Building and Environment 104 (2016) 92e101

Contents lists available at ScienceDirect

Building and Environment

journal homepage: www.elsevier.com/locate/buildenv

Clusters and exemplars of buildings towards zero carbon

Wei Pan a, *, Kaijian Li a, b
a
Department of Civil Engineering, The University of Hong Kong, China
b
School of Construction Management and Real Estate, Chongqing University, China

a r t i c l e i n f o a b s t r a c t

Article history: In addressing anthropogenic climate change many buildings worldwide have been designed and con-
Received 14 December 2015 structed towards zero carbon. However, their cross-context learning is largely constrained. The aim of
Received in revised form this paper is thus to achieve a better understanding of the clusters and the energy strategy and per-
31 March 2016
formance of buildings towards zero carbon worldwide. The research was carried out through the com-
Accepted 22 April 2016
Available online 23 April 2016
bination of a two-step cluster analysis of several hundreds of low or zero carbon buildings and case
studies of five exemplars. Five clusters of these buildings were revealed, which identified the gaps in the
knowledge of high-rise buildings towards zero carbon and of decarbonizing the building stock. The
Keywords:
Zero carbon building
results of the case studies of exemplars illustrate the systems integration of strategies for buildings'
Low carbon energy efficiency and energy generation and supply, but also indicate different concepts and calculating
Zero energy methodologies of ‘zero carbon’ or ‘zero energy’. This inconsistency significantly hampers the bench-
Energy performance marking of buildings' energy performance and carbon reduction practices. Net-zero carbon, particularly
Cluster analysis net-zero energy, was found to be technically difficult to achieve over the one-year period of operation.
The findings cast a shadow over the feasibility of achieving net zero carbon particularly for high-rise
buildings due to geographic constraints for use of renewable energies, which encourages the explora-
tion of emerging energy and carbon reduction technologies.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction
The approach of zero carbon building has been promoted in
many countries and regions as a government strategy for
addressing anthropogenic climate change [1e3]. This promotion
has made an essential contribution to the momentum of ‘zero
carbon buildings’; many have emerged worldwide and been re-
ported through various channels. Markedly, the International En-
ergy Agency's (IEA) Solar Heating & Cooling (SHC) Program Task 40
‘Net Zero Energy Solar Buildings’ has analyzed and reported the
conceptual approaches of several hundreds of ‘net zero-energy and
energy-plus buildings’ worldwide [4]; the Zero Carbon Hub's (ZCH)
Building Profiles showcase ‘zero carbon homes’ at national level
[5]; various researchers reported on individual or groups of projects
of ‘zero carbon buildings’ [6,7]. From the worldwide promotion and
reporting more ‘zero carbon buildings’ can be expected to be
developed for future years. Cross-context learning of the past and
existing practices is thus important, which however is largely
* Corresponding author. Department of Civil Engineering, The University of Hong
Kong, Pokfulam, Hong Kong, China.
E-mail address: wpan@hku.hk (W. Pan).
constrained. Also, there have been increasing government and ac-
ademic concerns about the gap between as-designed and as-built
energy performance of buildings [8,9], which is being further
explored and addressed by, e.g. the IEA-EBC Annex 66 ‘Definition
and Simulation of Occupant Behavior in Buildings’ that addresses
the gap between simulation and measurement. However, there is a
lack of study of the profiles, patterns and energy strategy and
performance of ‘zero carbon buildings’.
The aim of this paper is thus to achieve a better understanding
of the clusters and the energy strategy and performance of build-
ings towards zero carbon worldwide. There are three research ob-
jectives: (1) to reveal the profiles and clusters of the buildings
towards zero carbon; (2) to examine the energy strategy for
building ‘zero carbon’; and (3) to investigate the energy perfor-
mance of exemplary ‘zero carbon building’ cases. The research was
carried out through the combination of a two-step cluster analysis
of several hundreds of low or zero carbon buildings and case
studies of five exemplars that represent the clusters. Following this
introduction the paper reviews the concept of ‘zero carbon’. It then
reveals five clusters of the buildings studied, followed by the ex-
amination of the energy strategy and investigation of the energy
performance of exemplary building cases. The paper discusses the

http://dx.doi.org/10.1016/j.buildenv.2016.04.027
0360-1323/© 2016 Elsevier Ltd. All rights reserved.
 W. Pan, K. Li / Building and Environment 104 (2016) 92e101
implications of the findings before conclusions are drawn.
2. The concept of ‘zero carbon building’
Although research on buildings' energy and carbon emissions at
least dates back to the 1940s [10], the concept of ‘zero carbon
building’ was probably first formalized in 2006 in the UK Govern-
ment's policy publication of seeking for a step-change in sustain-
able homebuilding practice [11]. This policy well signposts the start
of governments' promotion for zero carbon buildings worldwide.
There have emerged a number of other similar but different terms,
often used interchangeably, which are mostly associated with their
relevant policy frameworks, e.g. ‘nearly zero-energy building’ in the
EU [12], ‘zero emission building’ in Australia [13,14], and ‘net-zero
energy buildings’ in the US [15,16]. Some researchers have
expanded the scope of the concept of ‘zero carbon building’ to also
include the many other building energy and carbon relevant con-
cepts/terms in their studies. For examples, the Concerted Action
report [17] presents different terms under the banner “high per-
formance buildings”, which are used in European Union (EU)
member states. They suggested that the many terms could broadly
be categorized as referring to low energy consumption, low emis-
sions or sustainable or green aspects. Riedy et al. [13] identified
many similar terms in common use, such as “near zero energy; zero
energy; zero net energy; passive house; energy plus; fossil fuel
free; 100% renewable; zero carbon; net zero carbon; carbon
neutral; climate neutral; climate positive and positive develop-
ment.” Most of these terms are used in real-life projects of building
towards zero carbon. The many terms, coupled with their associ-
ated policies, calculation methodologies and practices, make the
concept of ‘zero carbon’ so complicated that the learning of the
practices in different contexts becomes implicit and difficult.
3. Research method
This research was carried out through the combination of a two-
step cluster analysis of low or zero carbon buildings and case
studies of five exemplars. The cluster analysis identifies and ex-
amines the clusters of those buildings. The case studies elaborate
the clusters drawing on exemplary building cases, examine the
strategies for energy efficiency and energy generation and supply,
and investigate the buildings' energy performance.
3.1. Selection of building cases for cluster analysis
Three considerations were taken for the selection of the low or
zero carbon building cases for inclusion in this study. First, the
keywords of ‘net, nearly and/or lifecycle zero carbon’ or ‘zero en-
ergy building’ were used to describe and report the cases. This
consideration addresses the complexity of ‘zero carbon’ and its
related concepts. Second, information on the selected building
cases was obtainable from accessible sources or channels. Not many
buildings were reported on their designed energy performance and
carbon emissions, fewer with measured data. Those reported with
relevant data often did not provide details of the scope of energy
use and carbon emissions and their relevant calculation method-
ology. Third, the selection of the cases of low or zero carbon
buildings aimed to represent a worldwide overview of the prac-
tices, rather than being constrained to any specific countries.
Several important sources were identified in the selection of
building cases. First was the source in the form of Google maps
view developed by the research group within the IEA's ‘Towards
Net Zero Energy Solar Buildings’ Project [4]. This source analyses
the conceptual approaches of ‘net zero-energy and energy-plus
buildings’ worldwide as of June 2013 with a primary focus on
93
Germany and the rest of Europe. This source presents basic project
data illustrated and provides web-links to most of the projects
included. Second was the ‘Low Impact Housing’ in the form of a
website with search function that is the outcome of a survey of low
impact houses in the world with a focus on North America [18]. The
building profiles in this source were provided through a two-round
survey with relevant professionals and professional organizations.
Third was the source provided by the Zero Carbon Hub which is an
independent non-profit public/private partnership, established to
take day-to-day operational responsibility for coordinating the
delivery of low and zero carbon new homes in the UK [5]. These
sources were complemented by a wide search in public domains
taking into account the three abovementioned considerations for
building case selection. In total 600 buildings were selected, which
together provide a reasonable representation of low or zero carbon
buildings in the market.
3.2. Method of cluster analysis
The Statistical Package for the Social Sciences (SPSS) ‘TwoStep
Cluster’ method was applied to reveal the clusters of the selected
buildings. This method is designed to discriminate natural groups
from a set of variables stabilizing the nearness criterion, with a
hierarchical agglomerative clustering whose centers are far apart
[19]. Compared to classical cluster analysis methods, SPSS ‘TwoStep
Cluster’ can deal with both continuous and categorical attributes.
Also, this method can automatically determine the optimal number
of clusters. Likelihood was selected as the distance measure, which
defines the normal density for continuous variables and the
multinomial probability mass function for categorical variables. The
cluster analysis involved two steps:

Pre-clustering step: the data records were scanned one by one
and the algorithm decided whether the current record could be
added to one of the previously formed clusters or it started a
new cluster, based on the distance criterion;

Clustering step: the clustering stage had sub-clusters resulting
from the pre-cluster step as input and grouped them into the
optimal number of clusters. To determine which number of
clusters was optimal, each of these cluster solutions was
compared using Schwarz's Bayesian Information Criterion (BIC)
as the clustering criterion. An optimal number of clusters will
have a smaller value of the BIC, a reasonably large Ratio of BIC
Changes and a large Ratio of Distance Measures.
Silhouette Coefficient [20], a measure of density of all the data in
the cluster, was utilized to measure the goodness-of-fit of the
outcome. This index combines both, cohesion (based on the
average distances between all the objects in a cluster) and sepa-
ration (based on the average distance of any object to all the other
objects not contained in the same cluster), and can range
between 1 and þ 1; values below 0 are indicative of inappropriate
fit, between 0 and 0.2 are poor, between 0.2 and 0.5 are fair, and
above 0.5 are good.
3.3. Selection of variables for cluster analysis
Although previous studies have examined the ‘metrics’ of zero
energy buildings [21,22], the ‘characteristics’ of net zero energy
buildings [23] and the ‘system boundaries’ of zero carbon buildings
[7], no previous research has yet explicitly examined the variables
of low or zero carbon buildings. This seems to be attributed to the
many existent similar but different concepts relevant to building
energy and carbon and their different policy contexts. It may also be
due to the different methodologies for calculating the energy
94 W. Pan, K. Li / Building and Environment 104 (2016) 92e101
consumption and carbon emissions of buildings and the subse-
quent difficulty in benchmarking and cross comparison. However,
the variables or attributes of the selected building cases from
different contexts must be considered for achieving effective cross-
context learning. Drawing on literature this research selected 10
descriptive and explanatory variables for the cluster analysis. The
descriptive ones included year of completion, climatic zone,
building type, location, number of storeys and building size, and
were used to describe the tangible features of the building cases.

‘Year of completion’ denoted the year of construction or reno-
vation of the building.

‘Climatic zone’ was used as a proxy of the geographic location of
the building, which followed the Ko €ppen climate classification
scheme that divides climates into five main groups, each having
several types and subtypes, and combines average annual and
monthly temperatures and precipitation, and the seasonality of
precipitation [24].

‘Building type’ denoted the functional type of the building, e.g.
experimental buildings that were built to demonstrate or test
low or zero carbon technologies and were usually owned by
clean energy suppliers, institutions, or government sectors,
services buildings that included recreation and transportation
facilities such as sport centre and land port of entry.

‘Location as per density’ denoted the location of the building in
relation to urban density, e.g. urban areas that were character-
ized by higher population density and vast human features, sub-
urban areas stood for an outlying part of a city or town or a
smaller community adjacent to or within commuting distance
of the urban area, and rural areas that were neither urban nor
sub-urban areas.

‘Number of storeys’ referred to the maximum number of storeys
of the building.

‘Building size’ denoted the gross floor area (GFA) of the building,
inside the building envelope, including the external walls but
excluding the roof.
The explanatory variables included building ownership, status,
grid connection, and geographic boundary, and were used to
explain the less tangible features of the cases, and thus to enable a
better understanding of the profiles and clusters of the cases.

‘Building ownership’ referred to the sector of the leading
ownership of the building, e.g. public buildings that were
government-funded, including government offices, educational
buildings, transportation infrastructures, public affordable
houses, and recreation facilities, etc. and private buildings that
were owned by private developers or individuals, including the
private houses, commercial buildings and private offices, etc.
and mixed-ownership buildings that were developed jointly by
government and private developers or individuals.

‘Building status’ meant the major status of the building, e.g.
new-build, renovation and mixed.

‘Grid connection’ described if and how the building was con-
nected with power grid. Those buildings that produced energy
on-site and feed electricity into the mains or national grid were
categorized as ‘on-grid (two-way)’; those that only took elec-
tricity from grid but did not feed into the mains or national grid
were categorized as ‘on-grid (one-way)’; those that were not
connected with the mains or national grid were termed ‘off-
grid’.

‘Geographical boundary’ referred to the scope of the facility in
relation to energy and carbon, e.g. the building itself, the project,
the community or the city.
3.4. Method of case studies of exemplars
Five exemplars of the building cases (see Table 1) were selected
for the case studies. Each exemplar represented one of the five
revealed clusters of the buildings, i.e. the characteristics of the
selected exemplars were consistent with their associated clusters.
Information of the exemplars was first obtained in publicly acces-
sible domains and then verified with the project stakeholders
through site visits, meetings and/or email communications. Such
information included project specifics as well as data on the
buildings' energy performance and carbon emissions. Through the
case studies, the strategies for achieving ‘zero carbon’, including
strategies for buildings' energy efficiency and energy generation
and supply, were examined, and the buildings' energy and carbon
emission performance was investigated.
4. Results and analysis
4.1. Overview of the building cases
The results of analysing the 600 selected low or zero carbon
buildings indicate the diverse values, or the diverse ranges of
values, of the building cases against the variables (Table 2). These
results suggest complicated and unstraightforward profiles of the
buildings towards zero carbon. To effectively reveal the clusters of
the buildings, the two-step cluster analysis was carried out.
4.2. Results of the two-step cluster analysis and the refining process
The two-step cluster analysis took into account three strategies
in order to elicit the building clusters. First, variables that are
dominated by only one value of the variables are avoided.
Accordingly, because the building cases for study were dominated
by ‘new-build’ (506; 87.7%), the concerned variable ‘building status’
was not included for the cluster analysis. Second, variables that
include significant missing data are removed. Accordingly, because
of the missing data of a significant number of building cases against
the variables ‘grid connection’ and ‘building size’, 217 (36.2%) and
205 (34.2%), respectively, these two variables were also not
included for the cluster analysis. After this refining process, a pre-
liminary cluster analysis was conducted using the remaining seven
variables. The resultant Silhouette Coefficient was found equal to
0.1, which indicates that the goodness-of-fit was not achieved.
The third strategy is to use the resultant Silhouette Coefficient of
0.2 or above as a control for revealing meaningful clusters [20]. The
use of this control further narrowed down the number of variables
to five, and as a result, the number of the building cases was refined
to be 404 (67.3%) for being included in more detailed cluster
analysis. With the five finalized variables, each of the cluster so-
lutions was compared using Schwarz's BIC as the clustering crite-
rion in order to determine the optimal number of clusters. The
optimal number of cluster should have a smaller value of the BIC, a
reasonably large Ratio of BIC Changes and a large Ratio of Distance
Measures. The results of this analysis suggest that the optimal
number of clusters was five, with a smaller value of the BIC
(2353.658), a larger Ratio of BIC Changes (0.311) and a larger Ratio
of Distance Measures (1.549). These results suggest that goodness-
of-fit was achieved, with fair average Silhouette Coefficient equal to
0.30.
Through the ‘TwoStep Cluster’ analysis five clusters of the
buildings (404; 67.3%) were revealed, which are further examined
in the following sections. These clusters involved the use of the
variables: building ownership, climatic zone, building type, loca-
tion as per density, and number of storey. These clusters were with
the sample size of 86 (21.3%), 75(18.6%), 100 (24.8%), 54 (13.4%), and
 W. Pan, K. Li / Building and Environment 104 (2016) 92e101 95

Table 1
Overview of the exemplar buildings for case studies.

Building's name Eco Terra Pixel Green Lighthouse Elithis Tower Greenwatt way

Year of completion 2007 2010 2009 2009 2010

Concept adopted Near net zero energy solar house Carbon neutral building Carbon neutral building Near net zero energy building Zero carbon development
Country Canada Australia Denmark France UK
Building ownership Private Private Public Private Private
Climatic zone Cold Temperate Temperate Temperate Temperate
Building type Residential Commercial Educational Commercial Residential
Location of density Rural Urban Urban Urban Suburban
Number of storeys 2 4 3 10 2

Table 2
The range and values of selected buildings (n ¼ 600).

Variable Type Identified buildings' values and ranges of values

Year of completion Descriptive; Ordinal New construction or renovation in the period 1988e2014
Climatic zone Descriptive; Nominal Tropical, arid, temperate, cold, polar and breakdown climatic zones
Building type Descriptive; Nominal Residential, commercial, services, industrial, experimental, educational, mixed use
Location as per density Descriptive; Nominal Urban, sub-urban, rural
Number of storeys Descriptive; Scale Range from 1 to 71
Building size Descriptive; Scale Gross floor area ranges from 40 to 324975 m2
Building ownership Explanatory; Nominal Public, private, mixed
Building status Explanatory; Nominal New-build, renovation, mixed
Grid connection Explanatory; Nominal On-grid (two-way), on-grid (one-way), off-grid
Geographic boundary Explanatory; Nominal Low-rise building, med to high-rise building, development, project, community, city, island

89 (22.0%), with the ratio of sizes comparing the largest to smallest
cluster being 1.85. The distribution of the building cases grouped in
the five revealed clusters against the involved variables is illus-
trated in Fig. 1.
4.3. Revealed clusters and exemplar case studies
4.3.1. Cluster 1: private residential low-rise rural ‘zero carbon
buildings’ in cold zone
There were 86 building cases (21.3% of 404) falling in Cluster 1,
which were dominated by private (100% of 86 cases) residential
(98.8%) low-rise (2.31) rural (60.5%) ‘zero carbon buildings’ in cold
zone (65.1%) (Fig. 1).
4.3.1.1. Exemplar of cluster 1: EcoTerra house, Canada. EcoTerra
House, one of the 15 Canada Mortgage and Housing Corporation
Equilibrium Demonstration projects in 2007, is a 2-storey, 237.7 m2
private prefabricated home that was assembled in 2007 in Quebec,
Canada [25]. Previous research [26] revealed that this building
consumed 12.4% of the energy of a standard house of the same size
in Canada and produced nearly as much energy as it used on an
annual basis.
4.3.1.2. Strategies for building's energy efficiency. Due to the cold
winter of Quebec, EcoTerra House was optimized for passive solar
heating; about 40% of the south facing façade area consisted of
windows [26]. No need was considered for mechanical heating on a
sunny day and the excess heat gain in the day could be stored for
night use. Significant heat loss normally occurs through the
building envelope. To combat this problem, EcoTerra used triple-

Fig. 1. The revealed clusters of selected buildings (n ¼ 404). Notes: ‘Building typology’ includes residential, commercial, services, industrial, experimental, educational and mixed
use building types; ‘Ownership’ denotes public, private and mixed building ownership; ‘Location’ denotes as per density, i.e. urban, sub-urban and rural; ‘Climatic zone’ includes
tropical, arid, temperate, cold, polar and breakdown climatic zones; ‘Number of storeys’ refers to the building cases, ranging from 1 to 71.
96 W. Pan, K. Li / Building and Environment 104 (2016) 92e101

glazed, low-e, argon-filled windows and energy-efficient building

envelope; the air-tightness of the house was reported as 0.8 air-
changes per hour at 50 Pa [25]. An innovative building-integrated
photovoltaic/thermal (BIPV/T) collector was installed in the
house. Outdoor cold air was heated by passing under the PV panels.
The temperature of the solar-heated air was approximately 35  C
warmer than the outdoor [26]. The heated air was sent into the
house through insulated ducting and could be used for space
heating, water pre-heating and clothes drying. The heated air was
then ducted to the basement slab for storage. The slab contains air
channels which allowed the slab to absorb heat as the air passed
through. Once the slab was heated, it radiated heat to the basement.
Due to the slab's large thermal inertia, this process occurred
gradually and kept the house warm through the night. A heat re-
covery ventilator (HRV) was used to provide fresh air to the
building, which captured heat from the exhaust air and transferred
it to the incoming outside cold air [27].

4.3.1.3. Strategies for energy generation and supply. The south-
facing roof area was covered by PV panels with the capacity of
2.84 kWp [25]. When the house was producing more electricity
than it was consuming, the meter ran backwards. A ground source
heat pump (GSHP) was used when the BIPV/T system could not
produce enough heated air [27]. The heat pump exchanged heat
between the house and the vertical wells, and was reported to have
the coefficient of performance (COP) of between three and four
[27].
4.3.1.4. Building's energy performance. Data of this building's en-
ergy performance were collected from December 2009 to
December 2010. The energy consumption included energy for
space heating, water heating, lighting, cooling, appliances and
equipment (Table 3). During this occupied year, the house had the
net electricity use of 40.9 kWh/m2/year (Table 3). The overall net
carbon emissions were estimated to be 818 gCO2/m2/year. Although
EcoTerra did not realize net-zero energy consumption, it was re-
ported to have consumed only 12.4% of the energy of a typical
Canadian home [26].
4.3.2. Cluster 2: private commercial low-rise ‘zero carbon buildings’
in temperate zone
There were 75 building cases (18.6% of 404) falling in Cluster 2,
which were dominated by private (100% of 75) commercial (77.3%)
low-rise (3.41) ‘zero carbon buildings’ in temperate zone (80%).
4.3.2.1. Exemplar of cluster 2: Pixel, Australia. Pixel is a 4-story,
1000 m2 private office building (Fig. 2) in Australia. A team
comprising Grocon (developer, builder, tenant and owner), stu-
dio505 (architect), and Umow Lai (sustainability consultant) was
put together to deliver Pixel with the aim to become the first fully
Fig. 2. Pixel, Australia. (Courtesy of CBRE).
carbon neutral office building in Australia [28].
4.3.2.2. Strategies for building's energy efficiency. Reducing energy
demand was the first consideration when the building was
designed. The daylight glare control system provided daylight
penetration into the office space automatically whilst allowing
screen based technologies to be deployed without the need for
blinds [29]. The smart window control system enabled night
ventilation by opening the windows to allow cool night air to flood
into and cool down the building [29]. The absorption chiller using
ammonia refrigerant was reported as the first of its kind in a
commercial building in Australia [29]. Other energy efficiency
measures included individual addressable dimming on office
lighting, and light-emitting diode (LED) lighting in all non-office
areas of the building, double glazed windows and natural
ventilation.
4.3.2.3. Strategies for energy generation and supply. A strategy of
low carbon intensity fuel sourcing and renewable energy genera-
tion was utilized to supply the energy on-site. Gas, with a carbon
intensity of 0.21 kgCO2/kWh, was selected as the fuel to fire the
absorption chiller, and a back-up for anaerobic digester over the
traditional Victorian brown coal powered grid electricity, which has
a carbon intensity over six times greater (1.34 kgCO2/kWh) [6].
Owing to the reduction in energy demand and low carbon intensity
fuel source, relatively small amount of renewable energy was
needed to achieve carbon neutral. On the ground level of the Pixel,
an anaerobic digester tank system collected all the black waste
from facilities of toilets and kitchen; methane was extracted from

Table 3
Measured end-use energy and estimated carbon emissions of EcoTerra.a

End-use energy Energy use data (kWh/m2/year) Source of energy Carbon conversion factorb (gCO2/kWh) Carbon emission data (gCO2/m2/year)

Space heating 14.3 Electricity 20 286

Water heating 6.6 Electricity 20 132
Lighting 4.6 Electricity 20 92
Cooling 0.5 Electricity 20 10
Appliance 15.2 Electricity 20 304
Other equipment 12.9 Electricity 20 258
Solar electricity 10.8 Electricity 20 216
Solar thermal 2.4 Electricity 20 48
Total 40.9 e e 818
a
Data of measured end-use energy and estimated carbon emissions of the building from 2009 to 2010; the conditioned space of the building: 237.7 m2.
b
Carbon conversion factor of electricity being hydro power in Quebec, Canada.
 W. Pan, K. Li / Building and Environment 104 (2016) 92e101 97

the waste for water heating [29]. Renewable energy systems, con-
sisting of fixed and tracking photovoltaic and wind turbines on the
roof, were designed into the building [29], with any excess
renewable energy produced to be exported back to the grid.

4.3.2.4. Building's energy performance. It was reported that a set of

simulation tools, including Indoor Climate and Energy and Radi-
ance and Ecotect, were applied to model the Pixel's energy per-
formance [29]. Prior studies [6] estimated that the Pixel had a net
savings of approximately 16 t CO2-e per year, although the net
energy use of the building was simulated to be 10.4 kWh/m2/year
(Table 4). This is because of the different carbon conversion factors
used for carbon emission estimates for natural gas (200gCO2/kWh)
which was used for space heating and cooling of Pixel and for
electricity (1200gCO2/kWh) which was partly offset by the use of
PV and wind turbines (Table 4). It is worth noting that the Pixel was
designed for carbon neutral rather than net zero energy, which in
fact was energy positive.
Fig. 3. Green Lighthouse, Denmark.
4.3.3. Cluster 3: public low-rise ‘zero carbon buildings’ in temperate
zone
There were 100 building cases (24.8% of 404) falling in Cluster 3, stored solar heat in the ground via a heat pump. In total there were
which were dominated by public (99%) low-rise (2.42) ‘zero carbon 76 m2 of solar panels installed on roof, providing electricity for the
buildings’ in temperate (58%) zone. building's lighting, ventilation and pumps. The other 65% of heating
and power was supplied through eco-friendly district heating [32].
4.3.3.1. Exemplar of cluster 3: Green Lighthouse, Denmark.
Green Lighthouse, of 1019 m2 in size and on three floors, was
4.3.3.4. Building's energy performance. The overall measured end-
officially inaugurated in 2009 to provide meeting facilities, faculty
use energy of the building in 2012 was 63.2 kWh/m2/year, partly
offices and student services for the University of Copenhagen's
offset by the energy generated on-site by PV and solar thermal heat
Faculty of Science (Fig. 3). Green Lighthouse was reported as the
pump, yielding a net energy use of 39.6 kWh/m2/year (Table 5). The
first public carbon neutral building in Denmark [30].
energy consumption included energy for space heating, water
heating, lighting, ventilation, running installations and appliances.
4.3.3.2. Strategies for building's energy efficiency. The Green Light-
The relevant carbon emissions of the building in 2012 were esti-
house was designed to reduce energy consumption by natural light
mated to be 7640.1gCO2/m2/year.
and ventilation, and highly efficient envelope [31]. Its circular form
ensured the supply of natural light to the inner atrium through roof
windows. The fresh air from façade windows was channelled 4.3.4. Cluster 4: private urban ‘zero carbon buildings’ in temperate
through the atrium before escaping through roof windows. The sun zone
was the building's main source for space and water heating. Solar There were 54 building cases (13.4% of 404) falling in Cluster 4,
energy was also used for floor heating, and the excess energy could which were dominated by private (88.9%) urban (92.6%) ‘zero car-
be stored [32]. Efficient windows minimised the heat loss while bon buildings’ in temperate (77.8%) zone.
maximising passive solar gain.
4.3.4.1. Exemplar of cluster 4: The Elithis Tower, France. The Elithis
4.3.3.3. Strategies for energy generation and supply. The energy Tower, of 4500 m2 in total on ten floors, completed in 2009, was
concept of the Green Lighthouse was to make the building carbon reported as the first nearly net-zero energy office building in France
neutral. Around 35% of the heating and power for the building [33]. The Elithis Tower sets an example of building a medium-rise
relied on the solar energy from solar collectors on the roof and zero carbon building in dense urban contexts.

Table 4
Simulated end-use energy and carbon emissions of Pixel.a

End-use energy Energy use data (kWh/m2/ Source of energy Carbon conversion factorb (gCO2/ Carbon emission data (gCO2/m2/
year) kWh) year)

Space heating 4.7 Natural Gas 200 940

Space cooling 24.1 Natural Gas 200 4820
Lighting 1.1 Electricity 1200 1320
Elevator 0.6 Electricity 1200 720
HVAC fans 3.8 Electricity 1200 4560
HVAC pumps 0.4 Electricity 1200 480
Condenser fans 1 Electricity 1200 1200
Hydraulic pumps 0.2 Electricity 1200 240
Miscellaneous loads, controls, and so forth 0.2 Electricity 1200 240
PV 10.4 Electricity 1200 12480
Wind turbines 15.3 Electricity 1200 18360
Total 10.4 e e 16320
a
Data of simulated end-use energy and carbon emissions of the building in 2010 sourced from Refs. [6]; the size of the building: 1000 m2.
b
Carbon conversion factors are taken from Australian National Greenhouse Accounts Factors 2011.
98 W. Pan, K. Li / Building and Environment 104 (2016) 92e101

Table 5
Measured end-use energy and estimated carbon emission of Green Lighthouse.a

End-use energy Energy use data (kWh/m2/year) Source of energy Carbon conversion factor (gCO2/kWh) Carbon emission data (gCO2/m2/year)

Space heating 36.6 District heating/heat pump 104b 3806.4

Water heating 0.8 District heating/heat pump 104 83.2
Ventilation 2.9 Electricity 377 1093.3
Lighting 6.3 Electricity 377c 2375.1
Running installations 0.3 Electricity 377 113.1
Appliance 16.3 Electricity 377 6145.1
PV 12.9 Electricity 377 4863.3
Solar thermal heat pump 10.7 Heat pump 104 1112.8
Total 39.6 e e 7640.1
a
Data of measured end-use energy and estimated carbon emissions of the building in 2012; the size of the building: 1019 m2.
b
Carbon conversion factor taken from Danish District Heating Company HOFOR Environmental Declaration 2013.
c
Carbon conversion factor taken from energinet.dk Environmental Impact Statement 2013.

4.3.4.2. Strategies for building's energy efficiency. The building en- rise (2.36) suburban (100%) ‘zero carbon buildings’ in temperate
velope included wood, recycled insulating materials and bay win- (95.6%) zone.
dows [34]. The main feature of the building was the combination of
high glazing rate (75%) windows and solar shields to exploit the 4.3.5.1. Exemplar of cluster 5: Grenwatt way, UK. Greenwatt Way,
benefits of the sun without suffering from the detriments, such as located in Slough, Berkshire (Fig. 4), was home to Scottish and
overheat and glare effect [34]. A heat recovery ventilation system Southern Energy (SSE) employees, Slough Borough Council staff
was used to achieve energy savings by recovering the heating and local residents. It was a development of ten homes completed
produced by office equipment (e.g. computers and photocopiers), in 2010 by a partnership by SSE. All the homes were designed to
the users and the facilities in the kitchen [35]. All offices had a meet the May 2009 definition of Level 6 of the Code for Sustainable
direct access to daylight to minimize artificial lighting needs, and a Homes [36].
“nomad lighting system” was used to guarantee visual comfort
without energy waste [35].
4.3.5.2. Strategies for building's energy efficiency.
Energy-efficient building envelope was applied in Greenwatt Way
4.3.4.3. Strategies for energy generation and supply. The underlying
philosophy of the Elithis Tower was to minimize the energy con-
sumption before adopting an energy production technology. A
small area of PV panels was thus integrated to the roof to meet part
of the electrical needs, and a wood pellet boiler was used to provide
space and water heating [35].

4.3.4.4. Building's energy performance. Elithis Tower was reported

to have a total primary energy demand of 97.2 kWh/m2/year and
40.2 kWh/m2/year of total primary energy generation [33]. The data
was collected in 2009 and included energy for space heating, water
heating, lighting, cooling, ventilation, elevator and appliances
(Table 6). The net energy use of the building was measured as
57 kWh/m2/year, and the relevant carbon emissions of the building
were estimated to be 4263.7gCO2/m2/year.

4.3.5. Cluster 5: private residential low-rise suburban ‘zero carbon

buildings’ in temperate zone
There were 89 building cases (22% of 404) falling in Cluster 5,
which were dominated by private (97.8%) residential (100%) low-

Fig. 4. Greenwatt Way, UK. (Courtesy of Zero Carbon Hub).

Table 6
Measured primary energy and estimated carbon emissions of Elithis Tower.a

Primary energy Energy use data (kWh/m2/year) Source of energy Carbon conversion factor (gCO2/kWh) Carbon emission data (gCO2/m2/year)
b
Space and water heating 6.3 Wood pellet boiler 15 94.5
Cooling 6.2 Electricity 82.72c 512.9
HVAC fans 14.1 Electricity 82.72 1166.4
HVAC pumps 2.6 Electricity 82.72 215.1
Lighting 9.5 Electricity 82.72 785.8
Elevators 3.6 Electricity 82.72 297.8
Appliances (plug loads) 54.6 Electricity 82.72 4516.5
PV power generation 40.2 Electricity 82.72 3325.3
Total 57 e 4263.7
a
Data of measured primary energy of the building in 2009 sourced from Refs. [33]; the size of the building: 4500 m2.
b
Carbon conversion factor taken from Carbon and energy balances for a range of biofuels options. Elsayed, MA, Matthews, R, Mortimer, ND. Study for DTI URN 03/836. The
figure does not include sourcing the feedstock and any pre-processing such as drying.
c
Carbon conversion factor taken from 2012 Guidelines to Defra/DECC's GHG Conversion Factors.
 W. Pan, K. Li / Building and Environment 104 (2016) 92e101
to meet a heat loss parameter of 0.8 (required for Code Level 6) [36].
A north facing roof-light was provided above the stairs to allow
natural daylight in. The roof-light also acted as a “chimney”,
opening in summer to draw warm air out of the house. In addition
to the building fabric for reducing the need for heating and lighting,
energy efficiency strategies were also applied. These included a
Mechanical Ventilation and Heat Recovery (MVHR) system, ultra
efficient appliances, low water use fittings, smart metering and
various ‘smart kits’ [36]. Furthermore, the dishwasher and washing
machine were hot and cold fill to maximise the use of available zero
carbon heat [37]. Most homes included a grey water recycling
system to recycle bath and shower water to flush toilets and
recovered waste heat [37]. As a result of utilising all these measures,
the homes were expected to consume 90% less energy than a
typical existing home in the UK [36].
4.3.5.3. Strategies for energy generation and supply. Heating and hot
water was supplied to all the homes through a low temperature
and thus low heat loss district heating system serviced from the
Energy Centre, which ran the technologies including solar thermal
panels, an air source heat pump (ASHP), a ground source heat pump
(GSHP) and a biomass boiler [38]. The district heating system
operated at a temperature of 55  C. A plate heat exchanger was
installed within each house to provide domestic hot water on de-
mand, drawing heat from the district heating ring and feeding ra-
diators directly. The flow from the radiators was fed to the heater
battery within the MVHR unit. The series connection of the space
heating loads reduced the temperature of the return flow in the
district heating, which was critical to improving heat pump effi-
ciency, reducing pipe work losses, and minimising pipe sizing and
pumping energy. Crucially, the Energy Centre had a large stratified
thermal store that resulted in much greater flexibility and efficiency
of the renewable heating plant operation. The roofs of the homes
and flats were covered with solar PV tiles (63 kWp in total) that
provided renewable electricity to achieve net zero carbon emis-
sions in each of the homes [36]. Excess electricity was sold back to
the National Grid.
4.3.5.4. Building's energy performance. The results of the design
stage Standard Assessment Procedure (SAP) indicated that the
Dwelling Emission Rates (DER) of the 10 homes in Greenwatt Way
ranged from 19.71kgCO2/m2 to 58.97kgCO2/m2 [36]. However,
no energy use data were available for the case study.
4.4. Cross-case analysis of the exemplars
All exemplars of the ‘zero carbon buildings’ under the five
clusters have adopted a wide range of strategies and technologies,
for achieving both buildings' energy efficiency and energy gener-
ation and supply (Table 7). Despite the diversity of the specific
technologies adopted, the buildings' energy efficiency strategies
together addressed building envelope, ventilation, heating, cooling
and lighting and the energy generation and supply strategies
together included predominantly PV and also other technologies
such as solar thermal, wind, biofuel, geothermal, combined heat
and/or power district heating etc. These results demonstrate the
systems integration of design strategies that include passive design,
energy efficiency, and renewable energy technologies.
The cross-case analyses of the buildings' energy consumption
and energy generation show that ‘net-zero energy’ was not ach-
ieved in four out of the five exemplar cases while no energy use
data were available for the other one, although all cases were very
low or nearly net-zero energy. However, ‘net-zero carbon’ was
achieved for two out of the five cases due to the use of carbon
conversion factors for different types of energy source. Given the
99
typical low- or medium-rise feature of the buildings studied, it can
be asserted that it would be technically difficult to achieve the ‘net-
zero energy’ or ‘net-zero carbon’ over the one-year period of
operation. This finding casts a shadow over the feasibility of
achieving ‘net-zero carbon’ for high-rise in particular due to their
higher energy demand and geographical constraints to the use of
renewable technologies.
The cross-case analysis also reveals problems with the reporting
of buildings' energy use and carbon emissions. Some building cases
were reported with estimated energy use or carbon emissions but
not measured performance. Most did not specify the scope of the
energy use or carbon emissions, hence with no breakdown energy
and carbon data for lighting, ventilation, heating/cooling, hot water,
etc. Also, the underlying methodologies for calculating the build-
ings' energy use and carbon emissions varied from one case to
another.
5. Discussion
The results of the clusters of the buildings towards zero carbon
indicate the dominance of new-build low-rise residential buildings
in temperate zones. This dominance can be explained by several
factors. First, the bulk of research on ‘zero carbon building’ in
literature is focused on low-rise residential buildings [39,40]. There
exist general perceptions that it is infeasible to achieve ‘zero car-
bon’ or ‘zero energy’ for high-rise buildings due to their high energy
demand and limited renewable energy technology [41]. Second,
government policies on ‘zero carbon’ or ‘zero energy’ appear to
prioritize the focus on new-build homes in the residential sector.
For examples, the UK policy targets to achieve zero carbon new-
build homes from 2016 [42] followed by zero carbon non-
domestic buildings from 2019 [43]; the US policy in California
specifies that all new residential construction in California need to
achieve zero net energy by 2020, followed by all new commercial
construction in California to achieve zero net energy by 2030 [44].
These results verify the knowledge gap pointed out in previous
research [45] of retrofitting the existing residential building stock
that was found to be more effective on stabilizing the carbon
emissions trend.
The results of the paucity of high-rise cases substantiate the
perceived gap in the knowledge of high-rise buildings towards zero
carbon. The design and delivery of high-rise zero carbon buildings
in dense urban environments are still largely unknown [46]. Also,
many of the ‘zero carbon buildings’ were delivered as signature
projects to showcase state-of-the-art green design and technolo-
gies to the construction industry and raise community awareness of
sustainable living. Because of the very different building features, to
transfer the knowledge and learning from the showcases to the
mainstream practice of high-rise buildings faces tremendous
challenges in technical, commercial and socio-cultural aspects.
The results of examining the strategies for achieving zero carbon
show the systems integration of the design strategies that include
passive design, energy efficiency, and renewable energy technolo-
gies. Such systems approach has been widely recognized and
adopted for low or zero carbon buildings [7,47,48]. However, no use
of carbon offsetting measures was identified yet, e.g. tree planting
and carbon capture and storage technology. The results of investi-
gating the energy performance of exemplar cases suggest that net-
zero carbon or energy is technically difficult to achieve over the
one-year period of operation, which casts a shadow over the
findings of previous research [49] that zero carbon homes are
technically feasible in the long term. These results particularly
highlight the challenges for achieving zero carbon of high-rise
buildings due to the geographic constraints for use of renewable
energies [50]. The exploration of emerging technologies should
100 W. Pan, K. Li / Building and Environment 104 (2016) 92e101

Table 7
Strategies for achieving ‘zero carbon’ of the exemplars.

Strategy Element Building cases

Eco Terra Pixel Green Lighthouse Elithis Tower Greenwatt way

Energy Envelope
Insulated envelope
Double-glazed window
Compact form
Compact form
Insulated envelope
efficiency
Triple-glazed window
Sun blinds
Insulated envelope
Sun blinds
Thermal store

Slab storage
Seasonal storage
Increased glass surface
Ventilation
HRV
Night natural ventilation
Natural ventilation
Triple flow
MVHR

MVHR ventilation system
Heating
BIPV/T e
Passive
Solar and internal
ASHP

Passive solar heat heat gains
GSHP
solar heating
Cooling e
Absorption chiller e
Chilled beams e
Lighting e
Task lighting
Daylight
Task lighting
Daylight

LED lighting
Task lighting
Efficient appliances

Daylight
LED lighting
Energy Solar
BIPV/T
PV
PV
PV
PV
generation Wind e
Wind turbines e e e
& supply Biofuel e
Anaerobic digester tank system e
Wood boiler
Biomass boiler
Geothermal
GSHP e e e e
CHP e e e e e
Others e e
District heating e
District heating

therefore be critical to the achievement of net zero carbon.
It is worth noting that the analysis of buildings' energy use and
carbon emissions was only carried out of the five exemplars, but not
of all of the identified buildings included in the cluster analysis. This
was mainly because of the insufficient information on the build-
ings' energy use or carbon emissions and inconsistent definitions
and methodologies for calculating the buildings' energy use and
carbon emissions. Even for the five exemplars analyzed, some did
not provide estimated and/or measured energy use or carbon
emissions; most did not specify the scope of the energy use or
carbon emissions, hence with no breakdown energy and carbon
data for lighting, ventilation, heating/cooling, hot water, etc. These
results support the finding of previous research [21] that the un-
derlying methodologies for calculating the buildings' energy use
and carbon emissions varied from one country or region to another.
Also, the policy targets for the ‘zero carbon’ standards changed in
recent years even in the same countries/regions, e.g. the evolution
of the energy performance requirements towards net zero energy
buildings levels in Denmark [51], and the changes of the zero car-
bon building policies and targets in the UK [2]. These factors
together made the benchmarking and cross comparison of the
energy use and carbon emissions of buildings worldwide very
difficult. Pan and Ning [2] suggested that for obtaining and veri-
fying data on buildings' energy use and carbon emissions for
effective comparison, the unit of balance, period of balance, energy
scope and any connection with the grid should be explicitly iden-
tified. Future research should develop methodology for effective
benchmarking of buildings' energy performance towards net zero.
6. Conclusions
This paper has revealed the clusters of buildings towards zero
carbon that have emerged and been reported worldwide. The paper
has also examined the strategies for achieving ‘zero carbon’ of five
exemplar building cases that represent the five revealed clusters,
and investigated their energy performance. The paper concludes
that low-rise new-build residential buildings in temperate zones
dominated the buildings towards zero carbon. Five clusters of
buildings towards zero carbon were revealed, which were: (1)
private residential low-rise rural ‘zero carbon buildings’ in cold
zone; (2) private commercial low-rise ‘zero carbon buildings’ in
temperate zone; (3) public low-rise ‘zero carbon buildings’ in
temperate zone; (4) private urban ‘zero carbon buildings’ in
temperate zone; and (5) private residential low-rise suburban ‘zero
carbon buildings’ in temperate zone. The gaps of high-rise cases
and the building stock were identified in the body of knowledge of
building towards zero carbon.
The paper also concludes that systems integration of design
strategies is needed for achieving ‘zero carbon’, which should
include passive design, energy efficiency, and renewable energy
technologies. However, net-zero carbon, and particularly net-zero
energy, was found to be technically difficult to achieve over the
one-year period of operation, and particularly difficult for high-rise,
of which the feasibility warrants future research. The use of
emerging technologies should therefore be explored for possibly
closing the knowledge gap in delivering high-rise buildings to-
wards ‘zero carbon’. The findings should help to understand the
complex profiles of the ‘zero carbon buildings’ and support effec-
tive cross-context learning of the practices. Such learning should
take into account the different characteristics of buildings against
relevant variables. A challenging but important area of future effort
will be to collect and compare buildings' energy performance and
carbon emissions in a wide context, not only as designed but as
measured, with details of their scope. Future research should also
develop consistent definitions and calculation methodologies of
‘zero carbon’ to enable effective benchmarking of buildings' energy
use and carbon reduction practices.
Acknowledgements
We would like to acknowledge research funding by the Public
Policy Research Funding Scheme (Project Number:
2014.A8.020.14E) from the Central Policy Unit of the HKSAR Gov-
ernment, and by the Construction Industry Council (Project Num-
ber: CICR/01/13). Its contents are solely the responsibility of the
authors and do not necessarily represent the official views of the
funding bodies. We would also like to acknowledge Concordia
University, University of Copenhagen and Zero Carbon Hub for ac-
cess to information on relevant case buildings.
References
[1] A. Koch, S. Girard, K. McKoen, Towards a neighbourhood scale for low-or zero-
carbon building projects, Build. Res. Inf. 40 (2012) 527e537.
[2] W. Pan, Y. Ning, A socio-technical framework of zero-carbon building policies,
Build. Res. Inf. 43 (2015) 94e110.
[3] M. Panagiotidou, R.J. Fuller, Progress in ZEBsda review of definitions, policies
and construction activity, Energy Policy 62 (2013) 196e206.
[4] Musall E. Available at: http://www.enob.info/en/net-zero-energy-buildings/
map/(accessed 07/03/2015, 2015).
 W. Pan, K. Li / Building and Environment 104 (2016) 92e101
[5] ZCH. Available at: http://www.zerocarbonhub.org/building-profiles (accessed
07/03/2015, 2015).
[6] J. Zuo, B. Read, S. Pullen, Q. Shi, Carbon-neutral commercial building devel-
opment, J. Manag. Eng. 29 (2012) 95e102.
[7] W. Pan, System boundaries of zero carbon buildings, Renew. Sustain. Energy
Rev. 37 (2014) 424e434.
[8] P. de Wilde, The gap between predicted and measured energy performance of
buildings: a framework for investigation, Autom. Constr. 41 (2014) 40e49.
[9] ZCH. Closing the Gap between Design and As-built Performance. London: Zero
Carb Hub.
[10] G.C. da Graça, A. Augusto, M.M. Lerer, Solar powered net zero energy houses
for southern Europe: feasibility study, Sol. Energy 86 (2012) 634e646.
[11] DCLG, Code for Sustainable Homes: a Step-change in Sustainable Home
Building Practice, Department of Communities and Local Government, Lon-
don, UK, 2006.
[12] Commission E. Directive 2010/31/EU of the European Parliament and of the
council, Off. J. Eur. Union (2010) 13e34.
[13] C. Riedy, A. Lederwasch, N. Ison, Defining Zero Emission Buildingsdreview
and Recommendations: Final Report, Institute for Sustainable Futures, 2011.
[14] ASBEC. Net zero emission homes: an industry roadmap. Available at: http://
asbec.asn.au (accessed 07/03/2015, 2015).
[15] Act EIaS, CRS Report for Congress: Energy Independence and Security Act of
2007, US Government, Washington, USA, 2007.
[16] Obama B. Federal leadership in environmental, energy, and economic per-
formance. Executive Order (13514) of October. 2009; 5.
[17] H. Erhorn, H. Erhorn-Kluttig, Terms and Definitions for High Performance
Buildings, Concerted Action Energy Performance of Buildings, European
Union, 2011.
[18] LIH. Available at: http://sealevel.ca/lowimpact/index.htm (accessed 07/03/
2015, 2015).
[19] C. Fraley, A.E. Raftery, How many clusters? which clustering method? answers
via model-based cluster analysis, Comput. J. 41 (1998) 578e588.
[20] P.J. Rousseeuw, Silhouettes: a graphical aid to the interpretation and valida-
tion of cluster analysis, J. Comput. Appl. Math. 20 (1987) 53e65.
[21] A.J. Marszal, P. Heiselberg, J. Bourrelle, E. Musall, K. Voss, I. Sartori, et al., Zero
Energy BuildingeA review of definitions and calculation methodologies, En-
ergy Build. 43 (2011) 971e979.
[22] P. Torcellini, S. Pless, M. Deru, D. Crawley, Zero energy buildings: a critical look
at the definition, in: ACEEE Summer Study Pacific Grove, California, August
14e18, 2006.
[23] I. Sartori, A. Napolitano, K. Voss, Net zero energy buildings: a consistent
definition framework, Energy Build. 48 (2012) 220e232.
[24] M.C. Peel, B.L. Finlayson, T.A. McMahon, Updated world map of the Ko € ppen-
Geiger climate classification, Hydrol. Earth Syst. Sci. Discuss. 4 (2007)
439e473.
[25] S. Bucking, A. Athienitis, R. Zmeureanu, W. O'Brien, M. Doiron, Design opti-
mization methodology for a near net zero energy demonstration home, in:
Proceedings of EuroSun International Conference on Solar Heating, Cooling
and Buildings, Graz, Austria, Sept 28 e Oct 1, 2010.
[26] M. Doiron, P.E.W. O'Brien, A. Athienitis, P. Eng, Energy performance, comfort
and lessons learned from a near net-zero energy solar house, ASHRAE Trans.
117 (2011) 1e12.
[27] M. Noguchi, A. Athienitis, V. Delisle, J. Ayoub, B. Berneche, Net Zero Energy
Homes of the Future: a Case Study of the EcoTerra™ House in Canada,
Renewable Energy Congress, Glasgow, Scotland, 2008.
[28] J. Zuo, B. Read, S. Pullen, Q. Shi, Achieving carbon neutrality in commercial
building developmentseperceptions of the construction industry, Habitat Int.
101
36 (2012) 278e286.
[29] Sustainable Victoria. Carbon neutral offices- the Pixel building case study.
Available at: http://www.sustainability.vic.gov.au/(accessed 07/03/2015).
[30] K. Voss, E. Musall, Net Zero Energy Buildings: International Projects of Carbon
Neutrality in Buildings, Walter de Gruyter, 2013.
[31] COWI. Green lighthouse - sustainable building. Available at: http://www.cowi.
com/menu/project/Buildings/Sustainableandgreenbuildings/Pages/
Greenlighthousenewecofriendlybuilding.aspx (accessed 07/03/2015).
[32] VELUX. Sustainable living demo buildings-Green Lighthouse. Available at:
http://www.velux.com/sustainable_living/demonstration_buildings/green_
lighthouse (accessed 07/03/2015).
[33] J. Kurnitski, M. Achermann, J. Gra €slund, O. Hernandez, W. Zeiler, nZEB Case
Studies. Cost Optimal and Nearly Zero-energy Buildings (nZEB), Springer,
2013, pp. 135e176.
[34] A. Lenoir, F. Garde, E. Ottenwelter, A. Bornarel, E. Wurtz, Net zero energy
building in France: from design studies to energy monitoring. A state of the
art review, in: International Conference on Solar Heating, Cooling and
Buildings. Graz, 2010.
[35] H.Y.C.H.L. Baofeng, Strategies for building energy conservation of Elithis
Tower, New Archit. (2010) 2e020.
[36] SSE. Case study-Green Wattway. Available at: http://www.swalec.co.uk/
AboutUs/GreenwattWay/(accessed 07/03/2015).
[37] D. Saker, M. Vahdati, P. Coker, S. Millward, Assessing the benefits of domestic
hot fill washing appliances, Energy Build. 93 (2015) 282e294.
[38] H. Tol, S. Svendsen, Operational planning of low-energy district heating sys-
tems connected to existing buildings, in: 2nd International Conference on
Renewable Energy, Nagasaki, 2012.
[39] M. Bojic, N. Nikolic, D. Nikoli
c, J. Skerli
c, I. Mileti
c, Toward a positive-net-
energy residential building in Serbian conditions, Appl. Energy 88 (2011)
2407e2419.
[40] M. Leckner, R. Zmeureanu, Life cycle cost and energy analysis of a net zero
energy house with solar combisystem, Appl. Energy 88 (2011) 232e241.
[41] K. Fong, C.K. Lee, Towards net zero energy design for low-rise residential
buildings in subtropical Hong Kong, Appl. Energy 93 (2012) 686e694.
[42] DCLG, Build a Greener Future: Policy Statement, Department of Communities
and Local Government, London, UK, 2007.
[43] H.M. Treasury, Budget 2008 Stability and Opportunity: Building a Strong,
Sustainable Future, The Stationery Office, London, UK, 2008.
[44] Commission CPU, California Long Term Energy Efficiency Strategic Plan:
Achieving Maximum Energy Savings in California for 2009 and beyond: Cal-
ifornia Public Utilities Commission, 2008.
[45] N.C. Onat, G. Egilmez, O. Tatari, Towards greening the U.S. residential building
stock: a system dynamics approach, Build. Environ. 78 (2014) 68e80.
[46] W. Pan, Y. Ning, Delivering zero carbon buildings: the status quo and way
forward, Zero Carbon Build. J. 1 (1) (2014) 7e14.
[47] Y. Xing, N. Hewitt, P. Griffiths, Zero carbon buildings refurbishmenteea Hi-
erarchical pathway, Renew. Sustain. Energy Rev. 15 (2011) 3229e3236.
[48] R. Cole, L. Fedoruk, Shifting from net-zero to net-positive energy buildings,
Build. Res. Inf. 43 (1) (2015) 111e120.
[49] M. Osmani, A. O'Reilly, Feasibility of zero carbon homes in England by 2016: a
house builder's perspective, Build. Environ. 44 (2009) 1917e1924.
[50] W. Pan, H. Qin, Y. Zhao, Challenges for energy and carbon modeling of high-
rise buildings: the case of public housing in Hong Kong, Resour. Conserv.
Recycl. (2016). http://dx.doi.org/10.1016/j.resconrec.2016.02.013.
[51] H.-K.H. Erhorn, H. Erhorn-Kluttig, The Path towards 2020: nearly zero-energy
buildings, REHVA J. (2012) 20.