Recent advances in SEMERGY: A semantically enriched...
Transcript of Recent advances in SEMERGY: A semantically enriched...
Recent advances in SEMERGY: A semantically enriched optimization environment for performance-guided bui lding design
and refurbishment
U. Pont, F. Shayegenfar, N. Ghiassi, M. Taheri, C. Sustr & A. Mahdavi
Department of Building Physics and Building Ecology, Vienna University of Technology
S. Fenz, J. Heurix, A. Anjomshoaa, T. Neubauer & A.M. Tjoa
Inst. of Software Technology and Interactive Systems, Vienna University of Technology
Performance-Assessment-Guided Design – Barriers and Expectations
� Early-stage performance assessment can drastically affect final performance of
the building
� Currently assessment tools are used for certification and labeling of the final
product rather than to support design decisions
� Less than 10% of the developed tools are targeted for architects and designers
Source: Attia 2011
Performance-Assessment-Guided Design – Barriers and Expectations
� Intelligence: capability to analyse results, draw meaningful conclusions, and
suggest clear guidelines to improve design
� Usability: Facility of accumulation and entry of input information and performing
computations
Source: Attia 2011 Source: Mahdavi et al. 2005
A1
Folie 3
A1 add time consumption graphAdmin; 03.06.2013
SEMERGY: Concept
� Laymen� Architects� Municipalities
� Building products
� Performance assessment
� Life Cycle Analysis
Source: w
ww
.data2save.com
� Building products� Building codes� Subsidies
� Life Cycle Analysis� Cost estimation
OPTIMIZATION
Evolution of Web to Web of Data
5
Tim Berners Lee‘s Data star schemeQuality Level Description Format / example
� Available on the web but with an open license, to be Open Data
whatever format
�� Available as machine-readable structured data
Excel instead of image scan of a table
��� as (2) plus non-proprietary format CSV or JSON instead of
6
��� as (2) plus non-proprietary format CSV or JSON instead of Excel
���� All the above plus, Use open standards from to identify things
RDF, SPARQL
����� All the above, plus: Link your data to other people’s data to provide context
Linked RDF
Database Hugging Disorder (DBHD)Hans Rosling
• My basic idea is that the world has changed so much, what people need is a new mindset.
• To heal DBHD, we need to encourage:– those who provide the datasets– those who provide new technologies to access or process
them– and those who ‘play’ with them and give data meaning
http://www.thinkwithgoogle.co.uk/intl/en_uk/quarterly/data/hans-rosling-a-data-state-of-mind.html
SEMERGY: Structure
Use
r I
nte
rfa
ce
Source: Mahdavi et al. 2012
Use
r I
nte
rfa
ce
User Interface
� Novice users: SEMERGY Graphical Web-Based User Interface
� Experienced professionals (architects): Extended SEMERGY Graphical Web-Based
User Interface, with import possibility from BIM and CAD
� Area municipalities or other local authorities: Import from GIS data
Semantic Interface
� Acquisition of data from different sources
� Restructuring of the acquired data to enable specific queris
� Extension of restructured data with additional information to allow precise
search and implementation of rules
SE
ME
RG
Y
…
Baubook
Data Source : Baubook
<owl:NamedIndividual rdf:about="#solidwood_2142705277"><rdf:type rdf:resource="#SolidWood"/>
<rdfs:label l:lang="de">Brettsperrholzplatte</rdfs:label><rdfs:comment xml:lang="de„> Großformatige Massivholzpaneele aus
Kreuzweiseverleimten Brettlagen ... </rdfs:comment><semergy:heatTransferCoefficient>0.12</semergy:heatTransferCoefficient><semergy:acidificationPotential>0.00341</semergy:acidificationPotential><semergy:acidificationPotential>0.00341</semergy:acidificationPotential><semergy:notRenewablePEI>8.04</semergy:notRenewablePEI><semergy:gloablWarmingPotential>-1.26</semergy:gloablWarmingPotential>
</owl:NamedIndividual>
12
Reasoning Interface
� Incorporation of computational engines (performance simulation, normative
assessment tools,
� Identification of design alternatives
� Optimization of the initial design towards better performance within the limits of
the user defined constraints
SEMERGY
LCA, Cost estimation,…
Optimization procedure
Alternative Identification
SEMERGY Workflow
SEMERGY Building Data Model
Requirements:
� Compliance with integrated
computational engines
� Extensibility for future
Study of some existing building
data formats:
`
Cal
cula
tion
para
met
ers
Op
era
tio
na
l da
ta
data formats:
� IFC
� gbXML
� Shared Object Model for
SEMPERG
eo
me
try
Bu
ild
ing
co
mp
on
en
ts’
sem
an
tic
da
ta
Derivation of Detailed Simulation Model Based on Si mplified User Input
� Identification of spaces, space
enclosures and space boundary
conditions, detection of errors
� Generation of 3d vertex information
based on 2d user input and room
height. Generation of roof geometry
based on type,knee wall height, ridge
height and or angle and position of height and or angle and position of
gables
� Populating space properties and
material properties with default or
actual data based on user preferences
Occupied roof space
Type: Gabled roof
Knee wall height:…
Ridge angle:…
Ridge height:...
Identification of alternatives and optimization
� Study of common building constructions;
Creation of basic templates for various
constructions
� Determination of rules for products allowed in
each layer based on properties of products
(format, function, position)
� Extension of the ontology information with
Template: Massive wall with
outside insulation
Function: Insulation
� Extension of the ontology information with
additional properties:
� Imposition of the rule-based logic on the
ontology for identification of possible
candidates for each construction layer
� Creation of alternative layer combinations for
each building component; Filtering of the
results based on U-value regulations
Format: stable
Position: Outside wall, Outer
layer
Identification of alternatives and optimization
� Optimization of constructions in view of
investment costs, environmental impact and
energy use. Each individual consists of
constructions for each building component.
*Genetic algorithms are used to reduce
computations
* Initial optimization is done using normative
computational engines computational engines
� The Pareto optimal front is subjected to
simulation for the identification of the final
solution set
� Final solution set is suggested to the user
SEMERGY Graphical User Interface
� General Information: location, date of construction, building usage, number of
stories, roof form, heating system.
� Geometry and operational information: floor plans, doors, and windows
dimensions, wall types, space height, orientation, space functions.
� Semantic information and material properties: selection of construction systems,
determination of frameworks of optimization (user-defined constraints).
Ongoing Activities and Future Challenges
� BIM import possibilities for expert users
� Cost estimation
� HVAC
� GIS data import and generation of an SBM model based on this data
� Development of a data release format for product manufacturers to enable � Development of a data release format for product manufacturers to enable
seamless retrieval of product information
� Integration of active solar technologies
Institute of Building Physics and Building Ecology:
Univ. Prof. DI Dr. tech. Ardeshir Mahdavi
DI Dr. Tech. Ferial Shayeganfar
DI Ulrich J. Pont
M.Sc. DI Neda Ghiassi
DI Mahnameh Taheri
B. Sc. Christian Sustr
B.Sc. Arch. Kristopher F. Hammerberg
SEMERGY Team
Institute of Software Technology and Interactive Systems:
O.Univ.Prof. DI Dr.tech. A Min Tjoa
Mag. DI Dr. rer. Amin Anjomshoaa
DI Dr. tech. Stefan Fenz
Mag. Johannes Heurix
DI Dr. tech. Thomas Neubauer
B.Sc. Arch. Kristopher F. Hammerberg
B.Sc. Dawid Wolosiuk
B. Sc. Vaibhav Jain