [email protected], FUNITEC (ARC Engineering and Architecture La Salle)

Dimitrios Rovas, [email protected], UCL (University College London) / TUC

mailto:[email protected]




OptEEmAL GA no. 680676 | District Data Model

OPTIMUS – SWIMing Vocamp | London, 13–14 October 2016

Work programme

HORIZON 2020-WORK PROGRAMME 2014-2015

5. Leadership in enabling and industrial technologies

Call identifier

H2020-EeB-2014-2015 / H2020-EeB-2015

Topic

EeB-05-2015 Innovative design tools for refurbishment at building and district level

Title of the Proposal

Optimised Energy Efficient Design Platform for Refurbishment at District Level

GA no. 680676, Contact: [email protected]

List of participants

13 Partners – 4 RTO, 2 Universities, 2 IND, 3 SME and 2 Cities

Project Overview




IFC models

CityGML

Socio-economicdata

Weather data

Energy prices Monitoring data

Users’ objectives

CSV, SQL…

- BIMs data

- GIS data

- Contextual Data

Project Overview

A web-based platform for district energy-efficient retrofitting design

District Retrofitting

design

1

2

3

OptEEmALPlatform



- BIMs data

- GIS data

- Contextual Data

Addressing the district scale in modelling and simulation

1

2

3

Multiple data models, domains, formats…

ENERGY DPI’s

COMFORT DPI’s

ENVIRONMENTAL DPI’s

ECONOMIC DPI’s

SOCIAL DPI’s

URBAN DPI’s

GLOBAL DPI’s


design

…

IFC models

CityGML

CSV, SQL…

District Data ModelInput Data

?

Multiple tools, input formats…

EnergyPlus

CitySim

NEST



- BIMs data

- GIS data

- Contextual Data


1

2

3

ENERGY DPI’s

COMFORT DPI’s


ECONOMIC DPI’s

SOCIAL DPI’s

URBAN DPI’s

GLOBAL DPI’s


design

…

IFC models

CityGML

CSV, SQL…

CSV, SQL, ….

Socio-economic data

Weather data

Energy pricesMonitoring data

IFC model

CityGML


Users’ objectives



EnergyPlus

CitySim

NEST



ENERGY DPI’s

COMFORT DPI’s


ECONOMIC DPI’s

SOCIAL DPI’s

URBAN DPI’s

GLOBAL DPI’s


design

…


Energy Data Model

Economic Data Model

n Data Model

Simulation Data models

2. DataInteroperability



Semantic Web Technologies

EnergyPlus

CitySim

NEST


- BIMs data

- GIS data

- Contextual Data

1

2

3

IFC models

CityGML

CSV, SQL…

1. DataIntegration




designDistrict Data ModelInput Data


CityGML File

CityGMLOWL

IFC OWL

IFCFile

Contextual Data

CityGML RDF

IFCRDF

Energy Data

Model

SIMULATION DATA MODELS

SEMANTIC DATA MODELS

DATA MODELS

SimModelOWL

District extension

EnergyPlus

CitySim

NEST

SIMULATION MODELS

…

IDF

XML

Proprietary Format1. Data

Integration

2. DataInteroperability





CityGML File

CityGMLOWL

IFC OWL

IFCFile

Contextual Data

CityGML RDF

IFCRDF

Energy Data

Model

SEMANTIC DATA MODELS

DATA MODELS

1. DataIntegration

SimModelOWL

District extension

A) Finding relations (alignments) between CityGML OWL, ifcOWL and SimModelOWLOntology matching: LogMap, AML…

B) Transforming RDF data according to the ontologies and their alignments. RDF-To-RDF via SPARQL constructs: Alignment API, R2R, …







Energy Data

Model


SimModelOWL

District extension

EnergyPlus

CitySim

NEST

SIMULATION MODELS

…

IDF

XML

Proprietary Format

Ad hoc connectors between Energy Data Models and particular simulation models.

All particular data needed by Simulation models have been integrated in the Simulation Data Models. 2. Data

Interoperability




Ontologies

ifcOWL (Pauwels & Terkaj) - http://www.buildingsmart-tech.org/future/linked-data/ifcowl

- Is an ontology for IFC supported by BuildingSMART.

- Exploit the benefits of semantic web technologies in terms of data distribution,extensibility of the data model, querying, and reasoning,

CityGML Owl (Knowledge Engineering @ ISS UoG) - http://cui.unige.ch/isi/icle-wiki/ontologies

- A direct translation of the CityGML XMLSchema to OWL, manually tuned

SimModel OWL (Pauwels, Corry & O’Donnell, 2014) -http://www.lbl.gov/namespaces/Sim/

- It is a data model with a domain that covers the domain of energy simulation of theentire building.

- This is implemented as a data model (.XSD) that is interoperable through XML.

- Is “geometrically compatible” with IFC among other formats.


http://www.buildingsmart-tech.org/future/linked-data/ifcowl

http://cui.unige.ch/isi/icle-wiki/ontologies

http://www.lbl.gov/namespaces/Sim/



Data requirements

Data requirements for the OptEEmAL processes

• We have used ReqCap tool and Word/Excel documents to collect datarequirements

• Each WP (OptEEmAL processes) have collected their ownrequirements (e.g., data integration, scenario generation, simulation,model visualisation, data exportation…)

• Detailed simulations require particular data requirements: Secondlevel space boundary topology.

• We have started from the end of the process DPIS and we arecoming back defining the requirements.



Data requirements

BIM level



Data requirements

Districtlevel



Data requirements

Excel ReqCap tool



Data requirements

Excel ReqCap tool

Some OptEEmAL processes have templates (data integration, scenario generation, simulation…). • There are overlaps between templates• Each one with its own data structure



Data requirements

Excel ReqCap tool

Global requirements• They are the unification of the

different templates• They have an unique structure for all

the stages of the projectX

Some OptEEmAL processes have templates (data integration, scenario generation, simulation…). • There are overlaps between templates• Each one with its own data structure



Data requirements

Global requirements

OptEEmAL processes (Stages)

Current Data Models

Each of them (global requirements, data models, processes) have its own data structure. ReqCap tool helps to identify which requirement are needed for each stage and the mappings between the requirements and the data models. However, it is needed more details for the implementation of the processes:- Identification of all data items needed: Geometry, properties of materials…- Relations between the data models: IfcBuilding SimBuilding…



Data requirements: Mapping data models

IFC RDF data

inst:IfcBuildingStorey_941 rdf:type ifc:IfcBuildingStorey .

inst:IfcBuildingStorey_941 ifc:elevation_IfcBuildingStorey inst:IfcLengthMeasure_919 .

inst:IfcLengthMeasure_919 rdf:type ifc:IfcLengthMeasure ;

inst:IfcLengthMeasure_919 express:hasDouble "2.69999999999993".

SIMMODEL IFC RDF data

inst:IfcBuildingStorey_941 rdf:type sim:SimBuildingStory_BuildingStory_Default.

inst:ifcBuildingStorey_941 sim:BuildingStoryHeight "2699.99999999993" .

Structural and conceptual mismatches between models:- Different Units- Different structures

ifc:IfcBuildingStorey ifc:IfcLengthMeasure literal

sim:SimBuildingStory_BuildingStory_Default literal




IFC RDF data

SIMMODEL IFC RDF data

ifc:IfcBuildingStorey ifc:IfcLengthMeasure literal

sim:SimBuildingStory_BuildingStory_Default literal

Translation into R2R mapping format:

mp:BuildingStoryHeight rdf:type r2r:PropertyMapping ;

r2r:classMappingRef mp:Storey ;

r2r:prefixDefinitions "ifc: <http://www.buildingsmart-tech.org/ifcOWL/IFC2X3_TC1#> …. " ;

r2r:sourcePattern "?SUBJ ifc:elevation_IfcBuildingStorey [ express:hasDouble ?num]." ;

r2r:targetPattern "?SUBJ simbldg:BuildingStoryHeight ?num_mil" ;

r2r:transformation "?num_mil = ?num * 1000" .




A methodology is needed to address the problem of overcome the structural

differences between data models. It is important to maintain the data transformation

process in the future.

Steps:

1. Find direct mappings between the concepts of the two ontologies (ifcBuilding

simBuilding) Using semi-automatic processes (Ontology alignment tools)

2. Find the structural differences between the two ontologies and extract the

patterns. A manual process.

3. Find the particular aspects that need to be dealt in each pattern.



Data requirements

Domain: Specific requirements:

District Site Information about the site, including location, placement of buildings in OptEEmAL’s definition of districtDistrict Performance Information about the performance of the district, District Performance Indicators

Product Building Geometric properties, location, building envelopeBuilding ElementsWalls, Windows, SlabsMaterials Layering, Material Properties

Actor User Information about IPD users involved in the project

Energy SecondLevelSpaceBoundary Second-level space boundaries inside the building,building outside shading surfaces, inter-building second-level boundaries, shading groupsDistrict-level interactions Generation and distribution of energy along the district

Original source: D1.2 “Business Use Cases for the use of BIM-LOD in BLCEM – Phase II”, SWIMING project



Data requirements

Cross Domain: Specific requirements:

Identification Properties for uniquely identifying objects (i.e name and category)

Location Basic description of the placement of spatially-located things/building services/devices, etc.

Space Data regarding volumes of spaces, numbering (in case of rooms and whether a space is interior or exterior, whether the space contains other spaces, or is contained by a space).

Representation A method for visually representing an object, e.g. a geometric representation.

Project Information relevant to the entire project, e.g. units of measurement

OwnerHistory Provenance dataOriginal source: D1.2 “Business Use Cases for the use of BIM-LOD in BLCEM – Phase II”, SWIMING project



Conclusions

• Data requirement capture is an iterative process.- When the data integration process is carried out then you may need to modify

initial requirements.

• The requirements of the simulations tools are not easy to model in ReqCap tool (BIM*Q tool):

- Not all data requirements (particularly the detailed ones) can be represented.- The geometry of the elements of the building cannot be fully represented.

• Mappings between data models (IFC, SIMMODEL…) are needed- How a data model is transformed/translated into another one is a challenge.



Conclusions

Semantic Web Journal. Impact Factor 1.7 Q1Special issue: AEC/FM + Semantic WebDeadline: March 2017

THANK YOU FOR YOUR ATTENTION!

@[email protected] project has received funding from the European Union’s Horizon 2020

research and innovation programme under Grant Agreement No 680676

Gonçal Costa, [email protected], FUNITEC (ARC Engineering and Architecture La Salle)

Álvaro Sicilia, [email protected], FUNITEC (ARC Engineering and Architecture La Salle)

Leandro Madrazo, [email protected], FUNITEC (ARC Engineering and Architecture La Salle)

Dimitrios Rovas, [email protected], UCL (University College London) / TUC





OPTEEMAL data requirements experiences

Data & Analytics