![CO2 emissions track (billion tonnes per year) over the past 170 years (right vertical axis), comparison between observed and simulated earth surface temperature (left vertical axis) [6].](https://www.researchgate.net/profile/Mohsen-Sharifi-4/publication/362520071/figure/fig1/AS:1186036256374786@1659784487138/1-CO2-emissions-track-billion-tonnes-per-year-over-the-past-170-years-right-vertical_Q320.jpg)
![Key polices and milestones on the road to net zero emissions [6].](https://www.researchgate.net/profile/Mohsen-Sharifi-4/publication/362520071/figure/fig2/AS:1186036256374787@1659784487172/2-Key-polices-and-milestones-on-the-road-to-net-zero-emissions-6_Q320.jpg)
![The building sector's share of global annual energy use and carbon emissions in 2019 (left) [3] and share of different sub-sector from the energy use in the building sector in 2020 [10].](https://www.researchgate.net/profile/Mohsen-Sharifi-4/publication/362520071/figure/fig3/AS:1186036256391168@1659784487210/3-The-building-sectors-share-of-global-annual-energy-use-and-carbon-emissions-in-2019_Q320.jpg)
Content uploaded by Mohsen Sharifi
Author content
All content in this area was uploaded by Mohsen Sharifi on Jan 10, 2023
Content may be subject to copyright.
Early-Stage Integrated Design Methods for Hybrid GEOTABS Buildings
Mohsen Sharifi
Doctoral dissertation submitted to obtain the academic degree of
Doctor of Engineering
Prof. Jelle Laverge, PhD - Eline Himpe, PhD
Department of Architecture and Urban Planning
Faculty of Engineering and Architecture, Ghent University
Supervisors
July 2022
Wettelijk depot: D/2022/10.500/56
NUR 955, 978
ISBN 978-94-6355-615-6
Members of the Examination Board
Chair
Prof. Filip De Turck, PhD, Ghent University
Other members entitled to vote
Prof. Lieve Helsen, PhD, KU Leuven
Prof. Arnold Janssens, PhD, Ghent University
Prof. Doreen E. Kalz, PhD, Berliner Hochschule für Technik, Germany
Prof. Ongun Kazanci, PhD, Danmarks Tekniske Universitet, Denmark
Prof. Steven Lecompte, PhD, Ghent University
Supervisors
Prof. Jelle Laverge, PhD, Ghent University
Eline Himpe, PhD, Ghent University
Acknowledgement 7
On January 5th, 2022, when I was assisting Jelle in supervising his exam,
he approached me in the exam room and told me that the university has
"strongly recommended" the use of a specific face mask and handed me one
of them. I replaced my regular medical face mask with the mask provided
by the university. I am not sure how many layers that mask had, but there
seemed to be no current of air passing through it, and I felt like suffocating.
Unlike regular masks, however, no fog appeared on my glasses. I told Jelle
about my experience with the trade-off between the suffocation sensation
and the improved vision. Jelle confirmed my observation, saying, "well,
they say life is a trade-off between what you lose and what you gain".
Having a gentile smile when confronting my professor’s irony, I was
contemplating the extent to which I had control over trade-offs of my life.
I never undermined the role of “random turn of events” (smart people
simply call it “chance”) in my life. I never denied my chance in having
people who truly assisted me in my journey to this point.
I was working in R&D department in a company that once had an enviable
reputation in Iran's energy industry. After a while, the company entered a
drastic financial crisis and R&D activities were the first to suffer. I realized
that despite my efforts, my contribution to the company was negligible. I
had to leave the company where I had feelings for after five years of learning
and experiencing. I started to look for a new job and a destination, Ghent,
called my name.
I entered a H2020 project named hybridGEOTABS with numerous
experienced partners. I was rapidly accepted by my colleagues in the project
with a warm hug. Although I was one of the last people added to the
project, my colleagues Liz, Anne, Ongun, Héctor, Fernando, Qian, Iago,
Filip, Damien, Jan, and Pascal made me feel I had been always working
with them. During the project, I was excited to be part of a group led by
famous names of the field, Lieve, Wim, and Bjarne. I was delighted by the
fact that the hybridGEOTABS project gathered the most reputable names
and I could communicate with them. I would like to thank all my colleagues
in hybridGEOTABS project for their warm welcome, direct critics, and
insightful comments. There would be no PhD thesis about
hybridGEOTABS if there were no hybridGEOTABS project.
Building physics group (BFG) has furnished a friendly environment with
the minimum marginal issues. I would like to thank Arnold for creating
8 Acknowledgement
and leading such a group. I would like to thank all my colleagues in BFG
group for a desirable working place. When I entered the group, there were
around 10 PhD researchers and 3 post-doctoral researchers. The group has
almost tripled during the last four years and is rapidly growing with
multiple national and multinational projects in a variety of topics related
to the built environment.
I would like to thank my friends Begüm, Kata, Jota, and Siavash who were
my family in Ghent. With their warm support, I could get through the
coldest days of Gent. Without them, it would be impossible to continue
throughout the days of quarantine. We shared happiness, joy, laugh, and
tears. We, in our small group, celebrated weddings and mourned each
other’s loss during the past four years. We endured homesickness together.
We got mad at each other, worried about each other, and missed each other
when traveling home. We are now a small family of people from four
different countries. When I left my friends in Iran, with whom I had a 20-
year friendship, I thought I would never be able to make such a friendship
again. But I did not know that when a door closes, a new window, and
such a bright window, might open.
From thousands of kilometers away, you can check your supervisors’
scientific records to get insight about their scientific background. But what
can you know about your supervisors’ attitudes? I could not anticipate my
supervisors will be kind, supportive, and professional. I could not foresee
they will stand by me in my very personal unexpected issues. I would like
to thank my supervisors, Jelle and Eline. They trusted me and gave me the
freedom to choose my way and then supported my choice. They always
allowed me to experience, try, and even fail till I find an appropriate way.
It is only through this stumbling that you can learn. Believe me; you will
not learn anything if your supervisors cannot bear your failure.
Words cannot express how grateful I am to my wife, my father, and my
sister. Zahra, my wife, has been always the greatest support for me. In the
hardest days of my life, she was always there for me, and it is only because
of her that I could overcome my problems. My father invested on me with
his soul. He is my hero, even though that sounds like an old cliché for the
feelings between fathers and sons. He neglected his own desires to make
opportunities for me. And my sister, Mahsa, is the reason I feel that I still
have a family after my mother left us for an endless travel. Also, I would
like to thank my wife's family for their unwavering support over the last
ten years.
Acknowledgement 9
Finally, I would like to thank the jury of my PhD dissertation. They
invested substantial time in review process and provided insightful
comments from different perspectives. I am happy that I could improve the
book by fruitful discussions with the jury and by using their comments
about the text.
With the words above I tried to state the impact of those around me on
this book. I have been encouraged, guided, assisted, and disappointed by
those around me and I synthesized all the inputs to make this dissertation.
I believe that the only way I could effectively appreciate all the time and
efforts people around me invested on me is to make the investment fruitful.
I hope they accept this book as a token of my gratitude to their support.
March 8th, 2022
Ghent, Belgium
Mohsen Sharifi
10 Table of Contents
Acknowledgement ...................................................................... 7
Table of Contents .................................................................... 10
Nomenclature .......................................................................... 14
Summary ................................................................................. 18
Samenvatting ........................................................................... 22
List of Tables ........................................................................... 26
List of Figures ......................................................................... 27
.................................................................... 32
1.1 Introduction ......................................................................... 32
1.2 Context ................................................................................ 34
1.2.1 The so-called “trigger point” is approaching fast ................ 34
1.2.2 Carbon emission level and global warming are definitely
related .................................................................................. 35
1.2.3 Sustainable technologies can reduce carbon emissions ......... 37
1.3 Sustainable heating and cooling technologies ...................... 38
1.3.1 Sustainable emission systems ............................................... 40
1.3.2 Sustainable production systems ........................................... 43
1.3.3 Renewable energy source coupled to a storage-integrated
radiant emission heating and cooling system ...................... 46
1.3.4 What is hybridGEOTABS? ................................................. 47
1.4 Challenges for hybridGEOTABS design ............................. 48
1.4.1 So far so good! What is the problem? .................................. 48
1.4.2 Early-stage design ................................................................ 57
1.5 This PhD thesis ................................................................... 59
1.6 Organization of the book ..................................................... 62
.................. 66
2.1 Introduction ......................................................................... 66
2.2 Building energy performance modeling ............................... 67
Table of Contents 11
2.3 Methodology .........................................................................70
2.3.1 Model architecture ................................................................71
2.3.2 Parameter value estimation methodology .............................75
2.3.3 Implementation and verification ...........................................81
2.4 Results ..................................................................................84
2.4.1 Parameter values and interpretation ....................................84
2.4.2 High-level indicators to assess the modeling accuracy ..........85
2.4.3 Analysis of the residuals .......................................................87
2.4.4 Indoor temperature ...............................................................89
2.5 Discussion .............................................................................92
2.6 Conclusion ............................................................................95
........................... 96
3.1 Introduction .........................................................................96
3.2 Optimization problem ..........................................................98
3.3 Optimization algorithm ........................................................99
3.3.1 Initial guess ......................................................................... 101
3.3.2 Constraints .......................................................................... 102
3.3.3 Search algorithm ................................................................. 104
3.4 Case studies and baseline scenario ..................................... 104
3.5 Results ................................................................................ 106
3.5.1 Objective function value ..................................................... 106
3.5.2 Objective function track ..................................................... 107
3.5.3 Optimal load split ............................................................... 108
3.5.4 Time series comparison ....................................................... 110
3.5.5 Calculation time .................................................................. 113
3.5.6 Prediction horizon N ........................................................... 114
3.6 Discussion ........................................................................... 116
3.7 Conclusions ......................................................................... 118
...120
12 Table of Contents
4.1 Introduction ........................................................................120
4.2 Simulation-based design .....................................................121
4.3 Methodology .......................................................................124
4.3.1 Step 1: Building modeling and load calculation ..................125
4.3.2 Step 2: Calculating the Hourly Time Series of the GSHP and
Secondary System Loads ....................................................126
4.3.3 Step 3: Calculating Annual Loads and Sizing Key Components
...........................................................................................129
4.3.4 Step 4: Sizing of the Borefield .............................................130
4.3.5 Cost-benefit analysis ...........................................................133
4.4 Case studies ........................................................................134
4.5 Results ................................................................................137
4.5.1 Optimal solutions ................................................................138
4.5.2 Cost Breakdown and Optimal Load Split ...........................140
4.5.3 Annual cashflow and payback period .................................142
4.5.4 Sensitivity of the payback time to the Electricity and Gas
Price ...................................................................................144
4.5.5 Optimal sizing .....................................................................148
4.5.6 CO2 Emission sensitivity analysis .......................................156
4.6 Discussion ...........................................................................158
4.7 Conclusion ..........................................................................160
....... 163
5.1 Introduction ........................................................................163
5.2 Necessity of post-processing ...............................................164
5.3 Post-processing methodology ..............................................166
5.3.1 The influence of smoothing on the sizing and optimal load split
...........................................................................................167
5.3.2 From load splitting to sizing: borefield thermal balance ....171
5.3.3 Wrap-up of the sizing steps ................................................173
5.4 Performance estimation ......................................................176
Table of Contents 13
5.5 Design procedure verification ............................................. 176
5.6 Discussions ......................................................................... 181
5.7 Conclusions ......................................................................... 182
............184
6.1 Introduction ....................................................................... 184
6.2 Informed decision making with decision trees .................... 185
6.3 Data analysis ...................................................................... 187
6.4 Exploratory data analysis and preliminary results ............ 188
6.4.1 GEOTABS share ................................................................ 188
6.4.2 Sizing .................................................................................. 190
6.5 Methodology to derive decision trees ................................. 192
6.6 Application of design decision trees ................................... 195
6.7 Discussion ........................................................................... 201
6.8 Conclusion .......................................................................... 203
....................................................................205
7.1 Conclusions ......................................................................... 205
7.2 Limitations ......................................................................... 209
7.3 Perspectives ........................................................................ 210
List of publications .................................................................213
Annex A .................................................................................216
Annex B .................................................................................217
Annex C .................................................................................218
Annex D .................................................................................219
Annex E .................................................................................225
Annex F .................................................................................227
Annex G.................................................................................228
Annex H .................................................................................232
References ..............................................................................234
14 Nomenclature
Abbreviations
ADR
Active demand response
AHU
Air handling unit
AMLR
Adaptive multiple linear regression
ASHP
Air source heat pump
ASHRAE
American society of heating, refrigerating and air-
conditioning engineers
BES
Building energy simulation
BGI
Building grid integration
CART
Classification and regression trees
CBA
Cost benefit analysis
CCA
Concrete core activation
COP
Coefficient of performance
CTREE
Conditional inference trees
EED
Earth energy designer
EER
Energy efficiency ratio
ESS
Embedded surface system
FCU
Fan coil unit
FFT
Fast Fourier transformation
GA
Genetic algorithm
GEOTABS
Geothermal heat pump coupled to TABS
GSHP
Ground source heat pump
GSHX
Ground source heat exchanger
HP
Heat pump
HVAC
Heating, ventilation and air conditioning
hybridGEOTAB
S
GEOTABS coupled with secondary systems
LDC
Load duration curve
linprog
Linear programming solver in MATLAB
LOWESS
Locally weighted scatterplot smoothing
LQR
Linear quadratic regulator
LTI
Linear time-invariant
MPC
Model predictive control
NPC
Net present cost
NRMSD
Normalized root mean square deviation
nZEB
Nearly-zero energy building
Nomenclature 15
OLSA
Optimal load splitting algorithm
PCHX
Passive cooling heat exchanger
PCM
Phase-change materials
PV
Photovoltaic
PWM
Pulse width modulation
R²ES
Renewable and residual energy sources
RBC
Rule-based control
RC
Resistor-capacitor
RCP
Radiant ceiling panels
RES
Renewable energy sources
RHC
Radiant heating and cooling
SSE
State-space equations
TABS
Thermally activated building systems
TACO
Toolchain for automated control and optimization
TEA
Techno-economic assessment
UBB
Unknown but bound
Parameters
The primary system power injected to the TABS
at time step k (kW)
The secondary system power injected to the zone
at time step k (kW)
Power of TABS in the building side at time step k
(kW)
Building thermal loads at time step k (kW)
calculated with ideal heating and cooling system
with fixed set point (23 °C)
Building thermal loads at time step k (kW)
calculated with ideal heating and cooling system
with 22 °C and 24 °C set points for heating and
cooling, respectively
Vector of the upper bound of the model states
constraints (°C)
16 Nomenclature
Vector of the lower bound of the model states
constraints (°C)
Vector of the upper bound of the control variables
constraints (W or kW)
Vector of the lower bound of the control variables
constraints (W or kW)
A
Area (m2)
c
Specific thermal capacity (J/kg/K)
D
Distance between pipes inside TABS (mm)
d
Diameter of pipes inside TABS (mm)
hrad+conv
Overall heat transfer coefficient for radiation and
convention between the TABS surface and the
room (W/m2/K)
L
Thickness of the concrete layer of TABS (mm)
N
Prediction horizon (hours)
objFun
Specific objective function value for one year
converted to (kWh/m2/year)
T
Vectors of the model states (°C)
u
Vector of the control inputs in state-space model
(W or kW)
V
Volume (m3)
Effective borehole resistance (K/W)
Hourly effective thermal resistance of the ground
estimated with a g-function for the maximum
hourly load on the borefield (K/W)
Monthly effective thermal resistance of the ground
estimated with a g-function for the maximum
monthly load (K/W)
Yearly effective thermal resistance of the ground
estimated with a g-function for the yearly load
(K/W)
Nomenclature 17
Thermal conductivity of the ground (W/m/K)
Maximum hourly load on the borefield (kW)
Maximum monthly load on the borefield (kW)
Yearly imbalance load on the borefield (kW)
Greek letters
Diffusivity of the ground (m2/s)
Density (kg/m3)
λ
Thermal conductivity (W/m/K)
Overall efficiency of a system
18 Summary
The building sector is responsible for nearly 36% of total CO2 emissions in
Europe. According to the European Commission's roadmap for achieving a
low-carbon building sector, policymakers must seek a practical plan for
decarbonizing energy supplies and reducing final energy use in buildings.
Future-proof buildings are energy efficient and energy flexible, allowing for
maximum integration of renewable and residual energy sources (R2ES). As
a result, heat pumps—as the devices that convert electricity to heat—have
a pivotal role in the electrification of heating, ventilation, and air
conditioning (HVAC) systems. However, renewable electricity production
may be curtailed. Energy-flexible buildings allow the demand to be shifted
to periods of abundance of R2ES by using smart building system controls.
GEOTABS, the combination of a GEOthermal heat pump and a thermally
activated building system (TABS), is an example of a future-proof HVAC
concept. TABS is a radiant heating and cooling emission system in which
the heating/cooling pipes are embedded in the mass of the building
elements (for example, concrete floors), activating them as thermal storage;
this offers energy flexibility. TABS can provide very low-temperature
heating (as low as 23-28 °C) and high-temperature cooling (as high as 17-
23 °C) by converting entire floors or ceilings surfaces into emission systems.
Geothermal heat pumps operate efficiently at these temperatures, which
are found at shallow depths in the ground. The hybridGEOTABS system
is centered on GEOTABS, which is supplemented by secondary heating
and/or cooling emission and production systems to maintain thermal
comfort in every building. Although this hybridity expands the potential
application of the concept, it also leads to the first question every
hybridGEOTABS designer faces: how to split the load between GEOTABS
and the secondary systems. A discussion of this question is followed by
financial and environmental analyses of the design with a high degree of
freedom, resulting in decision-making difficulties.
Chapter 1 points to the ambiguity in the design of such a future-proof
HVAC concept. The dilemma is between a detailed and accurate but
prohibitive design procedure, and a traditionally simplified but inaccurate
design procedure. It is elaborated that the thermal inertia of TABS as the
source of energy-flexibility of hybridGEOTABS is neglected in simplified
design procedures. The thermal inertia of TABS can allow downsizing of
the production unit and can shift the building load, if it is optimally
Summary 19
exploited. TABS causes a time delay between the production unit’s power
and the building’s thermal loads. Moreover, heat transfer regime of TABS
consists of advection, conduction, convection, and radiation. The time
delay and the complex heat transfer regime of TABS must be integrated
into the design procedure by means of dynamic modeling and simulation
of the system in order to achieve a robust design. However, optimal design
of TABS is an intricate mathematical problem, obstructing the early-stage
design. To overcome this dilemma, a simplified design procedure that
optimally incorporates the dynamic behavior of TABS is required. To
achieve such a design procedure, simplifications in modeling and
optimization are required to keep the mathematical complexity and the
calculation time tractable. Then, the dynamic simulations are run for
140,000 case study buildings from an accessible database in order to offer
pre-engineering results to designers. Designers can therefore simply consult
a near-optimum design associated with a given building with limited inputs
at the early-stage of the design. The simulation results are also used to
obtain early-stage integrated design decision trees for architectural design
informed decision making. Through this simple and early-stage design
procedure, the back-and-forth relation between architectural and HVAC
designs and so an integrated design method can be achieved. As such, the
book chapters were organized with specific research targets as follows:
In Chapter 2, an automated method for generating a simplified model of
the building and TABS is developed. Using the results of detailed
simulations, an approach is proposed which uses an inverse modeling
technique to generate a simplified grey-box model of the building and
TABS. The simplified model is then used in the design procedure to
incorporate the dynamic behavior of TABS and the building. The
novelty of this chapter is in the incorporation of the building dynamic
behavior in the design without using detailed mathematical models.
In Chapter 3, a heuristic algorithm is developed to create an automated
optimal load splitting algorithm (OLSA). OLSA employs the
simplified model developed in Chapter 2 to supply the building's thermal
demand optimally with TABS and (only if necessary) a secondary system.
Incorporating the optimal control of the TABS in the early design
estimations, is the novelty of this chapter. As such, OLSA provides time
series of the primary and secondary system power for the entire year,
allowing for optimal component sizing in combination with the post-
processing which is elaborated in the following chapters.
20 Summary
In Chapter 4, the OLSA is used in an integrated dynamic simulation-based
design procedure to provide a techno-economic assessment (TEA)
model for hybridGEOTABS. The design procedure is based on numerous
dynamic borefield and building energy simulations for each case study. This
chapter incorporates a novel algorithm into the TEA model to guarantee
that all the optimal solutions considering different competing objectives are
investigated. The TEA model is applied in nine case studies, and the
definition of the optimal design is studied through the results.
Furthermore, some general guidelines for optimal hybridGEOTABS
designs are derived from the results. The results demonstrate the
importance of the early-design stage, as it is observed that an inappropriate
building design can turn GEOTABS infeasible for a building (32% more
expensive than a conventional solution) while an appropriate design can
save 19% of total costs, considering the energy prices in the European
market in 2021. The appropriate design is achieved through a back-and-
forth design process in which the architect is aware of the consequences of
each architectural design decision, also called informed decision-making.
This back-and-forth relationship, which is missing from simulation-based
designs, entails a quick HVAC design and is elaborated in the next
chapters.
Using the OLSA simulation results (from Chapter 3) and the lessons
learned from Chapter 4, in Chapter 5 a computationally efficient design
procedure is developed and applied to merely 140,000 case studies from an
available database of building stock simulations for three European
climates (represented by Madrid, Brussels, and Warsaw). As a result, a
new database with a near-optimal hybridGEOTABS design for all the
buildings in the original database is provided. The designer can thus receive
pre-engineering results from the new database provided. The designer can
then look up the GEOTABS design by using only high-level building
parameters such as climatic zone, building function, insulation level, glazing
area as inputs. Furthermore, this chapter provides a rich database for the
data analysis performed in the next chapter.
Chapter 6 distinguishes between appropriate design tools for HVAC
designers and architects. The database provided in Chapter 5 is itself a
design tool for HVAC designers. However, in the earliest steps of the
building design, decision trees are more informative and practical to use. A
design decision tree enables architects to rapidly compare the impact of
their design decisions on the energy performance of a building. As a result,
they can alter the building design to improve environmental indicators by
Summary 21
increasing the share of renewable resources, even at the beginning of the
design, without running simulations. This, indeed, integrates
architectural and HVAC early-stage design. To derive these decision trees,
a previously validated and automated supervised machine learning
algorithm known as a classification technique is used for automated data
analysis. The application of the decision tree is exemplified in three case
studies.
This book demonstrates that the optimal design of hybridGEOTABS is
highly sensitive to various factors such as building functionality,
architectural design, economic indicators, available investment funds, and
location. Hence, the design process of an innovative and sustainable
technology such as hybridGEOTABS entails a synergy between different
parties, including building owners, architects, HVAC designers, energy
suppliers, and policy makers. This book demonstrated that early-stage
integrated design methods provide ample room for the different
stakeholders to search for simultaneously feasible, sustainable, and
financially appealing solutions. With early iterations between architectural
and HVAC design, pitfalls such as high investment costs, low operation
cost savings, and low CO2 emission savings which can be due to sub-optimal
design can be avoided. As such, informed and integrated decision-making
from the earliest stages is an essential part of the design procedure of
hybridGEOTABS. This book provided integrated design tools for
hybridGEOTABS to help designers and architects optimize their decisions
with the limited data that are accessible at the early stage of the project.
22 Samenvatting
De bouwsector is verantwoordelijk voor bijna 36% van de totale CO2-
uitstoot in Europa. De Europese Commissie stelde een roadmap naar een
koolstofarme samenleving voor en stelt daarin dat beleidsmakers een plan
moeten voorstellen om de energievoorziening koolstofarm te maken en het
eindverbruik van energie in de bouwsector te verminderen.
Toekomstbestendige gebouwen zijn energiezuinig en flexibel, en laten een
maximale integratie van hernieuwbare en residuele energiebronnen (R2ES)
toe. Warmtepompen, die op een zeer efficiënte manier elektriciteit
omzetten in warmte, spelen een cruciale rol bij de elektrificatie van
verwarmings-, koelings- en luchtbehandelingssystemen (HVAC), die op
haar beurt toelaat om hernieuwbare elektriciteit te benutten in gebouwen.
Door een gebrek aan hernieuwbare bronnen kan de productie van
hernieuwbare elektriciteit echter worden beperkt. Energieflexibele
gebouwen spelen hierop in door de vraag te verschuiven naar perioden van
overvloed aan R2ES, waardoor het gebruik van fossiele brandstoffen wordt
verminderd. Deze verschuiving vereist ook een slimme regeling van het
HVAC-systeem.
GEOTABS is, als combinatie van een GEOthermische warmtepomp en
thermisch geactiveerd gebouwsysteem (TABS)een voorbeeld van zo’n
toekomstbestendig HVAC-concept. TABS, ook bekend als
betonkernactivering, is een emissiesysteem voor verwarming en koeling
waarbij de warmte-overdracht in belangrijke mate door straling gebeurt.
De verwarmings-/koelleidingen zijn ingebed in de massa van de
bouwelementen (bijvoorbeeld betonnen vloeren), waardoor ze worden
geactiveerd als thermische opslag, wat energieflexibiliteit biedt. TABS kan
verwarming leveren bij zeer lage aanvoertemperatuur (tot 23-28°C) en
koeling bij hoge aanvoertemperatuur (tot 17-23°C) door volledige vloer- of
plafondoppervlakken om te zetten in emissiesystemen met een hoge
thermische massa. Geothermische warmtepompen werken efficiënt bij deze
temperaturen, die vrij dicht liggen bij de temperaturen in de ondiepe lagen
van de ondergrond. Het hybridGEOTABS-systeem maakt gebruik van
GEOTABS als duurzame kern van het HVAC-systeem, en wordt aangevuld
met secundaire verwarmings- en/of koelingsemissie- en/of -
productiesystemen. Deze laten toe om het thermisch comfort te garanderen
ook wanneer GEOTABS niet aan alle vraag kan of moet voldoen, en bieden
ook een hogere flexibiliteit (op vlak van energiebronnen,
investeringskosten) wat de inzetbaarheid van het systeem voor een grote
Samenvatting 23
range van gebouwen mogelijk maakt. Voor de HVAC-ontwerper, die al
vroeg in het ontwerpproces de haalbaarheid (op vlak van comfort,
energieprestastie, CO2-uitstoot en economie) van het hybridGEOTABS
systeem moet kunnen inschatten, stelt zich dan ook al snel deze essentiële
vraag: welk aandeel van de warmte- en koelvraag van het gebouw kunnen
optimaal door het GEOTABS systeem en het secundaire systeem gedragen
worden?
Hoofdstuk 1 verduidelijkt de ambiguïteit in het ontwerp van een dergelijk
toekomstbestendig HVAC-concept. Er ontstaat een dilemma tussen een
gedetailleerde en nauwkeurige maar dure ontwerpprocedure, en een
vereenvoudigde en snelle maar erg onnauwkeurige ontwerpprocedure, die
belangrijke dynamische aspecten van het systeem niet meeneemt. Een
daarvan is de thermische traagheid van TABS die, indien optimaal benut,
het benodigde vermogen van het productiesysteem kan verkleinen door het
spreiden (en verschuiven) van de gebouwbelasting in de tijd
(piekvermogenreductie). TABS brengt namelijk een tijdsverschuiving mee
tussen het moment dat het opwekkingssysteem vermogen levert en de
energievraag van het gebouw. Die tijdsverschuiving moet worden
geïntegreerd in de ontwerpprocedure met behulp van dynamische
modellering van het systeem om tot een robuust ontwerp te komen.
Daarom is een vereenvoudigde ontwerpprocedure nodig die het dynamische
gedrag van TABS optimaal incorporeert. Om een dergelijke
ontwerpprocedure te verkrijgen, zijn vereenvoudigingen in de modellering
en optimalisatie vereist.
In Hoofdstuk 2 wordt een geautomatiseerde methode ontwikkeld voor het
genereren van een vereenvoudigd model van het gebouw en TABS. Op basis
van de gedetailleerde simulatieresultaten maakt de voorgestelde aanpak
gebruik van een inverse modelleringstechniek om een vereenvoudigd grey-
boxmodel van het gebouw en TABS te genereren. Het vereenvoudigde
model wordt vervolgens gebruikt in de ontwerpprocedure om het
dynamische gedrag van TABS en het gebouw te integreren.
In Hoofdstuk 3 wordt een heuristisch algoritme ontwikkeld om een
geautomatiseerd optimaal load-splitting-algoritme (OLSA) te creëren.
OLSA gebruikt het vereenvoudigde model dat in hoofdstuk 2 werd
ontwikkeld om optimaal te voorzien in de thermische vraag van het gebouw
met TABS en (indien nodig) een secundair systeem. OLSA genereert
tijdreeksen van het primaire en secundaire systeemvermogen voor het hele
jaar. Op basis daarvan kunnen de belangrijkste componenten van het
24 Samenvatting
hybride systeem optimaal ontworpen worden, met behulp van de
nabewerkingen die in de volgende hoofdstukken worden uitgewerkt.
In Hoofdstuk 4 wordt de OLSA gebruikt in een geïntegreerde, op
dynamische simulaties gebaseerde, ontwerpprocedure om een techno-
economisch beoordelingsmodel (TEA) voor hybridGEOTABS te maken. De
ontwerpprocedure is gebaseerd op talrijke dynamische van het BEO-veld
en gebouw voor elke casestudie. Het TEA-model wordt toegepast op negen
casestudies en met de resultaten wordt de definitie van het optimale
ontwerp bestudeerd. Verder worden uit de resultaten enkele algemene
richtlijnen afgeleid voor het optimale ontwerp van hybridGEOTABS. De
resultaten tonen de grote invloed van gebouwkarakteristieken op de
prestatie van het hybridGEOTABS concept, en dus tonen zij ook het
belang aan van een vroege integratie van de HVAC-keuze in het
ontwerpproces. Het juiste ontwerp wordt bereikt door een iteratief
ontwerpproces waarbij de architect zich bewust is van de gevolgen van elke
architecturale ontwerpbeslissing op de prestatie van het HVAC-systeem.
Methodes die zo’n iteratief HVAC-ontwerp op tijds-efficiënte wijze mogelijk
maken, komen in de volgende hoofdstukken aan bod.
In hoofdstuk 5 wordt met behulp van de OLSA-simulatieresultaten
(hoofdstuk 2) en de lessen uit hoofdstuk 4, een rekenkundig efficiënte
ontwerpprocedure ontwikkeld die toepasbaar is op duizenden case studies
uit een beschikbare database van gebouwenergiesimulaties van de
gebouwenstock. Het resultaat is een nieuwe database met een bijna-
optimale hybridGEOTABS-ontwerpen. De ontwerper kan dus
voorgedimensioneerde ontwerpen opzoeken in een nieuwe database. De
ontwerper kan zo het gebouwontwerp en het GEOTABS-ontwerp evalueren
aan de hand van high level gebouwkarakteristieken (vb. isolatiegraad,
beglazingsaandeel). Bovendien resulteert deze aanpak in een rijke database
als basis voor de gegevensanalyse die in het volgende hoofdstuk wordt
uitgevoerd.
Hoofdstuk 6 maakt het onderscheid tussen een geschikte ontwerptool voor
HVAC-ontwerpers en architecten. De database in hoofdstuk 5 is zelf een
ontwerptool voor HVAC-ontwerpers. Architecten kunnen de grootte-orde
van de prestatie van hun gebouwontwerp echter sneller raadplegen via een
beslissingsboom, waarvan de takken zijn opgesplitst volgens high level
gebouwkarakteristieken zoals de locatie, isolatiegraad, beglazingsaandeel….
Zo kunnen ze het ontwerp van het gebouw wijzigen om de milieu- en
energieprestatie-indicatoren te verbeteren door het aandeel van
Samenvatting 25
hernieuwbare bronnen zelfs aan het begin van het ontwerp te vergroten en
zonder simulaties uit te voeren. Dit maakt een geïntegreerd gebouw- en
HVAC-ontwerp mogelijk waarbij het gebouwontwerp kan worden
aangepast om een beter HVAC-ontwerp te bereiken. Om de
beslissingsbomen af te leiden, worden classificatietechnieken, gebruikt voor
geautomatiseerde gegevensanalyse. De toepassing van de
ontwerpbeslissingsboom wordt geïllustreerd aan de hand van drie
casestudies
26 List of Tables
Table 1-1 Organization of the book. ........................................................ 66
Table 2-1 Conditioned floor area (m2) of ................................................. 84
Table 2-2 Characteristics of different case studies ................................... 84
Table 2-3 comparison between the verified and estimated peak loads and
annual load for the three case studies. ..................................................... 87
Table 3-1 Characteristics of the three groups of case studies. ................ 106
Table 3-2 Comparison between the J values calculated by the OLSA and
the baseline ............................................................................................. 108
Table 4-1 Building envelope characteristics of the 9 different case studies
used in this chapter. ................................................................................ 136
Table 4-2 Parameters values used in this chapter. ................................. 137
Table 5-1 Production efficiencies for the different systems..................... 177
Table 5-2 Primary energy and CO2 emission conversion factors. ........... 177
Table 6-1 Estimated values of design indicators for the three case studies
in Brussels based on the proposed decision tree ..................................... 201
Table 1 The main parameter values for TABS used in modeling in the
book. Position of the TABS are changed in modeling by changing the
heat transfer coefficient between zone and the TABS surface [88, 24, 91].
................................................................................................................ 217
Table 2 The building parameters for different typologies used for
estimation of the design ........................................................................... 221
Table 3 Maximum daily average solar irradiations (Ihor.max) in different
locations .................................................................................................. 223
Table 4 A summary of the parameters used in the modeling and
simulation of the four typologies. ............................................................ 228
Table 5 Dimension of the building and geometrical variables used in the
nine case studies. ..................................................................................... 233
Table 6 Three insulation groups used in the nine case studies in the book.
................................................................................................................ 234
List of Figures 27
Figure 1-1 CO2 emissions track (billion tonnes per year) over the past
170 years (right vertical axis), comparison between observed and
simulated earth surface temperature (left vertical axis) [6]. ..................... 37
Figure 1-2 Key polices and milestones on the road to net zero emissions
[6]. ............................................................................................................. 39
Figure 1-3 The building sector’s share of global annual energy use and
carbon emissions in 2019 (left) [3] and share of different sub-sector from
the energy use in the building sector in 2020 [10]. ................................... 40
Figure 1-4 Schematic 2D drawings of radiant systems [48] ..................... 43
Figure 1-5 Simplified schematic of the flow process in a heat pump. ..... 45
Figure 1-6 Share of installed GSHP from total HVAC installations in
Europe [39] ................................................................................................ 47
Figure 1-7 Key components of hybridGEOTABS .................................... 49
Figure 1-8 conceptual relation between the thermal power of the primary
system production unit, TABS, secondary system and the building
cooling loads for 5 days from Friday to Tuesday in an office building,
regions A and B represent examples of different definition of optimal
behavior of TABS. .................................................................................... 54
Figure 1-9 The conceptual relationship between design decisions and the
costs of the product during its life cycle (left) adapted from [200], and
design options in the various stages of design procedure (right). ............. 60
Figure 2-1 Flowchart of a general framework of the grey-box approach. . 70
Figure 2-2 The RC model developed by Gwerder et al.[88] (a), ............... 73
Figure 2-3 Grey-box modeling trends based on literature review study by
Li et al. [85]. ............................................................................................. 75
Figure 2-4 Schematic of the RC model used in this study. ...................... 76
Figure 2-5 Flowchart of the parameter estimation methodology. ............ 82
Figure 2-6 Different zones and floors in the case studies. ......................... 84
Figure 2-7 Normalized objective function value attributed to each set of
parameter values, one point is shown as an example, note that the 3D
representation is to observe the sensitivity only. ...................................... 86
Figure 2-8 Comparison between validated ideal demand calculated by the
detailed model and the residuals between the estimated ideal demand and
the validated ideal demand for the three case studies. ............................. 89
28 List of Figures
Figure 2-9 Distribution of the residuals between validated and estimated
ideal demands divided by the maximum ideal demand in the three case
studies. ..................................................................................................... 90
Figure 2-10 The operative temperature evaluation flowchart. ................. 91
Figure 2-11 Operative temperature TZV in the detailed Modelica model
using and calculated by simplified RC model
(horizontal axis shows the different zones e.g. Z1 MF represents Zone1
middle floor). ............................................................................................ 93
Figure 3-1 High level flowchart of the implementation of the OLSA. .... 102
Figure 3-2 Load duration curves for the nine case studies. .................... 107
Figure 3-3 Tracking the value of normalized objFun (kWh/m2/year)
during the entire simulation (left to right) and a comparison with the
final objFun calculated by the baseline (right). ...................................... 115
Figure 3-4 Annual load split between the primary and the secondary
systems (kWhth/m2/year) and a comparison between the OLSA and
baseline for nine case studies. ................................................................. 110
Figure 3-5 Comparison between the time series of the primary and
secondary systems (left axis) and the zone temperature (right axis)
calculated using the OLSA and the baseline (simulation results for 7
days, beginning from July 13th ) ............................................................. 114
Figure 3-6 NRMSD between the hourly time series of ,
, and calculated by the OLSA and the baseline............ 112
Figure 3-7 Optimized objFun (kWh/year/m2) and calculation time (s) as
a function of prediction horizon N (hour) for nine case studies using
OLSA (note that the calculation time and objFun have the same order of
magnitude. The objFun obtained for the baseline for each case study is
shown separately on the right side). ....................................................... 116
Figure 3-8 FFT of the primary system power, calculated using the OLSA
and baseline (center) compared to the FFT of the building hourly ideal
heating and cooling loads for one year. ................................................... 118
Figure 4-1 Flowchart of simulation-based optimal design procedure. .... 123
Figure 4-2 Pareto optimal solutions for multi-objective problems [118]. 124
Figure 4-3 Flowchart of the proposed simulation-based design
methodology. ........................................................................................... 126
Figure 4-4 Example of the outputs of step 2 of the design procedure,
representing time series of the primary and the secondary system power
for one week of simulations. .................................................................... 128
List of Figures 29
Figure 4-5 LDCs for cooling (left) and heating (right) for the nine cases
studies. .................................................................................................... 136
Figure 4-6 Comparison between the present value of operational costs
over 25 years and investment costs, for different shares of GEOTABS,
together with the carbon dioxide emissions for nine case studies. .......... 141
Figure 4-7 Break down of the specific net present costs for 25 years (bar
graph related to the left vertical axis) and the share of GEOTABS (black
line related to the right vertical axis) for the 9 case studies; investment
costs are distinguished by darker colors ................................................. 143
Figure 4-8 Cashflow of the nine case studies with different design
scenarios (DesignID) for 25 years. .......................................................... 145
Figure 4-9 Relation between electricity price (€cent/kWh) with payback
period (a) and with GEOTABS optimal share (b). ............................... 148
Figure 4-10 Relation between the share of GEOTABS and the production
units relative sizing. ................................................................................ 151
Figure 4-11 Contribution of the peak, monthly and annual loads to the
borefield length in a) absolute terms, and b) relative terms. Building
imbalance (above each plot) and share of GEOTABS (black line) are also
indicated. (Courtesy of Iago Cupeiro Figueroa). .................................... 154
Figure 4-12 The relationship between the annual thermal imbalance (%)
and the average power of the borefield (W/m) in heating and cooling
modes for the nine case studies under 15 DesignIDs. ............................. 157
Figure 4-13 The impact of the carbon dioxide emission factor on the
carbon dioxide emission and CO2 saving for the nine case studies (CO2
conversion factor for Poland is 751 g/kWh which is not shown in the
graph). .................................................................................................... 158
Figure 5-1 Admittance, transmittance and stored energy in TABS versus
time [105] ................................................................................................ 167
Figure 5-2 Deviation between maximum loads with and without moving
average in heating and cooling modes for the primary and secondary
systems (deviation in % is the difference between maximum loads of the
original signal (AVG1) and smoothed signals (AVG2, AVG6,…, AVG168)
divided by the maximum loads of the original signal (AVG1) to be
normalized............................................................................................... 169
Figure 5-3 Normalized deviation of thermal loads the primary and
secondary systems between scenarios with and without moving average in
heating and cooling modes for all the case studies. ................................ 172
30 List of Figures
Figure 5-4 Steps to go from load duration curves to the final sizing of the
key components. ...................................................................................... 176
Figure 5-5 The algorithm for final sizing of the hybridGEOTABS
components. ............................................................................................ 176
Figure 5-6 Component sizing comparison between MPC and RBC
scenarios, and OLSA with and without applying the balancing
assumption. ............................................................................................. 179
Figure 5-7 Annual load split with different methodologies, MPC and
RBC, OLSA with and without applying the balancing assumption ...... 181
Figure 5-8 comparison of the borefield length with different approaches,
MPC, RBC, OLSA-Balanced, and OLSA-Imbalanced ........................... 181
Figure 6-1 Traditional performance-based design (left) and informed-
decision design procedure (right) [173]. .................................................. 187
Figure 6-2 Distribution of the share of the total energy demand that can
be covered by GEOTABS in the office buildings included in the
database (a), distribution of the share of the total energy demand that
can be covered by TABS or secondary system (system share) in the
office buildings included in the database (b). ......................................... 190
Figure 6-3 Distribution of the nominal power for HP, boiler and chiller in
office cases. .............................................................................................. 191
Figure 6-4 Distribution of the HP power as the output of the sizing
methodology for all office cases (a) and office cases with more than 80%
share of GEOTABS (b). ......................................................................... 192
Figure 6-5 Distribution of the specific HP power in all office cases ....... 193
Figure 6-6 Distribution of the specific maximum HP power in office cases
in function of different building design parameters. ............................... 193
Figure 6-7 Maximum power required from heat pump (vertical axis) vs.
designheating for all the office building case studies from the database.
Difference between different regression lines for different groups according
to subgroups. ........................................................................................... 194
Figure 6-8 Classification algorithm leading to a decision tree. ............... 195
Figure 6-9 Sizing decision tree for (a) boiler and for (b) heat pump,
derived from automated classification algorithm applied on simulations
results from Chapter 5. ........................................................................... 198
Figure 6-10 Final decision tree for office typology in Brussels. .............. 199
Figure 1 Detailed flowchart of the Optimal Load Split Algorithm (OLSA)
................................................................................................................ 218
Figure 2 Regression lines for the GSHP and borefield prices ................. 219
List of Figures 31
Figure 3 Ratio between maximum steady state and dynamic thermal
loads for low occupancy (top row) and high occupancy (bottom row)
office buildings in the database............................................................... 224
Figure 4 Comparison between regression lines derived from OLSA
simulations and the theoretical explanation of the heating curve. ......... 230
Figure 5 Comparison between the optimal energy use using heating
curves estimated with MATLAB solvers and the OLSA........................ 232
Figure 6 Geometrical shape of the nine case studies used in this book
during verification steps where n is the number of floors and h1 is the
height of each floor. ................................................................................ 233
32 Chapter 1
1.1 Introduction
Future-proof buildings are energy flexible, meaning that they can shift their
heat demand to times when sustainable energy supply is abundant, thus
reducing the use of fossil fuels. Such buildings necessitate innovative
heating, cooling, and air conditioning (HVAC) solutions. Being a future-
proof HVAC concept, hybridGEOTABS is the combination of a
GEOthermal heat pump (with an underground borefield as a heat source
and sink) and a thermally activated building system (TABS), and is
combined with an additional secondary heating and cooling system. TABS
is a radiant heating and cooling emission system activating the building
structure as thermal storage and providing energy flexibility. TABS
provides low temperature heating and high temperature cooling and thus
can bring higher efficiency for the heat pump system.
While heating and cooling technologies have improved significantly, the
improvement has not been reflected equally in the market. It is
demonstrated in this book that the feasibility and performance of
innovative HVAC solutions (i.e. hybridGEOTABS) are highly sensitive to
architectural design. This thesis defies conventional HVAC design
Introduction 33
methodologies that overlook the synergy between different actors such as
architects and HVAC designers and thus cannot be used for optimal design
of innovative and sustainable HVAC solutions. Moreover, it is shown that
the synergy has the biggest impact at the early stages of the design when
the degree of freedom in HVAC and architectural design is still high. In
response, an integrated design procedure for hybridGEOTABS as a
sustainable and innovative HVAC solution is developed in this PhD thesis.
The overall context for future-proof buildings, the specific properties of
hybridGEOTABS buildings and the general research objective are further
elaborated in this introductory chapter, that concludes with the detailed
research objectives addressed in the rest of the book. This book argues that
the early stage of the design is where important decisions are made, and
the detailed design is grounded. The decisions in this step can be optimized
with an integrated design method that facilitates iterations between
architectural and HVAC design. Integrated design methods require easy-
to-use tools that are compatible with limited data available at the early-
stage design steps. However, the integrated design method cannot be
oversimplified. In chapter 1, it is argued that the dynamic behavior of the
TABS is indispensable to the design procedure. Moreover, the decision on
the share of GEOTABS and the secondary systems from the total annual
demand of the building can significantly impact the component sizing,
energy performance, environmental impacts, and financial aspects of the
design. Furthermore, optimal control of the thermal inertia of TABS is a
crucial part that must be incorporated into the design procedure to achieve
a robust design. All of this, however, complicates the design procedure and
makes the optimal design inaccessible at the early stages. Moreover, the
iteration between the architectural design and HVAC design will become
time consuming and hence neglected. As a result, architectural designs that
prohibit GEOTABS adoption may emerge, and HVAC designers may opt
for less efficient but more conventional and conservative solutions. Different
chapters of this book provide means to achieve an easy-to-use integrated
early-stage design tool that incorporates the dynamic behavior and optimal
control of thermal inertia of TABS. The tool easily enables the designer to
evaluate the consequence of a variety of architectural decisions on the
energy performance, environmental impacts, and components sizing of
hybridGEOTABS.
A simplified model of the building and TABS that incorporates the
dynamic behavior of the system into the early-stage design is developed in
chapter 2. In chapter 3, an optimal load split algorithm is developed that
34 Chapter 1
incorporates the optimal control of TABS into the early stage of the design.
In chapter 4, optimal design of hybridGEOTABS is investigated with a
techno-economic assessment (TEA) model developed in this book. The
TEA model is relatively complex and research oriented. Thus, in chapter
5, a computationally efficient design procedure is developed and applied on
nearly 140,000 case studies from a previously developed database to provide
a new pre-engineering database of near-optimum design of
hybridGEOTABS for HVAC designers. In chapter 6, the design tool
required for architects and HVAC designers are first differentiated and then
architects are targeted. An integrated design tool (design decision tree) is
created by applying classification techniques to the dynamic simulation
results available from the pre-engineering database. The design decision
trees facilitate simple iterations between architectural design and energy
performance estimation at the very beginning of the building design.
1.2 Context
1.2.1 The so-called “trigger point” is approaching
fast
The Sixth Assessment Report (AR6) from Working Group 1 (WPI) of the
Intergovernmental Panel on Climate Change (IPCC) [1] has confirmed a
fearful crisis on the basis of over 6000 recent peer-reviewed scientific
publications. According to AR6 part 1 [1], the debate over the existence of
global warming has reached a clear conclusion: not only is the planet
warming, but it is already warmer than it should be. We are experiencing
global warming consequences such as reductions in crop yields, depleted
freshwater reserves, melting ice caps, and more frequent heat waves. The
contribution from working group 2 (WGII) [2] states that global warming
is unequivocally threatening people’s lives. AR6 emphasizes the critical
importance of taking immediate action to stop the global warming process.
The unvarnished truth implies that previous human actions were
insufficient to slow down global warming. We are approaching "the point
of no return", also known as the "trigger point", at which we will no longer
be able to control or reverse global warming. After the trigger point is
reached, the process will enter an independent self-sustaining loop in which
causes and effects empower one another. For example, demand for cooling
in buildings has recently increased significantly. Regions where no cooling
systems have been previously used during the summer are now installing
Introduction 35
cooling systems, as heat waves become more frequent [5]. Using more
inefficient cooling technologies will increase energy use and CO2 emissions,
accelerate global warming, and repeat the cycle. WGII clearly concludes in
AR6 [2] that CO2 emissions must be drastically reduced. Higher targets for
reducing CO2 emissions must be set. That is the only well-established
strategy for preventing additional global warming in the time remaining
before the trigger point.
1.2.2 Carbon emission level and global warming are
definitely related
The average surface temperature of the earth and the CO2 emission rate
have been unanimously accepted as two key indicators which can be used
to track the status of the global warming crisis [6]. The average surface
temperature of the planet has experienced a relatively continuous increase
in the course of the 20th century (Figure 1-1). Importantly, this increase
became steeper after the 1960s. It is widely held that the average surface
temperature of the earth has an direct relation with the CO2 emission rate,
as shown in Figure 1-1 [6]. It can be seen from the figure that the average
temperature of the planet was stable when the CO2 emission level was low
(prior to the 19th century). According to simulation results (Figure 1-1, the
green line), had the CO2 emission rate stayed within its natural bounds,
the average temperature would also have remained within a safe range.
Limiting global warming to around 1.5 °C (2.7 °F) requires global
greenhouse gas emissions to peak before 2025 at the latest, and be reduced
by 43% by 2030 [4].
36 Chapter 1
Figure 1-1 CO2 emissions track (billion tonnes per year) over the past 170
years (right vertical axis), comparison between observed and simulated
earth surface temperature (left vertical axis) [6].
Fortunately, the relation between the CO2 emission level and the average
surface temperature of the planet has been widely acknowledged. As a
conclusion of the recent discourse about global warming, a vast group of
scientists and policy makers have reached a consensus: to shift the discourse
focus away from a search for more evidence to prove the existence of global
warming and its anthropogenic causes towards making plans to resolve this
crisis. Working group 3 (WGIII) of the IPCC explicitly states that the
global GHG emissions can be halved by 2030 if appropriate polices are
taken [4]. Many human activities emit an enormous amount of CO2 to the
environment. Policy makers are thus committed to limiting and decreasing
these emissions by means of appropriate legislation which aims to limit
carbon-intensive human activities [7, 8]. Meanwhile, innovative
technologies are being introduced, demonstrated, and commercialized to
guarantee the smooth replacement of carbon-intensive technologies with
low-carbon technologies. Scientists have also proposed changes to humans'
"lifestyles" in order to slow global warming. Note that this book does not
address these changes, which include, but are not limited to food diets,
travel habits, mobility and transportation, consumption habits, and so on.
Introduction 37
1.2.3 Sustainable technologies can reduce carbon
emissions
The CO2 emissions of different sectors have been tracked, and the energy
sector has been found to be responsible for 75% of total CO2 emissions [6].
Accordingly, various policies have been proposed to transform the energy
system and its sub-sectors into a low carbon emission system (Figure 1-2).
A variety of actions, such as decreasing carbon intensive human activities
and improving the energy efficiency of human activities through sustainable
technologies, have been considered as means of reducing emissions.
Sustainable technologies are those that provide specific services to society
while having minimal environmental impact. The green region in Figure
1-2 shows the CO2 emissions associated with the building sector. For
instance, the installation of new gas boilers should be banned and replaced
by low-carbon technologies such as heat pumps as of 2025, according to
proposed plans by IEA. Appropriate policies and innovative technologies
are envisaged to prevent the negative repercussions of a massive transition
in the socio-economic system. In each sub-sector, emerging technologies will
help society to pass these milestones without sacrificing people’s welfare.
Figure 1-2 Key polices and milestones on the road to net zero emissions [6].
-5
0
5
10
15
20
25
30
35
40
2020 2025 2030 2035 2040 2045 2050
Buildings Transport Industry Electricity and heat Other
2045
150 Mt low-carbon hydrogen
850 GW electrolysers
435 Mt low-carbon hydrogen
3 000 GW electrolysers
4 Gt CO2captured
Phase-out of
unabated coal in
advanced economies
2030
Universal energy access
60% of global car
sales are electric
1 020 GW annual solar
and wind additions
All new buildings are
zero-carbon-ready
Most new clean
technologies in heavy
industry demonstrated
at scale
All industrial
electric motor sales
are best in class
No new ICE car sales
2035
Overall net-zero
emissions electricity in
advanced economies
Most appliances and
cooling systems sold
are best in class
50% of heavy truck
sales are electric
7.6 Gt CO2captured
No new unabated
coal plants approved
for development
2021
2025
No new sales of
fossil fuel boilers
2040
More than 90% of
heavy industrial
production is
low-emissions
2050
Almost 70% of
electricity generation
globally from solar PV
and wind
More than 85%
of buildings are
zero-carbon-ready
50% of heating demand
met by heat pumps
Phase-out of all
unabated coal and oil
power plants
Net-zero emissions
electricity globally
50% of fuels used
in aviation are
low-emissions
Around 90% of
existing capacity in
heavy industries
reaches end
of investment cycle
50% of existing
buildings retrofitted
to zero-carbon-ready
levels
No new oil and gas
fields approved for
development; no
new coal mines or
mine extensions
38 Chapter 1
1.3 Sustainable heating and cooling technologies
According to the latest reports about carbon emissions in the building
sector (e.g., [10]), the building sector is devouring a substantial part of the
energy produced worldwide, and releasing an equally substantial amount
of CO2 to the environment (Figure 1-3, left). Indirect building emissions
are due to electricity use in the building sector, and thus these emissions
are from power plants, given that building operations are responsible for
nearly 55% of global electricity consumption [9]. Direct building emissions
are mainly due to the use of fossil fuels for heating systems. Overall, the
heating and cooling of buildings is responsible for almost a third of the total
energy use of this sector (Figure 1-3, right). At the same time, the building
sector has been observed to be deviating from the path towards carbon
neutrality, and may not achieve the required CO2 emissions reduction
targets if sustainability plans are not accelerated [9].
Regardless of the unsatisfactory current record of CO2 emission reductions
in the building sector, explicit pathways have been outlined for this sector
to reach the intended level of CO2 emission reductions. A combination of
improving the energy efficiency of buildings and HVACs, electrifying
heating and cooling systems, and increasing the share of renewables in
electricity production has been drafted as a composite solution to
Figure 1-3 The building sector’s share of global
annual energy use and carbon emissions in 2019
(left) [3
] and share of different sub-sector from
the energy use in the building sector in 2020 [
10
].
Introduction 39
decarbonizing heating and cooling systems [11, 12]. The European Union
(EU) aims for all new buildings (as of 2020) to be “nearly Zero-Energy
Buildings” (nZEB) [11, article 9]. Future-proofed buildings are energy-
efficient and energy-flexible, allowing for the maximum integration of
residual and renewable energy sources (R2ES). However, due to a lack of
renewable resources, for instance during nights with lower electricity
demands and high production from wind, renewable electricity production
may not be available. On the other hand, there are periods where renewable
electricity production is more than the demand and thus the electricity
supply is curtailed. Energy-flexible buildings allow demand to be shifted to
periods of R2ES abundance. This shift also necessitates smart building
system controls. The challenge for building owners, architects, and
engineers in this sector is to provide a functional and healthy indoor
environment which is also as sustainable as possible. To achieve this,
innovative heating and cooling technologies are being introduced and put
into practice in order to provide thermal comfort in buildings in a
sustainable manner.
1.3.1 Sustainable emission systems
Heating and cooling emission systems in buildings deliver the heat and cold
produced by the production units to the buildings. It has been widely
discussed (e.g., [30, 31, 32, 33, 35]) that the temperature level of the
emission system influences the efficiency of the production unit. That is,
lower temperatures for heating and higher temperatures for cooling are
more favorable for the production unit. These efficient and sustainable
methods to satisfy occupants’ thermal comfort are called low-temperature
heating (LTH) and high-temperature cooling (HTC) systems. These
systems provide temperatures close to indoor comfort temperatures at the
emission level, e.g. within 23–35 °C for LTH and 15–22 °C for HTC [31].
Radiant heating and cooling (RHC) emission systems are LTH and HTC
systems which have been proven to maintain a higher level of overall
thermal comfort [41, 42, 43, 56] and higher efficiency than all-air systems
(see e.g., [44, 45]). ASHRAE [46] defines RHC systems as emission systems
in which radiant heat transfer accounts for more than 50% of the heat
exchange within a conditioned space. In radiant emission systems, the
heating and cooling fluid (normally water) flows into the pipes from where
the heat is being transferred from/into the building to maintain thermal
comfort [23, 24]. Ning et al. [48] categorized radiant emission systems on
the basis of their response time: 1) radiant ceiling panels (RCP) which offer
40 Chapter 1
a quick response time (1.8–6.5 min) since the pipes are attached to thin
conductive metal panels fixed to the ceiling structure; 2) embedded surface
systems (ESS) characterized by pipes embedded within the surface layer
and insulated from the ceiling structure, with a medium response time (0.8–
8.7 h); and 3) thermally activated building systems (TABS) in which the
pipes are thermally coupled and embedded in the central core of the
ceiling/floor structure, exploiting the thermal mass potential of the massive
concrete layer with, consequently, a slow response time in the range of 8.7–
18.8 h (Figure 1-4). The latter system, TABS, has a high time constant
due to the large thermal capacity, and thus storage capacity, integrated
into the system. Thanks to its large thermally activated surfaces and high
thermal storage capacity, TABS can store the energy and release it
gradually and thus the productions system can work in lower temperature
for heating and higher temperatures for cooling than other radiant systems.
TABS can effectively provide thermal comfort with LTH (23 °C-29 °C) and
HTC (18 °C-22 °C) [16] and thus can efficiently work in combination with
production units that use renewable energy sources.
Incentives for integrating storage capacity into heating and cooling systems
have been widely discussed in the literature (e.g., [16, 17, 18, 19, 20]).
Traditionally, storage capacity (such as a storage tank) assists a local heat
production unit (such as a boiler) to deliver heat with an assigned water
temperature to the emission system continuously. In fact, storage capacity
can shift the demand in the building or local production so that it matches
the supply. The advanced application of storage capacity is referred to as
an active demand response (ADR) feature [21]. ADR regulates the demand
to match with the production since renewable electricity production
inherently contains uncertainties. ADR entails a potential flexibility on the
demand side which can be provided by storage capacity. Moreover, ADR
can help the district-level production system to shift demand and avoid
simultaneous peaks (peak shaving) for all the buildings it covers. Similarly,
the production unit can be downsized in the building level when a storage
capacity is integrated to the system. These bring about the opportunity to
integrate more R2ES into the electricity consumption of the district [193]
and saving in investment costs on the production side [53]. In the context
of advanced heating and cooling systems, storage capacity includes, but is
not limited to, batteries, storage tanks, and TABS.