S.1.c Building Energy Performance Estimation

Large-scale Residential Energy Maps:
Estimation, Validation and Visualization
Project SUNSHINE: Smart UrbaN ServIces
for Higher eNergy Efficiency
Umberto di Staso1
, Luca Giovannini2
, Marco Berti1
, Federico Prandi1
,
Piergiorgio Cipriano2
, and Raffaele De Amicis1
1
Fondazione Graphitech, Via alla Cascata 56/C, Trento (TN), Italy
{umberto.di.staso, marco.berti, federico.prandi, raf-
faele.de.amicis}@graphitech.it
2
Sinergis Srl, Via del Lavoro 71, Casalecchio di Reno (BO), Italy
{luca.giovannini, piergiorgio.cipriano}@sinergis.it
Abstract. This paper illustrates the preliminary results of a research project
focused on the development of a Web 2.0 system designed to compute and vis-
ualize large-scale building energy performance maps, using: emerging plat-
form-independent technologies such as WebGL for data presentation, an ex-
tended version of the EU-Founded project TABULA/EPISCOPE for automatic
calculation of building energy parameters and CityGML OGC standard as data
container. The proposed architecture will allow citizens, public administrations
and government agencies to perform city-wide analyses on the energy perfor-
mance of building stocks.
Keywords: CityGML, WebGL, 3D city model, smart city, interoperable ser-
vices, energy efficiency, city data management, open data, building energy effi-
ciency, energy performance index.
1 Introduction
One of the current hottest topics in the information technology research area is cer-
tainly the one about “smart-cities”. But, what is a smart-city? Many definitions exist
in the current literature [3][6,7,8][18] and all of them have a factor in common: the
existence of an underlying ICT infrastructure that connects the physical infrastructure
of the city with web 2.0 capabilities and enables innovative solutions for city man-
agement, in order to improve sustainability and the quality of life for citizens.
The energy consumption efficiency of residential houses is an important factor
having an impact on the overall city ecosystem and quality of living and it would
greatly benefit from an ICT-enabled smart approach. In fact, increasing building en-
ergy efficiency would not only mean a cut-down in energy expense for citizens, but
would also have an impact on the overall production of CO2 at energy plants and
also, even if less intuitively, on the city air pollution. Among of the major causes of

poor air quality are actually industries and domestic heating systems, not the quantity
of vehicles circulating in the urban area, as one might more typically think [6].
So, what kind of smart service can be designed in order to support the increase of
building energy efficiency and improve the city quality of life in this respect? What
we present in this paper is a specific answer to this question. The paper will illustrate
the concept and the development of smart services, which allow the assessment of the
energy performance of all the residential buildings in a city, its validation and visuali-
zation in a format accessible to citizens and urban planning experts alike.
The development of these services is part of the scope of the SUNSHINE project
(Smart UrbaN ServIces for Higher eNergy Efficiency, www.sunshineproject.eu), that
aims at delivering innovative digital services, interoperable with existing geographic
web-service infrastructures, supporting improved energy efficiency at the urban and
building level. SUNSHINE smart services will be delivered from both a web-based
client and dedicated apps for smartphones and tablets. The project is structured into
three main scenarios:
 Building Energy Performance Assessment: Automatic large-scale assessment of
building energy behavior based on data available from public repositories (e.g. ca-
daster, planning data etc.). The information on energy performances is used to cre-
ate urban-scale maps to be used for planning activities and large-scale energy pre-
certification purposes.
 Building Energy Consumption Optimization: Having access via interoperable
web-services to real-time consumption data measured by smart-meters and to lo-
calized weather forecasts, it is possible to optimize the energy consumption of
heating systems via automatic alerts that will be sent by the SUNSHINE app to the
final users.
 Public Lighting Energy Management: Interoperable control of public illumina-
tion systems based on remote access to lighting network facilities via interoperable
standards enables an optimized management of energy consumption from a web-
based client as well as via the SUNSHINE app.
This paper focuses on the preliminary results for the first of the three scenarios.
The aim of the service for Building Energy Performance Assessment is to deliver an
automatic large-scale assessment of building energy behavior and to visualize the
assessed information in a clear and intuitive way, via the use of what we call energy
maps.
Energy maps will be made publicly available via a 3D virtual globe interface based
on WebGL [14] that leverages on interoperable OGC standards, allowing citizens,
public administrations and government agencies to evaluate and perform analysis on
the building energy performance data. The presentation of these energy performance
data in a spatial-geographic framework provides a global perspective on the overall
performance conditions of the residential building stock as well as on its fluctuations
on the neighborhood and block scale. It is thus the key for an efficient maintenance
planning to increase the overall energy efficiency, allow citizen to save more money
and, ultimately, improve the quality of living of the city.

2 Energy Maps: State of the Art
The current availability of relevant technologies and standards has encouraged the
development of many research projects in the area of building energy performance
estimation based on publicly available data with the aim of creating energy map.
The main challenge in this task is related to effectively providing data for the
whole city area. For example, building certificates, adopted by many of EU countries
to describe building efficiency, can provide a very detailed insight on building energy
properties, but on the other hand, these certifications are not mandatory for all the
residential buildings and their availability is thus very sparse.
So, given the fact that publicly available data generally do not include all the in-
formation needed for the energy performance calculation, one of the most common
approaches to energy map creation is to estimate the missing information in a reliable
way, using the basic input data that is typically available, such as building geometry,
building use, construction year, climatic zone, etc. A solid example of this approach is
described in [15], where the City Geography Markup Language (CityGML) standard
[9] is used to semantically describe the set of objects that compose the urban envi-
ronment, a building typology database is exploited to statistically estimate the energy
performance properties of buildings and, finally, an Application Domain Extensions
(ADE) to the CityGML model is defined to store the estimated information for each
building [4][11].
A radically different approach is described in [10], where thermal images acquired
by airborne thermal cameras are used to measure the heating lost by buildings via
their roofs and windows and from that the energy performance of the buildings is
estimated.
Both approaches have merits and deficiencies. In the former case, input data are
publicly available, requiring no additional cost; however, having to rely on typologi-
cal databases to estimate the most of the energy parameters yields a result that is typi-
cally not very statistically reliable at the building scale and is usually confined to
residential buildings (where performance typologies are easier to define). Moreover,
the overall software architecture is typically desktop based, so the access to the results
is often limited to a small number of users with advanced GIS skills. Another limit is
related to the dissemination and exploitation activities of the computed results: for
performance reasons, the visualization is commonly provided via a conversion to
KML [19], where the link between the building performance data and its geometry is
color-coded in each building-style parameter and the other information stored in the
starting CityGML file is lost.
In the thermal image approach, instead, all the building use typologies are taken in-
to consideration, but the cost to collect thermal images to cover an entire city is hardly
negligible. Furthermore, only the roof surface is evaluated in terms of energy perfor-
mance, ignoring the full contribution of walls. Moreover, the use of proprietary stand-
ards does not encourage the adoption of the same solution by the research community
The approach we present in this article belongs to the typological kind, but makes
an effort to reduce the common drawbacks that have been delineated. As will be de-
scribed in more details in the following sections, our approach is in effect hybrid,

leveraging on the outcomes of project TABULA-EPISCOPE [2] but limiting the use
of building parameters estimated typologically.
3 SUNSHINE Approach
3.1 System Architecture
In this chapter, the system architecture of the SUNSHINE platform is presented. As
reported in the introduction, the SUNSHINE project covers three different scenarios;
however, given the focus of the paper on the first scenario, the system architecture
description has been rearranged to focus on the components that are meaningful in
this context.
Fig. 1. System Architecture
The chosen approach for this scenario was that of leveraging on a building typolo-
gy database and the system architecture that has been designed to comply with it (see
Fig. 1) is based on a Services Oriented Architecture (SOA) with three tiers (data,
middleware and application layers). A SOA-based approach allows accessing to the
resources through a middleware layer in a distributed environment and thus avoids
single access-points limitations that are instead typical for desktop-based architec-
tures.
Data Layer.
The bottom level of the SUNSHINE system architecture is aimed at storing geometry
and semantic information about buildings and thematic raster data. The two funda-
mental components of this layer are the 3D CityDB [5] and the raster map repository.

The 3D City Database [17] is a free 3D geo database designed to store, represent,
and manage virtual 3D city models on top of a standard spatial relational database, up
to five different Levels of Detail (LoDs, see Fig. 2). The database model contains
semantically rich, hierarchically structured, multi-scale urban objects facilitating GIS
modeling and analysis tasks on top of visualization. The schema of the 3D City Data-
base is based on the CityGML [9] standard for representing and exchanging virtual
3D city models.
The 3D City Database is described via a PostGIS relational database schema and
specific SQL scripts are provided to create and drop instances of the database on top
of a PostGIS DBMS.
Fig. 2. CityGML LOD example [courtesy of http://www.simstadt.eu]
The raster map repository is a data file store aimed at containing geo-located or-
thophoto and elevation files in raster file format.
Middleware Layer.
The middleware layer is the core component of the SUNSHINE platform. Its duty is
to manage the connection between the application layer and the data layer, providing
access to the resources stored in databases and map repository. The middleware layer
is composed by the 3D CityDB [5] Importer/Exporter Tool and Apache Tomcat.
The 3D CityDB Importer/Exporter Tool allows interacting with the 3D City Data-
base or external data. The Apache Tomcat [1] is an open source software implementa-
tion of the Java Servlet and JavaServer Pages technologies.
Application Layer.
A further challenge that the SUNSHINE project took into high consideration is the
dissemination and exploitation of the reached results. A smart-city will become
smarter only if all the involved stakeholders (citizens, public administrations and
government agencies) are aware about the outcomes of the research activities in that
particular scope. For this reason, a great effort was put in designing and implementing
a client platform that would be usable by the majority of devices, both mobile and
desktop-based.

To achieve the widest dissemination possible for the project’s results, the emerging
WebGL technology [14] has been employed, in conjunction with HTML5 elements,
as the main component of the application layer. WebGL is a cross-platform royalty-
free web standard for a low-level 3D graphics API based on OpenGL ES 2.0, exposed
through the HTML5 Canvas element as Document Object Model interface.
3.2 Energy Map Generation
The aim of this section is to describe how each building is associated with an energy
class index. The energy class is an index describing the energy performance of a
building and it is usually computed from a series of detailed information on building
energy properties that are not available in general as public domain data. Publicly
available data is usually limited to more basic information, such as building geometry,
year of construction, number of building sub-units, etc. So, the approach we followed
was to estimate the energy parameters needed for performance calculation from the
few publicly available data.
More specifically, the data necessary to the estimation are:
 Geometrical data: i.e. footprint, height, number of floors, etc. From these data,
using specific geoprocessing procedures, other geometrical properties are derived,
such as the building volume of the extent of the building walls shared with neigh-
boring buildings.
 Thermo-physical data: i.e. period of construction, prevalent building use, refur-
bishment level. From these data, using a typological approach and leveraging on a
sample of representative buildings for the different thermo-physical typologies, the
thermal properties of each building are estimated, such as U-values of envelope el-
ements and the percentage of windowed surface.
 Climatic data: i.e. the extent of the heating season and the average external tem-
peratures. These data are derived from national and local directives.
As the following sub-sections will describe with more details, using the geomet-
rical, thermo-physical and climatic data and applying a simplified computation proce-
dure based on ISO 13790 and ISO 15316 (international standard protocols for the
energy sector), the following parameters are computed for each residential building:
 The energy need for heating;
 The energy need for heating and domestic hot water;
 The corresponding index for energy performance.
There are some relevant aspects to highlight about this approach. The first is relat-
ed to the fact that the building typological classification currently applies to residen-
tial buildings only and thus cannot be used to assess the energy performance of build-
ings with a predominant use that is other than residential (commercial, administrative,
industrial, educational, etc). As a consequence, the energy map itself will carry infor-
mation only for residential buildings. This seems to us a reasonable compromise as
residential buildings are among the major causes for energy consumption and air pol-
lution [6].

A second important aspect is the use of thermo-physical typologies in order to es-
timate building properties that would be otherwise hardly obtainable on a large scale
without employing a great deal of resources (money and time) and whose knowledge
is instead necessary to determine an estimate of energy performance. The definition of
these typologies is based on the results of project TABULA [13], integrated and ex-
tended to adapt to the specificities of SUNSHINE. Project TABULA defined a set of
building typologies for each of the countries participating into the project, basing on 4
parameters: country, climate zone, building construction year, building size type (i.e.
single family house, terraced house, apartment block, etc). A building stereotype,
described in all its geometrical and thermo-physical properties, is associated to each
class, with the aim of representing the average energy behavior for buildings of that
class. So, if the 4 parameters are known, than it is possible to associate the building to
a specific typology class and thus to its estimated energy performance class.
As anticipated, the energy performance estimation approach developed in
SUNSHINE is hybrid, it differs from the typological approach followed in TABULA
because it limits the use of typological estimation only to the thermo-physical proper-
ties, using instead a deterministic approach for the geometrical properties, measured
or computed, of the involved buildings. A fully typological approach has in fact the
intrinsic limitation that the statistical significance of the performance estimation di-
rectly proportional to the scale at which the approach is applied, so very low at the
scale of the single building. A hybrid approach that takes into account the real geo-
metrical properties of the building makes the estimate of the building energy perfor-
mance more accurate. The validation step that will be described in section 3.3 makes
this approach even more robust.
Input Data Model.
To implement the conceptual idea explained in the previous section, a set of data re-
lated to the buildings in the pilot urban scenarios of the cities of Trento and Cles,
Italy, has been collected and the SUNSHINE energy map estimation workflow has
been executed on it. Table 1 contains the data model created to collect such buildings
data.
Table 1. Energy maps input data model.
Attribute Name Type
Building identifier string
Building geometry geometry
Begin construction year integer
End construction year integer
Building height real
Floors integer
Average floor height real
Refurbishment level {no/standard/adv}

Use string
Energy Performance Estimation.
This section is the core of this article, here is described the automatic process for
large-scale building energy performance estimation, that is the base for the energy
maps. The workflow illustrated in this section is implemented by using the following
software:
 A Postgres relational database with PostGis extension, where an input shapefile,
structured in accordance with the input data model, is loaded by the SHP2PG tool.
 FME 2014 Professional edition, ETL tool for manipulating spatial data. The same
results can be achieved also by the use of free ETL tools, for example GeoKettle.
The procedure implemented by FME 2014, connected with the PostGIS database
where input shapefiles are loaded, is represented in Fig. 3.
Fig. 3. Implemented energy maps generation workflow
More in details:
1. The workflow starts initializing the system;

2. For each building, all the data gathered in the input data model is used and, in addi-
tion, additional geometries parameters are computed such as area, perimeter,
shared and exposed walls perimeter;
3. The building typology, according to the categories provided by TABULA, is esti-
mated with the algorithm described in figure Fig. 4;
4. Using the previous estimated parameters, it is possible to query the TABULA da-
tabase in order to obtain the set of U_VALUES in accordance with the climatic
zone, typology, construction year and refurbishment level;
5. Having the set of U_VALUES and the real geometry proprieties it will be possible
to estimate the Energy Performance Index (EPI) according with the EN ISO 13790
regulation.
6. An output shapefile extending the input data model with the new geometrical and
thermo-physical data is produced. More details regarding the output data model
will be provided in the next section.
Fig. 4. Building size type estimation
Output Data Model.
The workflow illustrated in the previous section produces a new data model that is
described in Table 2. As it is possible to see, the workflow’s output is an extension of
the input data model, where thermo-physical and additional geometrical data are pro-
vided. It is important to underline the fact that the set of U_VALUES are the ones
provided by TABULA, while other data are computed starting from geometrical and
topological information.

Table 2. Energy maps output data model
Attribute Name Type
Building identifier string
Building geometry geometry
Begin construction year integer
End construction year integer
Building height real
Floors integer
Average floor height real
Refurbishment level {no/standard/adv}
Use string
Area real
Perimeter real
Shared perimeter wall real
Exposed perimeter wall real
U_roof real
U_floor real
P_win real
U_wall real
U_win real
EPI real
EPGL real
CRS string
Delta_U_bridge real
Building_typology int
Heating_days int
Pilot_id int
Irradiation real
Climatic_zone_id int
The final data model chosen to host the 3D buildings and their data is CityGML [9]
at the LoD-1 level of detail (see Fig. 2) that will be extended with a new Application
Domain Extension (ADE) defined with the aim of describing the properties of the
buildings in the energy domain. The definition of this ADE is the outcome of a joint
effort of a consortium of research institutions and private companies interested in
expanding CityGML model to support thermal building modeling at different level of
detail and complexity. The authors of this paper are members of this working group
which is led by the Modeling Working Group of the Special Interest Group 3D [16].
3.3 Energy Map Validation
Validation of the model is carried out comparing the estimated energy performance
with real energy certifications, for residential building built after 1900 in the urban

environment of Trento and Cles, located in the north of Italy. There are several con-
siderations to take into account:
1. The SUNSHINE Workflow provides an estimation based on the whole building
geometry while energy certifications are apartment-based;
2. The apartment position (ground floor/middle floor/last floor) influences the heat
loss. For this reason, the energy performance estimation workflow is refined as fol-
lows and performed for the three above mentioned conditions. According by EN
ISO 13790:
( ) ( ) (1)
( ) (2)
( ) ( ) (3)
With:
{ }
In particular:
{ }
{ } { }
{ }
3. Different software used to calculate the energy performance index produces simi-
lar, but not identical, results. The differences between these results are variable
and can arrive to 20% in the worst case.
Table 3 reports a subset of the whole set of buildings involved in the two urban en-
vironments where, for each entry, the following information is reported:
 Building typology, in according with the TABULA classification;
 City (TN/CL);
 Number of floors;
 Construction year;
 Real EPI [KWh/(m2
year)]: energy performance index provided by real certificates;
 Estimated EPI – whole building [KWh/(m2
year)]: estimated by the analysis on the
entire building. This is the output of the unrefined SUNSHINE energy map work-
flow;
 Final EPI [KWh/(m2
year)]: best estimated EPI as output of the refined energy map
workflow.
Real data of some building are listed in Table 1 shows, where hypothetic floor
number was calculated by comparing the reference area and shape one.

Table 3. Samples of validation result
ID Type City Floors
Year
Real EPi Est. EPi Fin. EPi
Building 1 SFH TN
3 1934 394,29 390,38
390,38
Building 2
MFH
TN
3 1920 362,01 519,38
400,48
Building 3 MFH TN 4
1990 256,97 301,8
259,56
Building 4 MFH TN
2 1961 230,3 342,22
238,12
Building 5
MFH
CL 2
1973 82,26 128,44
76,88
Building 6 SFH TN 1
2004 89,32 89,95
89,95
Building 7
SFH
TN
1 2012 94,28 112,03
112,03
Building 8
SFH
CL 4 2007 162,27 130,09
130,09
Building 9
MFH
TN
1
2012 42,92 128
53,56
Building 10 MFH CL
2
2012 56,01 96,4
59,04
Block 1 AB TN
4
1992 73,4 68,63
68,63
Block 2 AB TN
5
2004 87,41 94,64
94,64
Block 3
AB TN
1
1950 191,21
232,92 232,92
Block 4 AB
TN 1
1950 314,29 282,14
282,14
The same results are reported in Fig. 5.
Fig. 5. Validation against buildings reported in Table 1
Some considerations on the validation process:
 For old buildings, TABULA overestimates the set of U_VALUES;
 The real energy certificates do not include information on the refurbishment level
of the envelope.
 The Delta U Bridge for recent buildings is, in general, overestimated: around 10%
independently by the construction year.

Fig. 6 represents the comparison of real and estimated data on the whole dataset
composed by two cities, by year of construction.
Fig. 6. Validation graph against the entire dataset
The percentage gap between two values is much greater in the recent buildings, be-
cause uncertainties in glazing area are more relevant than U-values.
The following diagram represents the normal distribution of absolute errors. As it
is possible to see, the average error between estimated and real data is near the 21%
but, taking into account the previous considerations on the validation process, it is
easy to understand that the error factor can be effectively decreased.
Fig. 7. Errors distribution against real and estimated data
3.4 Energy Map Visualization
As described in the previous section about system architecture, via the WebGL-
enabled visualization the project stakeholders can easily discover, compare and per-
form statistics on the estimated energy map data by accessing to a classical HTML
web page.

Fig. 8. Example Building color-coding scale [courtesy of
http://www.greenbuildingadvisor.com]
Energy maps are generated merging geometry LOD-1 information from the
CityGML of the displayed city with the output of the energy performance estimation
procedure. More specifically, the color of each extruded KML polygon will be de-
pendent on the estimated building energy class. The reference between each building
in the KML file and the corresponding building in the 3D CityDB is ensured by stor-
ing the unique GML UUID of the building in the KML polygon name property. By
the use of a web service it will then be possible to retrieve the energy-related parame-
ters corresponding to the selected object.
The following code shows an example of how each building is specified in the
KML file.
<Style id="F">
<PolyStyle>
<color>FF0000FF</color>
<fill>1</fill>
<outline>0</outline>
</PolyStyle>
</Style>
<Placemark>
<description>Building extruded test 1</description>
<name>UUID_a2017297-d0cf-45ee-ae6d-
94a5d4fcda03</name>
<styleUrl>#F</styleUrl>
<Polygon>
<altitudeMode>absolute</altitudeMode>
<extrude>1</extrude>
<outerBoundaryIs>
<LinearRing>
<coordinates>
11.1263299929,46.0683712643,213.486
11.1263070656,46.0684180254,214.313
11.1262141309,46.0684011600,208.837
11.1262401711,46.0683529640,213.74
11.1263299929,46.0683712643,213.486
</coordinates>
</LinearRing>
</outerBoundaryIs>
</Polygon>
</Placemark>
Referring to the code listed above, the first part is used to make a visual representa-
tion of the energy class determined by the estimation procedure. Fig. 8 shows the

color coding of each building based on its estimated energy class. The second part of
the KML code is used to describe the extruded geometry of each building contained
in the source file.
Fig. 9 shows, the energy map visualizer is composed by two interconnected parts:
1. An HTML5 canvas that displays the WebGL virtual globe in which KML energy
maps, based on CityGML LOD-1 geometries, are loaded;
2. A classical HTML tab, displaying the detailed energy data corresponding to the se-
lected building. Comparisons between building energy efficiency characteristics
can easily be performed using the “radar” diagram placed in the bottom-left part of
the page. The diagram allows the comparison of the most important building pro-
prieties between the current and the previously selected building.
Fig. 9. Energy map visualization
3. For each involved building, the web client allows the possibility to refine the main
algorithm input data, increasing the energy performance index estimation accuracy
and providing some information for the apartment-level estimation (position of the
apartment: ground, middle, last: already discussed in the previous chapter). Fig. 10
shows how the web client allows stakeholders to use the building analysis and
simulation tool in order to run the client-side EPI estimation.

Fig. 10. Building Analysis and Simulation Tool
4 Conclusion and Future Developments
In this paper we have presented some of the preliminary results of the SUNSHINE
project. The use of a part of the TABULA building typology database with real build-
ing geometry information allows, for a large-scale, application of the building energy
performance assessment and the underlying service-oriented system architecture sup-
ports a distributed access to the related services. Moreover, the use of the emerging
WebGL technology ensures the largest available audience in terms of devices, both
desktop and mobile, avoiding the development of device-dependent custom clients for
3D city map visualization.
Future developments will be linked to the estimation of energy performance data
related to historical buildings. On the side of data structure and visualization, im-
provements will be focused on increasing the quality of the geometry displayed, mak-
ing it possible to render buildings based on CityGML LoD-2 level of detail and on the
development of more detailed building size type estimation procedures.
Acknowledgements.
The project SUNSHINE has received funding from the EC, and it has been co-funded
by the CIP-Pilot actions as part of the Competitiveness and innovation Framework
Programme. The authors are solely responsible of this work, which does not represent

the opinion of the EC. The EC is not responsible for any use that might be made of
information contained in this paper.
References
1. Apache Community, Apache Tomcat. http://tomcat.apache.org/
2. Ballarini, I., et al. Definition of building typologies for energy investigations on
residential sector by TABULA IEE-project: application to Italian case studies.
Proceedings of the 12th. International Conference on Air Distribution in Rooms,
Trondheim, Norway, p. 19-22 (2012)
3. Bowerman, B., et al. The vision of a smart city. 2nd International Life Extension
Technology Workshop, Paris (2000)
4. Carrión, D., et al. Estimation of the energetic rehabilitation state of buildings for
the city of Berlin using a 3D city model represented in CityGML. ISPRS interna-
tional conference on 3D Geoinformation, p. 4 (2010)
5. Dalla Costa, S., et al. A CityGML 3D geodatabase for buildings’ energy efficien-
cy. International Archives of the Photogrammetry, Remote Sensing and Spatial
Information Sciences 38.4: C21 (2011)
6. Fenger, J. Urban air quality. Atmospheric environment, 33.29: 4877-4900 (1999)
7. Giffinger, R. Smart cities: Ranking of European medium-sized cities. Final re-
port, Centre of Regional Science, Vienna UT (2007)
8. Giffinger, R., Gudrun, H. Smart cities ranking: an effective instrument for the
positioning of the cities? ACE: Architecture, City and Environment, vol. 4, num.
12, p. 7-26 (2010)
9. Gröger, G., et al. OpenGIS city geography markup language (CityGML) encod-
ing standard. Open Geospatial Consortium Inc (2008)
10. Hay, G., et al. HEAT - Home energy assessment technologies: a web 2.0 residen-
tial waste heat analysis using geobia and airborne thermal imagery. The Interna-
tional Archives of the Photogrammetry, Remote Sensing and Spatial Information
Sciences, Vol. XXXVIII-4/C7 (2010)
11. Kaden, R., Kolbe T. City-wide total energy demand estimation of buildings using
semantic 3D city models and statistical data. ISPRS Annals of the Photogramme-
try, Remote Sensing and Spatial Information Sciences, Volume II-2/W1 (2013)
12. Krüger, A., Kolbe, T. Building Analysis for urban energy planning using key
indicators on virtual 3D city models - The energy atlas of Berlin. International
Archives of the Photogrammetry, Remote Sensing and Spatial Information Sci-
ences, Volume XXXIX-B2 (2012)
13. Loga, T. Use of Building Typologies for Energy Performance Assessment of
National Building Stocks: Existent Experiences in European Countries and
Common Approach. IWU (2010)
14. Marrin, C. WebGL specification. Khronos WebGL Working Group (2011)
15. Nouvel, R., et al. CityGML-based 3D city model for energy diagnostics and ur-
ban energy policy support. Proceedings of BS2013, 13th Conference of Interna-
tional Building Performance Simulation Association, Chambéry, France (2013)

16. Special Interest Group 3D, http://www.sig3d.org/index.php?&language=en
17. Stadler, A., et al. Making interoperability persistent: A 3D geo database based on
CityGML. 3D Geo-Information Sciences, Springer, p. 175-192 (2009)
18. Washburn, D., Sindhu, U. Helping CIOs Understand “Smart City” Initiatives.
Report by Forrester Research, Inc (2009)
19. Wilson, T. OGC® KML. OGC Encoding Standard, Version 2.0 (2008)

S.1.c Building Energy Performance Estimation

More Related Content

What's hot (20)

Viewers also liked (18)

Similar to S.1.c Building Energy Performance Estimation (20)

More from SUNSHINEProject (6)

Recently uploaded (20)

S.1.c Building Energy Performance Estimation