A co-simulation based framework for the analysis of integrated urban energy systems: Lessons from a Swedish case study

(1)

Doctoral Thesis in Energy Technology

A co-simulation based framework for

the analysis of integrated urban

energy systems

Lessons from a Swedish case study

MONICA ARNAUDO

Stockholm, Sweden 2021

(2)

A co-simulation based framework for

the analysis of integrated urban

energy systems

Lessons from a Swedish case study

MONICA ARNAUDO

Doctoral Thesis in Energy Technology KTH Royal Institute of Technology Stockholm, Sweden 2021

Academic Dissertation which, with due permission of the KTH Royal Institute of Technology, is submitted for public defence for the Degree of Doctor of Philosophy on Brinellvägen 68 - Room M263 - 16th April 2021 - 8 am, Stockholm

(3)

ISBN 978-91-7873-803-8 TRITA-ITM-AVL 2021:9

(4)

Ai miei genitori che mi hanno crescuta nella persona di cui oggi sono orgogliosa di essere.

(5)

(6)

Abstract

As major responsible for CO2 emissions, the energy sector is urgently called to take action against climate change. The integration of renewable energy resources is a solution that, however, comes with a challenge. In fact, renewables are often variable, unpredictable and distributed. These characteristics add an extreme complexity to the design and control of energy systems. Sector-coupling is nowadays strongly supported as a promising approach to increase the flexibility of these systems. For example, wind power curtailment can be reduced by using the power surplus to operate heat pumps. When the wind does not blow, the heat stored in the thermal mass of the buildings and waste heat recovery can be used instead. These solutions are largely available at district-to-city level. However, a suitable framework to design these integrated urban energy systems is missing.

This thesis work proposes such a framework, as a set of methodological steps and integrated modelling tools. Among them, the modelling and simulation approach is a fundamental aspect. Given the heterogeneity of integrated energy systems, dedicated technology-specific models are developed and used to achieve the required level of detail. A co-simulation method is implemented when time step coordination and data exchange are necessary. Scenarios are developed to compare the techno-economic and environmental performance of alternative solutions, based on sector-coupling. Levelized cost of energy and CO2 emissions are used as main performance indicators for this purpose. In order to show the applicability of this methodology, Hammarby Sjöstad (Stockholm, Sweden) is selected as a case study. This also allows to tackle a real local open issue, which is the definition of the best solution between district heating and domestic heat pumps for multi-apartment buildings.

The proposed framework was successfully applied to the case study. Case specific results allowed to formulate more general conclusions applicable to similar multi-apartment residential districts, in a Swedish context. It could be shown that co-simulation is a useful approach to capture sector-coupling bottlenecks and opportunities. Respective examples are electricity grid overloadings caused by installations of heat pumps and the control of thermal mass in buildings to replace the use of heat peak boilers. However, co-simulation should be strictly limited to cases where control feedback loops need to be taken into account, such as in the previous examples. This is because it involves a higher implementation complexity and a higher computational time. Thus, for example, the models of a heat network and of an electricity grid with no coupling technologies, such as heat pumps and electric boilers, should be preferably analyzed sequentially. The levelized cost of heat was found to be a game-changer parameter when comparing energy infrastructures, beyond the specific business aspects. For example, the replacement of a district heating tariff with its levelized cost of heat clearly showed the economic advantage of heat networks against domestic heat pumps. The CO2 emissions factors of different energy resources (waste, biomass, electricity mix) were shown to be highly critical for two main reasons. Firstly, different assumptions for these factors led to opposite findings regarding the carbon footprint of specific technologies. For example, heat pumps could be estimated as both more and less polluting than district heating, depending on the assumed emission factors. Secondly, control strategies based on the CO2 emission factors of the electricity supply mix (power-to-heat) were found to be a promising sector-coupling solution. By analyzing integrated energy systems, it was possible to assess uncovered bottlenecks and suggest new options. In particular, it was shown that the installation of a large number of distributed heat pumps can overload the electricity distribution grid in a district. Demand side management, through the thermal mass in buildings and vehicle-to-grid, could help alleviating this problem. On the other hand, district heating was found to be an even more promising alternative, by integrating demand side management and heat recovery. Heat pumps were shown to be a suitable partner technology for supporting heat recovery and enabling power-to-heat. .

(7)

Sammanfattning

Då energisektorn ansvarar för huvuddelen av växthusutsläppen finns det behov att omgående vidta åtgärder för att motverka klimatförändringar. Integrationen av förnybara energiresurser är en lösning, dock kommer den med en del utmaningar. Förnybar energi är ofta variabel, oförutsägbar och distribuerad. Dessa egenskaper medför en ökad komplexitet när det gäller utformning och styrning av energisystemsystemet. Ökad sektorkoppling mellan olika energislag är ett lovande tillvägagångssätt för att öka flexibiliteten i dessa system. Till exempel kan driftinskränkningar för vindkraft minskas genom att använda kraftöverskottet för att driva värmepumpar. När vinden inte blåser kan värmen som lagras i byggnaders termiska massa och återvinning av spillvärme användas istället. Dessa lösningar finns mestadels på stads- och distriktsnivå. Ett lämpligt modelleringsramverk för att utforma dessa integrerade stadsenergisystem saknas dock.

Det föreliggande arbete föreslår ett sådant ramverk som en uppsättning av metodologiska steg och integrerade modelleringsverktyg. Tyngdpunkten ligger på modellering och simulering. Med tanke på de integrerade energisystemens heterogenitet utvecklas dedikerade teknologispecifika modeller för att uppnå önskad detaljnivå. En samsimuleringsmetod implementeras när tidsstegskoordinering och datautbyte är nödvändiga mellan olika modeller. Scenarier utvecklas för att jämföra den tekno-ekonomiska och miljömässiga prestandan hos alternativa lösningar baserat på sektorkoppling. Nivellerade energikostnader och koldioxidutsläpp används som huvudindikatorer för detta ändamål. För att visa tillämpbarheten av denna metod väljs distriktet Hammarby Sjöstad (Stockholm, Sverige) som en fallstudie. Detta gör det också möjligt att ta itu med en real öppen fråga, nämligen huruvida fjärrvärme eller värmepumpar är den bästa lösningen för hushåll för flerbostadshus.

Det föreslagna ramverket tillämpades framgångsrikt på fallstudien. Fallspecifika resultat gjorde det möjligt att formulera mer generella slutsatser som är tillämpbara på liknande flerbostadsområden i ett svenskt sammanhang. Det visas att samsimulering är ett användbart tillvägagångssätt för att fånga upp flaskhalsar och nya möjligheter i sektorkopplingen. Exempel är överbelastningar av elnät orsakade av installationer av värmepumpar och kontroll av termisk massa i byggnader för att ersätta användningen av toppvärmepannor. Dock bör samsimulering begränsas till fall där regleråterkoppling måste tas i beaktande, såsom i de föregående exemplen. Detta beror på att samsimulering innebär en signifikant högre implementeringskomplexitet och en längre beräkningstid. Således bör exempelvis modellerna för ett värmenät och av ett elnät utan kopplingsteknik- såsom värmepumpar och elektriska pannor- företrädesvis analyseras sekventiellt. När det gäller nyckelprestationsindikatorer visade sig den i detta arbete införda nivellerade värmekostnaden vara en viktig ny parameter när man jämför energiinfrastrukturer utöver de specifika affärsaspekterna. Exempelvis visade byte av fjärrvärmetaxa till nivellerade värmekostnad tydligt den ekonomiska fördelen med värmenät jämfört med lokala värmepumpar. CO2-utsläppsfaktorer för olika energiresurser (avfall, biomassa, elmix) visade sig vara mycket viktiga av två huvudskäl. För det första ledde olika antaganden för dessa faktorer till motsatta slutsatser angående koldioxidavtrycket för specifik teknik. Till exempel kan värmepumpar uppskattas vara både mer och mindre förorenande än fjärrvärme beroende på de antagna utsläppsfaktorerna. För det andra befanns reglerstrategier baserade på koldioxidutsläppsfaktorerna i elmixen (kraft-till-värme) vara en lovande sektorkopplingslösning. Genom att analysera integrerade energisystem var det möjligt att fånga upp flaskhalsar i infrastrukturen och föreslå nya alternativ. I synnerhet visades att installationen av ett stort antal distribuerade värmepumpar kan överbelasta elnätet i ett distrikt. Styrning av efterfrågesidan, genom tex användning av den termiska massan i byggnader och elfordons lagringskapacitet, kan hjälpa till att minska detta problem. På andra sidan visade sig fjärrvärme vara ett ännu mer lovande alternativ genom att integrera både styrning av efterfrågan och värmeåtervinning. Värmepumpar visade sig i detta fall vara en lämplig partnerteknik för att stödja värmeåtervinning och möjliggöra kraft-till-värme-kopplingen.

(8)

Preface

This thesis was completed at the Heat and Power Technology (HPT) Division at KTH Royal Institute of Technology under the main supervision of Professor Björn Laumert. Research at HPT is focused on the fields of poly-generation, aero elasticity, turbomachinery, biofuels in gas turbine cycles, combined heat and power generation, concentrating solar power and integrated energy systems. The work was also co-supervised by Dr. Monika Topel, experienced scientific researcher, leader of the integrated energy systems group. The research has been funded by the European Union, under the ERA-NET IntegrCiTy project, the Swedish Energy Agency, under the Digitalt beslutsstöd för fossilfritt Stockholm project, and the Royal Institute of Technology. This doctorate thesis is a compilation based on six scientific appended articles. The thesis provides a summary of the articles as well a synthesis of the dissertation project as a whole.

(9)

Acknowledgements

My supervisor Björn Laumert gave me a precious guided freedom. Under his supporting supervision, I could take my decisions, make my mistakes, learn my lessons and achieve my goals.

Monika officially became my co-supervisor towards the last period of my PhD path. Truth is, she’s always been there. Monika is to me a mentor, an example of strength (or cardio?? XDXD), resilience and endurance and last, but not least, a precious friend.

Rafael got me ready to walk this PhD path with his deeply educative supervision during my master thesis project. He believed in me from the beginning and supported me throughout.

All my colleagues and friends in EGI created an environment that felt home. Fikas, lunches, celebrations and seminars mixed fun and educative moments. Once upon a time, there was this special group of PhD students friends: Rafael, Monika, Maria, Johan (Dahlqvist), Tobias, Jorge, Mauri. I could look at these guys as my big siblings. Dado came as a visiting PhD for a little while, bringing back the spirit of the master thesis group. Moritz prefers Barcelona, but his little time in Stockholm was enough to create a long lasting friendship. Saman has been a caring colleague, classmate and friend. Osama supported me as a project collaborator and as a friend, in my first challenges as a PhD student. Johan Dalgren joined in the second phase. He brought with him genuine enthusiasm, dedication and support, which I am immensely grateful for. I got the opportunity to work on my last PhD publication with Fabio. I thank him for his professional collaboration, availability and support. Thank you to Marco and Davide for keeping an eye on me all along. From Switzerland, Austria and Sweden, the IntegrCiTy team accompanied me through the first phase of my PhD path. Massi led us with his enthusiastic and engaged spirit. Pablo shared unimaginable passion and patience.

I cannot begin to tell how enriching was to be part of the InnoEnergy PhD School. The trips that I took left me an unrepayable luggage of human, cultural and educative values. Through this channel, I could also get the chance of setting up industrial collaborations that gave me the opportunity to use my research beyond academia. I thank José Acuña for involving me in his group’s work and environment. Cecilia Ibánez introduced me to a teamwork experience that allowed me to put myself in the game.

I made several friends throughout these years in Stockholm. Fikas, hikes, meals, trips, parties, movies and all kinds of events would have not been as much fun without them. I would like to especially thank those who supported me on a daily basis: Hanna, for understanding and sharing; Silvia, for being an exceptional gym-buddy and motivator; Eftyhia, for her enthusiastic and proactive spirit; Letizia, for her genuine friendship; Stefania, for being a reference point from the beginning of my adventure in Stockholm. In Stockholm, I also realized how much I love group cross-training. The toughest strength, cardio and core challenges from my favourite instructors made my day countless times.

I feel so lucky to have one-call away guardians of my soul. Roberta, grazie per quelle immense chiacchierate a cuore aperto, in quel di Copenhagen. Angioletto, grazie per quelle telefonate dai mille aggiornamenti e ragionamenti. Salvatore, grazie per la tua capacitá di ascoltare e comprendere veramente. Francy, grazie per credere in me e per il genuino supporto. Marti, grazie per essere una di quelle amiche su cui so di poter sempre contare. Miki, aspetta che ti mando un audio kilometrico.

La mia famiglia in quel della Valle Stura e de la Provence é la base solida della mia vita. Un immenso grazie per il continuo supporto va ai miei preziosi cugini, zii e nonni (Franca e Marcello di Rialpo, Zeta e Marcello dei Perdioni). Grazie alla mia amica-cugina-coetanea Lori, lontana ma vicina. Grazie a zio Marco, perché ci si capisce tra chi viaggia. Grazie a Mimi per l’instancabile affetto.

Grazie Vale, perché ci sei. Grazie Nesrine perché ti prendi cura di mio fratello. Per Mamma e Papá le parole mancano.

(10)

Nomenclature

Abbreviations

BB Building Block

CAPEX Capital expenditure

COP Coefficient of performance

DH District heating

dsHP Domestic scale heat pump

dsTES Domestic scale thermal energy storage

DC Data center

EB Electrical backup/boiler

EV Electric vehicle

HOB Heat only boiler

HP Heat pump

HR Heat recovery

HTDH High temperature district heating

KPI Key performance indicator

IUES Integrated urban energy systems LCOE Levelized cost of energy LCOH Levelized cost of heat

LTDH Low temperature district heating MILP Mixed integer linear programming MINLP Mixed integer non-linear programming MOO Multi-objective optimization

OPEX Operational expenditure

PV Photovoltaic panel

RQ Research question

SM Supermarket

SO Single optimization

SOC State of charge

TES Thermal energy storage

TM Thermal mass

V2G Vehicle-to-grid

(11)

1. Introduction

Climate change linked to human behavior is not a new problem. However, the raise of its global awareness has taken almost three decades to generate official consensus. Between the first public declarations in the ’80s ([1], [2]) and the Paris Agreement in 2015 [3], the urgency of taking action has increased [4] and today has reached the societal concern [5].

The energy sector, including transport, is responsible for almost 80% of the greenhouse gas emissions in Europe [6]. The European Commission has recently reacted to this situation by publishing a strategy for

Energy System Integration [7] in order to:

“[…] rely on an ever growing share of geographically distributed renewable energies, integrate different energy carriers flexibly, while remaining resource-efficient and avoiding pollution and biodiversity loss.”

The use of local renewable resources avoids transporting energy over long distances. However, this leads to a more decentralized system, whose complexity increases when the energy sources have a fluctuating availability (for example wind and sun). Thus, the flexibility of the energy infrastructures becomes a key player that enables shifting the energy use in time and/or magnitude [8]. Energy system integration, or sector-coupling, offers a large variety of solutions to implement this flexibility. When looking at multi-energy systems, cogeneration plants and heat pumps (HPs) can be operated in district heating (DH) networks to accommodate variable renewables-based electricity. HPs can be recruited when surplus electricity is available. Instead, power generation in cogeneration plants can be prioritized in the opposite situation, thus enabling a power-to-heat control strategy [9]. The thermal mass of buildings can be taken into account as a thermal storage device. This solution can be used to shift the heat peak load and avoid the operation of expensive and polluting boilers.

These examples become more evident and numerous when taking a district-to-city level perspective ([10]). The district scale is of particular interest in relatively large cities, where expansion planning is done by neighbourhoods. The construction of a new district, and thus a new energy infrastructure, offers the opportunity for an easier implementation of new solutions. This is where the following challenge comes in:

the application of sector-coupling to the planning and design of energy infrastructures at district level requires a comprehensive framework that is today still missing [11].

This framework should be based on methodological steps, modelling and simulation tools and key performance indicators (KPIs) able to support the decision making process of relevant stakeholders. Profits, savings and comfortable indoor temperatures are examples of objectives that are also often in contrast with each other. City planners, energy utilities and consumers need these decision-support tools not only to match their needs but also to urgently take the next steps towards the solution of the climate crisis. In the context of integrated urban energy systems (IUES), a particular challenge is presented by the achievable level of detail for modelling and simulations. A level of detail indicates the depth that can be reached when studying a system. For example, a DH network can be looked at as a black box with a supply and a return heat flow (kW and kWh based analysis). The same technology can be analyzed as an interconnection of pipes with specific thermal and hydraulic properties (temperatures and mass flow rates based analysis). For reliable levels of detail, different energy carriers and technologies often require the use of different modelling tools and/or programming languages. For this reason, the proposed framework is expected to have a modular setup. In this way, suitable tools can be selected and plugged in for each case. This modularity, however, leads to an additional challenge. This is the possibility of capturing time interdependences between models implemented in different environments. Control functions are often the reason of such links that should be captured within the simulation of IUES. A typical example is the coordination of a DH central supply mix and demand response in buildings. Finally, given the involvement of multiple stakeholders in IUES, a clear side should be defined for the perspective taken within the proposed framework. In particular, there is a need for differentiating the outcomes of a system perspective from the business interest of different stakeholders. The former one aims at selecting a solution (an energy

(15)

infrastructure) by excluding profit interests that typically distort the original benefits of the available options. At the same time, these business-related aspects should not be forgotten since they ultimately define ownership and management. The scale of the systems perspective, in this thesis work, is set on a district level, with a potential to be upgraded at city level.

Stockholm is a front runner city in the fight against climate change. The ambitious target of zero net carbon emissions by 2045 calls for effective measures to be implemented as soon as possible. When it comes to the planning of new energy infrastructures, Stockholm and its districts offer the place for a lively debate. The heating system of multi-apartment residential buildings, in particular, is currently the object of a discussion based on two solutions perceived as in competition with each other. On one hand, the traditional DH system makes progress towards a cleaner carbon footprint [12]. On the other hand, electricity is promoted as an already cleaner resource [13] and electricity prices are relatively low in Sweden. Thus, domestic HPs, at multi-apartment building scale, become an interesting option [14], along with attractive short term economic savings on the energy bill of the heat consumers. From a system perspective, however, a large adoption of this power-based technology rises a concern related to the loading limitations of the electricity distribution grid. Currently, in fact, the incoming electric transmission capacity to Stockholm is close to its saturation [15]. In the short term future [16], this is expected to force the system operators to impose constraints on the distribution level as well. Projects for the power grid’s upgrade take time and huge investments, which slows down the process for achieving the climate targets.

One direction to take, to increase the decarbonisation process speed, is a more efficient use of existing resources [17]. Energy demand side management offers such a potential with the possibility of being implemented both for heat and power peak load shifting. The control of the thermal mass in buildings [8] and of the electric vehicles’ batteries (vehicle-to-grid, [18]), as energy storage devices, are two promising options for this purpose. Another possible direction is to show the potential of DH to offer a lower heat generation cost compared to domestic HPs for multi-apartment residential districts. Moreover, by integration, rather than competition, with power-based and distributed technologies, it can be demonstrated that the current carbon footprint of DH can be lowered even further. Power-to-heat [9] and buildings’ thermal mass control can be shown as key players for this purpose. The former enables a control strategy based on the idea of supplying electricity when its generation mix includes a large share of renewables, and vice-versa [19]. In this sense, power-based technologies, like HPs, can be considered a useful load, especially when electricity surplus should be absorbed to avoid curtailment. Fuels-based DH can be used in the opposite situation, as long as its carbon footprint can be shown to be lower than the electricity mix’s one. The buildings’ thermal mass can be used as a thermal storage device so that heat peak demand can be shaved [20]. This strategy has the potential of reducing the need of generally expensive and polluting central boilers.

This discussion, around the case of Stockholm, is clearly a relevant example that requires a combined sector-coupling and district level approach. This thesis aims at contributing to the development of a modular framework for planning and design of IUES. This framework should tackle the challenges of reaching reliable levels of detail and capturing potential time interdependences between different simulation environments. The development of such a framework is achieved by answering general-level research questions (RQs) from a system perspective. A comparison with the perspective of relevant stakeholders is included. A typical district in Stockholm, Hammarby Sjöstad, is used as a case study to demonstrate the real world application of this framework. Different scenarios are proposed and analysed in the form of sets of design parameters, such as technologies’ sizes and operation strategies. The findings from a systematic scenario analysis answer case-specific RQs.

(16)

1.1. Research questions and objectives

On the basis of the above discussion, this thesis work is based on the following RQs:

1. How can a modular framework be designed to analyze IUES by providing an analysis:  flexible to different levels of detail for diverse energy carriers and networks

 able to capture time interdependences that are created when sector-coupling control strategies involve feedback loops

2. Within this framework, which KPIs should be used to enable an analysis of IUES from a system perspective?

3. How can this framework be applied to solve relevant IUES challenges?

a) Can distributed domestic scale HPs be considered an independent heating infrastructure for multi-apartments residential districts?

b) How can the carbon footprint of DH be reduced?

c) How can DH and distributed HPs be compared from a system perspective? d) How does the system perspective compare to the perspectives of:

i. Energy consumers ii. Energy utilities

The questions that form RQ 3 are based on Hammarby Sjöstad as a case study. The main related objectives are the following ones:

:

1. Develop a modular framework, for the design of IUES, which is adaptable to different levels of detail and able to capture time interdependences

2. Identify KPIs to compare energy infrastructures’ scenarios from a system perspective, as part of the aforementioned framework

3. Apply this framework to a relevant case study, by proposing IUES solutions:

a. Energy demand side management to unlock distributed domestic scale HPs as an independent infrastructure

b. Energy demand side management and heat recovery to reduce the carbon footprint of DH c. Analysis of competition vs. synergy between DH and distributed HPs

(17)

1.2. Appended papers

The above RQs are answered by the following list of appended papers. The methodology chapter 4 clarifies the connection between the RQs and the papers.

• Paper I - [21]

Arnaudo, M.; Zaalouk, O.A.; Topel, M.; Laumert, B. Techno-economic Analysis Of Integrated

Energy Systems At Urban District Level – A Swedish Case Study. Energy Procedia 2018, 149,

286–296. (Awarded for research excellence in district heating and cooling at the 16th International Symposium on District Heating and Cooling in Hamburg, Germany).

• Paper II - [20]

Arnaudo, M.; Topel, M.; Puerto, P.; Widl, E.; Laumert, B. Heat demand peak shaving in urban

integrated energy systems by demand side management - A techno-economic and

environmental approach. Energy 2019, 186, 115887. • Paper III - [22]

Arnaudo, M.; Topel, M.; Laumert, B. Techno-economic analysis of demand side flexibility to

enable the integration of distributed heat pumps within a Swedish neighborhood. Energy

2020, 195, 117012 • Paper IV - [23]

Arnaudo, M.; Dalgren, J.; Topel, M.; Laumert, B. Waste heat recovery in low temperature

networks versus domestic heat pumps - A techno-economic and environmental analysis.

Energy 2020, 219, 119675. • Paper V - [24]

Arnaudo, M.; Topel, M.; Laumert, B. Vehicle-to-grid for peak shaving to unlock the

integration of distributed heat pumps in a Swedish neighborhood. Energies 2020, 13.

• Paper VI - [19]

Arnaudo, M.; Giunta F., Dalgren, J.; Topel, M.; Sawalha S.; Laumert, B. Heat recovery and

power-to-heat in district heating networks – A techno-economic and environmental analysis.

Applied Thermal Engineering 2020, 185, 116388.

In these papers, the main author (also author of this thesis) led the development of the projects by taking care of the main part of:

- ideas elaboration;

- models development, validation and implementation; - results analysis and discussions;

- papers’ writing.

Active discussions and collaboration with the co-authors contributed to the overall work development. For each paper, the main author received and integrated (when considered applicable) feedback on the writing from the co-authors and from the respective journals’ reviewers.

1.3. Thesis structure

The remaining chapters are organized as follows. Chapter 2 gives a general introduction about IUES. Details are provided for the concepts that are the focus of this thesis work. Chapter 3 discusses the state of art concerning frameworks for planning and design of IUES. At the end of this part, the contributions and limitations of this thesis are compared to the state of art. Chapter 4 introduces the methodology of this thesis work. Chapter 5 describes the application of the proposed framework to the case study. Conclusions and future work close the thesis in chapter 6 and 7, respectively.

(18)

2. Integrated urban energy systems

This chapter introduces the concept of IUES from a general point of view. The terminology “Integrated urban energy systems” goes along several synonyms, among which “sector coupling”, “multi-energy systems”, “integration/synergies/cooperation/interactions/impact between diverse energy systems”. In this document, such expressions are used, with no distinction, to indicate a general concept. This concept is found comprehensively defined in the recent communication from the European commission, Powering

a climate-neutral economy:

“Energy system integration – the coordinated planning and operation of the energy system ‘as a whole’, across multiple energy carriers, infrastructures, and consumption

sectors” [7]

This definition highlights that the idea of IUES does not only refer to the interaction among diverse energy carriers, namely heat, electricity and gas. The concept of IUES involves also different infrastructures in terms of energy generation, conversion, distribution, storage and consumption. Last, but not least, any energy consumption typology can be included, such as residential, work-place, commercial and industrial energy consumptions.

All these elements have the potential of being designed and thus operated in an integrated way. This means that their planning cannot occur in silos any longer. Moreover, their operation should be coordinated with at least one level of centralized control (block-chain hypothesis are not considered in this discussion). This can happen with either a pre-defined or a feed-back loop control strategy.

When two or more energy carriers are involved, the integration typically occurs through coupling technologies. The most widespread are cogeneration plants (COGs), HPs, electric heating boilers, gas heating boilers and fuel cells [9]. COGs involve the simultaneous supply of both electricity and heat. Polygeneration systems can combine multiple energy carriers and products [25]. The rest of the above mentioned technologies are examples of x-to-x solutions. Power-to-heat [26], power-to-gas, gas-to-heat and gas-to-power ([9], [10]) are not new concepts. However, in the context of IUES, they represent key devices to enable the coordination of multi-energy systems.

When considering a single energy carrier, synergies can be enabled between different stages of its delivery. For example, the supply of DH can be integrated with heat recovery (HR) from local resources, such as data centers (DCs), supermarkets (SMs), industrial processes. The thermal mass (TM) of heat networks and buildings can be used as a thermal energy storage (TES) to perform heat load shifting. Within the power sector, the same type of service and other ancillaries can be offered by vehicle-to-grid (V2G). Examples are conceptually shown in Figure 1. This level of analysis that goes across generation, distribution and demand can also help to plan the allocation of future investments coherently [10].

The analysis of IUES can capture not only positive interactions, but also sector-coupling constraints. This helps preventing unexpected bottlenecks and faults [27]. An example is offered by the loading impact that the installation of a large number of domestic scale HPs (dsHPs) can have on the electricity distribution grid, in multi-apartment residential districts.

(19)

Figure 1 – Conceptual illustration of IUES. Figure adapted from a free resource at [28]

Based on the above discussion, Figure 2 illustrates examples of IUES with coupling of heat and electricity and with cross-interactions among energy supply, storage, distribution and demand. For each technology, the left side label indicates the primary function, while the top side label indicates an additional function. For example, the HP supplies heat, while it generates a demand of electricity as well. Buildings, indicated as building blocks (BBs), can represent demand of heat and electricity only. If their thermal mass is activated, they can also be operated as TES devices.

(20)

The design and control of these IUES introduce a larger system flexibility [29], both in time and space. This means that more renewable resources with unpredictable nature, like solar and wind, can be accommodated across multiple sectors [11]. A strictly related consequence is thus a widespread reduction of the carbon footprint [26]. Moreover, by assuming business and policy support, IUES represent a promising solution to reduce the overall system cost ([29], [27]). Examples are the reduced use of expensive peaking technologies and curtailment of renewables, with higher utilization of existing generation and distribution capacities [26].

The planning and design of IUES face challenges on technical, economic, business, policy and social levels. In particular, from a techno-economic point of view, IUES bring in several layers of complexity, which are difficult to fully capture with traditional simulation models and hard to manage in real life. This complexity is generated by interactions between sectors that are often of different temporal nature [11]. Moreover, the synergies typically involve distributed technologies and networks, thus different spatial scales [11]. New solutions, even if based on old technologies, are typically more expensive at their initial implementation stage. Both hardware and software equipment should be developed and installed as well [29]. This costs’ uncertainty constitutes a further challenge for reliable costs estimations. Sections 2.1-2.4 provide background details about the main sector-coupling mechanisms treated in this thesis.

2.1. Loading constraints on local electricity grids

A dsHP is a relatively small load for the electricity distribution grid. The burden changes when there is an aggregated number of new installations. When considering the increasing demand for electricity [30], the risk of hitting the power grid capacity limits becomes a serious one. dsHPs are a proven suitable technology for single family houses. However, multi-apartments building’s residents look at this option for short term savings (monthly bill) compared to the DH tariff. While considering economic benefits, the technical limitations are easily forgotten, especially if they impact a different sector. Thus, assuming a growing number of dsHPs to be connected to the grid, a solution should be found to avoid power outage issues. The expansion of the power infrastructure requires expensive capital investments and a long construction time. Thus it is considered as a solution that will have an impact in the further future. This sector-coupling challenge has the potential of being mitigated instead by sectors’ integration approaches with impact in the near future. In this context, demand side management is a promising option. The control of the thermal mass in buildings and of V2G, described in the following sections, are relevant examples for shaving the heat and power demand peak. This can help reducing the use of typically expensive and polluting heat and power peak boilers.

2.2. Thermal mass in buildings

Demand side management offers a solution for adapting an energy demand to the supply rather than the opposite. This load shifting approach is typically implemented to shave an energy demand peak that would be instead covered with expensive and often polluting technologies. Depending on the technology, a complementary valley filling enables a load increase when either a base/medium load capacity or a renewables-based surplus is available.

Energy storage can perform this full load shifting task. A particular type of energy storage, still to be further explored, is the use of the thermal mass of the buildings [31]. This solution can be implemented to perform heat peak shaving, for instance, in combination with dsHPs and in integration with DH. This second case is taken as an example to describe the functioning of the concept.

The thermal inertia of the construction materials and of the interior furniture can shift the need for DH away from the peak level. This is conceptually illustrated in Figure 3 (a). The heat stored in the thermal mass can be discharged so that the use of heat only boilers (HOBs) can be reduced or even avoided, depending on the available potential. This makes the aggregation of more buildings, for example on a district level, an interesting solution. An example of control is based on a feedback loop between the indoor temperature and its set point regulation (Paper II - [20]). As a control strategy, the latter can be adjusted within a specific range (Tmin-Tmax) depending on the current indoor condition (Tindoor). A generally

(21)

accepted oscillation of the indoor temperature falls in between ±0.5 °C, around the comfort level of 20 °C. Larger temperature oscillations can cause complaints from the inhabitants, thus may require, for example, economic incentives for acceptance. Given the comfort level, the thermostat set point can be lowered to allow the discharge of the thermal mass to replace DH. When the indoor temperature falls below a pre-defined lower-end level (Tmin), the set point is raised to make use of DH. The HOBs are activated if the heat load if higher than the baseload level, as indicated by the comment on the left side of Figure 3 (a). If the heat load is lower than the baseload level, as illustrated in Figure 3 (b), the baseload plants (e.g. COG) are operated on a continuous nominal level to charge the thermal mass. In the best case where the thermal mass completely replaces the HOBs, the DH supply is operated at a constant nominal baseload level (as shown in the diagrams in Figure 3 (a) and (b)).

When coupled with dsHPs, the management of the buildings’ thermal mass can work in a similar way as in the DH case. The main difference is the objective, which becomes to alleviate the electricity grid from peak power demand, instead of replacing HOBs.

Figure 3 – Thermal mass (TM) in buildings as TES: discharge (a) and charge (b).

2.3. Vehicle-to-grid

V2G [32] is used as an electricity storage technology to provide ancillary services to the power grid. The interconnection can be established at different voltage levels depending on the required task and on the entity controlling the operation [33]. An option is related to the possibility of using V2G to peak shave the new electric loads. This solution can be used to alleviate potential grid overloadings, as the buildings’ thermal mass option, described in section 2.2. The conceptual illustration in Figure 4 shows this idea. In a residential district, when electric vehicles (EVs) are parked, the remaining electricity in their batteries can be discharged to the grid. The value of this solution is supported by an increasing share of EVs infiltrating the transport market [34] and by the awareness that cars, including EVs, are currently parked on average over 90-95% of their lifetime ([35],[36]). The implementation of V2G as electricity storage option also implies that the increased charging requirements should be taken into account. Tailored control strategies can optimize both charging and discharging schedules and loads to provide ancillary services for the power grid. From the point of view of the EVs, assuming that their future cost will allow a full market penetration,

(22)

barriers remain concerning the batteries degradation and disposal. The implementation of V2G will increase the weight of these challenges because of more frequent charging and discharging cycles. Extensive research work is active to solve these issues [37].

Figure 4 – V2G to peak shave the grid overloadings caused by domestic HPs (Paper V - [24])

2.4. Heat recovery

Heat recovery from local resources is a promising solution to reduce the carbon footprint of DH [38]. This concept implies the cooperation of two systems, which are the DH supply/distribution and a heat recovery source. Moreover, this source often involves interactions with other energy carriers. Examples are data centers, supermarkets and ice-rinks that can provide heat as a by-product of their main function, which is cooling.

Waste heat recovery (WHR) in low temperature DH is a particularly attractive option ([39,40]) for two main reasons. Firstly, it enables the use of resources that would be otherwise wasted. Secondly, a low temperature network technology allows the integration of a larger number of sources, without additional equipment. When the temperature level of this heat does not meet the DH network requirements, a HP is used as a temperature booster [41]. A conceptual illustration of an example of such a system is shown in Figure 5, where the focus is set on the SM component of Figure 2. The cooling circuit of the SM refrigeration system offers the possibility of recovering heat to satisfy the internal space heating (SH) and domestic hot water (DHW) demand and to supply DH. A HP guarantees that the supply temperature matches the network’s requirements. This further link with the electricity sector opens the possibility of extending the power-to-heat management of DH operators.

(23)

Figure 5 – WHR from a SM refrigeration system with a HP for temperature boosting to supply DH

Through the integration of large scale HPs, COGs and hot water tanks in the supply mix, DH utilities are already able to operate their resources by following the electricity spot market prices [42]. High electricity prices push the full-load operation of COGs’ turbines. Low electricity prices are linked with the activation of the HPs and of the direct condensers (turbines bypassed). The hot water tanks provide load shifting between the two modalities. Smaller scale HPs for heat recovery from local resources can play a similar role, assuming that their control can be coordinated with the central DH supply.

Heat recovery is not only about WHR. An example is offered by a synergy between DH and CO2 refrigeration systems with geothermal storage. This technology is today considered an interesting option in Scandinavia [43], for supermarkets that aims at fully covering their internal heat demand all year around. The geothermal storage is also used during summertime to provide sub-cooling and, thus, also to support the heat recharging of the ground. Since the cooling loads are higher than the heating ones, a spare heating capacity is available. This can be further used, during wintertime, to supply heat to a nearby-located DH network.

(24)

3. State of art

The aim of this chapter is to depict the state of art of planning and design frameworks for IUES. The first section 3.1 provides a general overview within the timeframe of the last five years. In section 3.2, the focus is set on the Swedish context, for which all scientific literature concerning IUES is included, from its beginning in 1992. The last section 3.3 clarifies how the framework presented in this thesis stands in comparison with the state of art.

3.1. General overview of planning frameworks

Several in-depth reviews from previous literature should be considered as complementary to the state of art presented in this section. With an overview on current practices, ref. [27] points out the need of involving the energy utilities from the very beginning of the planning process for IUES. The focus is on the electricity distribution grid in interaction with the heating, cooling and transport systems. A main conclusion is that a collaborative design process from earlier stages can help to integrate more synergies at a lower cost. The review in ref. [26] presents technologies and related modelling approaches for power-to-heat strategies. In particular, the different roles played by HPs and energy storage are discussed. It is concluded that the potential of power-to-heat to decarbonize the energy sector is high, especially if HPs and thermal storage are integrated. Power-to-heat is also part of the review in ref. [9], which extends the discussion to the gas sector. This paper provides a comprehensive description of technologies, modelling approaches, operational strategies and trends. A highlight, often neglected, concerns the challenges involved in analyzing the integration of systems with different physical, technical, spatial and temporal nature. By considering similar aspects, the review in ref. [29] focuses on the transport sector and on the technical limitations of the power sector. The dominant role of EVs’ batteries, HPs, photovoltaic panels (PVs) and power-to-heat management is highlighted. Ref. [10] offers a broad review of studies that deal with the trade-off of investments between the supply and the demand sides, the potential of DH to decarbonize the heating sector, multi-generation systems and industrial WHR. It is stated that the 100% share of renewables in the power sector could be achieved sooner than expected, even though technical limitations should be taken into account. An important contribution is provided by reviews about existing tools for modelling and simulation of IUES ([44], [45], [46], [47], [48], [49]). In the effort of representing the complexity of these systems, it is fundamental to be aware of already existing tools to avoid re-inventing the wheel. In particular, ref. [46] specifies which tools present co-simulation potential, while ref. [47] and ref. [48] focus on user-friendliness. Other reviews ([50], [51], [11]) point their attention on the methods implemented by the different tools. General conclusions [52] can be made about the already existing potential, however this is limited by remaining challenges. Major ones are the combination of different temporal and spatial details at district-to-city scale, in combination with reasonable computational times and reliability of the results.

The review presented in this section focuses on planning and design frameworks that take a perspective that is either technical, techno-economic, techno-environmental or techno-economic and environmental. This is also the first layer of classification adopted for the state of art frameworks, as indicated in Table 12 in Attachment 1. The second layer, also reported in Table 12, corresponds to the dominant modelling and simulation method of each framework. These methods, described in their dedicated following subsections, are:

- Co-simulation - Energy hubs

- EnergyPLAN and EnergyPRO - P-graph

- CitySim

- City Energy Analyst (CEA) - Life cycle analysis (LCA)

- National Institute of Standards and Technology approach (NIST)

- Equations systems, which stands for in-house frameworks based on systems of equations (energy balance)

(25)

- Mehods based on optimization algorithms, namely single objective (SO), mixed integer linear programming (MILP), mixed integer non-linear programming (MINLP) and multi-objective optimization (MOO)

- Hybrid, which stands for combinations of the above methods

The scope of this review includes sector-coupling approaches. In relation to this concept, a further classification of the state of art frameworks is shown in Table 13 in Attachment 1. A difference is made between the approaches that include technical performance of energy networks in their modelling level of detail.

The first column in Table 13 contains a level of detail for:

- supply of different types of energy carriers and, thus, different types of demand. Coupling technologies are the reason for interaction and impact;

- supply of one type of energy carrier, and thus one type of demand, with interaction between the two sides. This interaction is typically represented by distributed solutions, for example distributed supply (HPs, PVs and others) and storage technologies (water tanks, thermal mass of buildings and others)

The second column in Table 13 contains a level of detail for:

- interactions among networks of different types of energy carriers;

- interaction between supply, distribution and demand of one type of energy carrier.

Another important part of the scope of this review is the scale of analysis for the state of art frameworks. Table 14 in Attachment 1 divides the reviewed literature between district and city level scales. Table 15, in Attachment 1, reports specific districts and cities used as case studies in each reference.

In the following subsections, each reference is briefly commented and the main methods are introduced. This review does not include:

- frameworks that do not include the heating sector - frameworks based on long term scenarios.

3.1.1. Co-simulation – based

As a general definition, co-simulation can be described as an approach “to enable global simulation of

a coupled system via the composition of simulators” [53]. In other words, if two (or more) parts of a system

should be modelled in two (or more) simulation environments, co-simulation enables a coordinated simulation of these parts. Multiple simulation environments are needed when, within a single system, it is possible to identify parts that have different physical nature or require different levels of detail. By simulating these system’s parts, any feedback interaction among them can be captured. A general basic co-simulation is composed of [54]:

- Two or more simulators - A master algorithm

Each simulator is constituted by a model (or multiple models) and a solver. Within each domain, a simulator is usually characterized by an interface language that simplifies the implementation of a model. This is directly translated into the related programming language and the solution is computed by the solver. For example, the models can be implemented as:

- Algebraic equations - Differential equations - A mix of algebraic equations - Discrete events models - Finite elements models

(26)

- Behavioral models

The master algorithm is defined as the orchestrator of the co-simulation. This module is mainly responsible for coordinating the simulators’ time steps and enabling the exchange of variables among them. From the orchestrator perspective, each simulator is a black box with input and output variables to be connected to others simulators. In order to interface a simulator with the orchestrator, the functional mockup interface (FMI) is considered as the standard approach [54,55], [56], [57]. The connection can be implemented either on the same machine or on a communication network among different machines [54]. In relation to the field of district energy infrastructures, several simulators are required to model phenomena belonging to different sectors, like heat and electricity, supply and demand, control and sizing. In this context, the word “system” is often confused with “part”. For example, heat and electricity networks linked by a heat pump can be considered as a coupled systems that could be modelled in different environments.

In the literature, four main frameworks based on co-simulation could be identified, within the field of district energy infrastructures:

- the Orpheus project - the Mescos project - the ETH – EMPA project

- the net zero energy district framework

Within the Orpheus project, models of DH networks and thermal consumption in buildings are implemented in Dymola/Modelica and models of power grids are implemented in DigSILENT Power Factory. The proposed framework is characterized by a wide flexibility to different types of control algorithms, both rule-based and optimization-based ones. An example of the former is provided in [58] and in [59], where constrains are set on the operation of specific technologies, eg. PV panels surplus generation, electric boilers power, TES state of charge and distribution grid voltage levels. The latter is applied in [60], where a linear optimization program is used to minimize the heat production operational costs of a combined heat and electricity networks system. In [61], two optimal control strategies aim at phasing out a fossil fuel resource and minimize the electricity operational costs, respectively. These four studies demonstrate the presented approach by using a residential district in Ulm (Germany) and Skellefteå (Sweden) as case studies. The study in ref. [62] focuses on providing a link between design optimization and advanced optimal control. As co-simulation environment, Fumola is chosen to handle the time-stepping of the simulators and the data exchange by interfacing through the FMI standard.

The aim of the platform described in [63] is to design net zero energy urban districts. For this reason, specific control algorithms are implemented and coupled to detailed models of energy demand in buildings and energy distribution systems. In the study presented in [63], the electricity demand is modeled in a tool called URBANopt, while the model of the power grid and the overall simulation are implemented in OpenDSS. A feedback loop with the control algorithms is enabled by a co-simulation link with OpenDSS. The ownership of the generation and distribution systems can be modified by considering different business scenarios. In each case, decision support is provided through the analysis of pareto frontier generated by an integrated multi-objective optimizer. A district in Denver (Colorado, USA) is studied as a use case.

The co-simulation platform proposed by the Mescos project [64] combines NEPLAN for electricity network models, Matlab/Simulink coupled to IBM ILOG Optimization Studio for the control algorithms, SimulationX for the buildings. The approach is applied to an ideal test case with 146 buildings with different heating systems, heat storage units and PVs. Power consumption, voltage drop and indoor temperature levels are estimated with a focus on the platform computational performance.

The data and time synchronization is performed by a commercial software package called TLK Inter Software Connector (TISC), which offers flexible connections to different software programs. Further flexible interfaces are available to integrate simulators not supported by default. Moreover, the TISC

(27)

environment can handle distributed simulations on multiple computers for large scale applications. An ideal district located in Ruhrgebiet (Germany) is used as test case.

3.1.2. Energy hubs – based

An energy hub is defined as an interface between energy producers, consumers, and the transportation infrastructure [65,66], [67], [68], [69], [70]. An energy hub framework includes generation, conversion and storage technologies that handle different types of energy carriers [66]. Moreover, this approach explores the potential for synergies among these systems [66]. An energy hub unit (or interconnection of units) can correspond to systems with different scales. Within the context of this review, for example, the size of an energy hub can range from a single building to a whole city. The energy hub approach on its own enables a load flow analysis. This allows determining the steady state performance conditions at each node of the modeled system. In order to estimate performance indicators that support the design process of energy infrastructure, hubs-based load flow analysis are coupled with optimization techniques. In the literature, three main studies were found, which propose a comprehensive approach for the design of IUES by using an energy hub – based method.

Orehounig et al. [71] use an energy hub model to formulate an optimal supply dispatch problem, which is solved with the Matlab Optimization toolbox. A special feature of this framework is the combination with Energy Plus, an open source building energy performance simulation tool. This is used to generate hourly time steps curves that are given as inputs to the energy hub – optimization process. As a case study, a set of scenarios were developed for a Swiss mountain village with 29 buildings. These were used to determine the best combination of COGs, HPs, wind turbines, boilers and PVs. CO2 emissions and level of autonomy are used as main performance indicators.

Ayele et al. [72] proposes an extended energy hub approach to analyse the load flow performance of IUES. The energy hub coupling matrixes are linked to detailed thermo-hydraulic heat networks and electric power grid models. This approach is complemented by an energy system design optimization process, which is based on a nested particle swarm technique [73]. This method allows the determination of the optimal size and placement of different supply technologies. This was demonstrated with an ideal case involving HPs and HOBs and taking into account energy losses from connected heat and electricity networks. Operational costs are included in the analysis.

Liu et al. [74] developed a Matlab-Excel VBA tool to perform the load flow analysis of integrated heat, electricity and gas networks. The models of the networks and of the energy conversion technologies are implemented in different coupling matrixes. The overall load flow problem is solved by a Newton-Raphson method. The results are systematically presented by using Sankey diagrams that show how the supply and the demand mixes are coupled to each other. As a case study, networks details (e.g. voltage levels, electricity and heat losses, supply and return temperatures, gas pressure levels, gas losses), CO2 emissions and operational costs are assessed for the Campus of the University of Manchester.

3.1.3. EnergyPlan – based

EnergyPLAN is defined as a deterministic model that aims at estimating the consumption, environmental and economic impact of different energy system’s design scenarios [75]. The initial focus was on combined heat and power technologies and renewable resources integration. Later on, the application of the model has been expanded to several other technologies and to the integration of multi-energy grids [76]. EnergyPLAN is programmed in Delphi-Pascal and it is characterized by a user friendly tabs-interface [75]. The platform has been developed and expanded since 1999 [75].

As a simulation tool, EnergyPLAN runs analytical computations on an hourly step, for a time range of one year [75]. The simulations are based on priority lists of the supply units to balance the energy demand

(28)

[77]. The plants efficiencies are constant. EnergyPLAN is based on an input/output logic. Examples of main inputs that can be set by the user are:

 different types of energy demand

 different types of renewable energy sources  energy technologies capacities

 costs figures  regulation strategies

Limited market analysis are also possible. This is based on an optimization approach applied to the generation plants profits on a short term planning horizon [76]. EnergyPLAN is used in several studies. The ones discussed in the following part of this section were selected within the scope of this discussion. This means that each study is considered as a contribution in terms of methods for the design of IUES.

The framework proposed by De Luca et al. [78] is based on EnergyPLAN. Field measurement data are used as inputs for the performance of already existing energy generation and conversion technologies. For new systems, detailed models are implemented in the TRNSYS software, which is specifically linked to EnergyPLAN. The Italian city of Silentina is used as a case study to determine the mix of PVs and thermal solar panels, biogas COGs and wind turbines systems. The objective is to reduce the CO2 emissions by taking into account investment and operation costs. This approach is extended by Bonati et al. [79] with a sizing optimization algorithm that aims at maximizing the exergy performance of the integrated energy technologies. When no economic feasible solutions are found, suggestions are formulated in terms of incentives. The same environmental and economic indicators are estimated for a different Italian case study, Pompei.

The approach adopted by Brandoni et al. [80] aims at minimizing the fossil fuel consumption of an integrated energy infrastructure by implementing a dispatch optimization feature offered by the EnergyPlan model. The focus of a case study (Corinaldo, Italy) is on the impact of an increasing share of micro-generation, like PV panels and micro-COGs. CO2 emissions are assessed as main performance indicators.

Prina et al. [81] coupled EnergyPlan with a thermal model tailored to study the interactions among PV panels, large scale HPs and TES units. Two parallel approaches are proposed for performance simulation and analysis. By the first approach, several scenarios are explored in a deterministic way to compare different peak shaving solutions and save the excess electricity generation. Within the second approach, the PV-HP-TES model is linked to a system design optimizer, which is implemented as a multi-objective evolutionary algorithm. The objective is the minimization of both CO2 emissions and costs. The city of Bressanone-Brixten (Italy) is used as a case study.

Drysdale et al. [82] propose a systematic and comprehensive framework around EnergyPlan. The approach includes the development of structured templates for data collection and the definition of Smart Energy System Sustainability Factors. These indicators correspond to the integration of different energy sectors, technical, socio-economic and energy security feasibility, efficiency, resource utilization and environmental performance factors. Different design scenarios based on DH, PVs, solar collectors, wind turbines, HPs and storage units are tested for the case study of Sondeborg, in Denmark. The technical analysis is complemented by investment costs calculations.

In ref. [83], EnergyPlan is used to systematically compare the performance of an energy infrastructure under two main scenarios. On one hand, a traditional non-integrated approach is used. On the other hand, synergies are introduced to link the operation of cogeneration and concentrating solar plants, boilers, TES units, PV panels, geothermal HPs and batteries. The focus is the achievement of a 100% renewable energy infrastructure by a specific year. As a case study, the best achievable energy supply mix is assessed for Zagreb (Croatia).

(29)

The methods introduced in [77] and [84] involve another tool besides EnergyPlan, which is called EnergyPro [85]. This is a commercial-licensed software that is oriented towards a business-financial analysis of complex energy projects with combined supply of electricity and thermal energy. Østergaard et al. ([77]) propose a two-stages approach. Firstly, EnergyPlan is used to perform an energy balance and an economic evaluation of an integrated system (DH networks, COGs, HPs, TES units, wind turbines and PV panels). Sensitivity studies are applied on the size of specific technologies, for example TES units’ capacities. Secondly, EnergyPRO optimizes the design of the infrastructure from a business-financial perspective against and external electricity market. This is demonstrated for the case study of Samso (Denmark).

In another study [84], the same author limits the models implementation to the EnergyPRO environment. Algebraic models are integrated to couple the business-financial optimization with details about the technical performance of the system components (e.g. DH temperature levels and mass flow rates, power grid losses and primary energy consumption). The focus of this approach is to achieve a high temporal resolution for simulating the market participation mechanisms. This is shown for an ideal case study of 900 typical Danish houses.

3.1.4. P-graph – based

The Process Network Synthesis (PNS) approach was initially developed for the analysis of process technologies. Maier ([86]) applied it to the design of smart energy systems. The method is based on the utilization of special graphs, called process-graphs, which represent possible solutions in terms of mass and energy flows through networks. This is done by implementing combinatory rules that lead to multiple feasible scenarios. A dedicated software tool, PNS Studio, is available at [87]. This approach is complemented by an evaluation of the ecological life-cycle footprint of selected design options. CO2 emissions and ecological implications are estimated by using a specific analysis based on the Energetic Long-term Assessment of Settlement Structures (ELAS) tool. The ELAS calculator can be used online at [88]. The Reininghouse District in Graz (Austria) is used as a case study. Besides the environmental and ecological implications, annual costs (investments, operational and resources’ costs) are estimated for different combinations of heat, electricity and gas networks, COGs and HPs, solar collectors and photovoltaic systems.

3.1.5. LCA – based

Bartolozzi et al. [89] adopt a full life cycle perspective in order to evaluate the performance of different heating and cooling infrastructure scenarios. The approach is based on the steps defined by the standard ISO 14040–14044. Firstly, the goal of the analysis should be defined and the relevant data should be collected. Secondly, an impact assessment should be performed according to pre-defined criteria. Finally, the results should be interpreted and framed within recommendations. In order to cover this workflow, Bartolozzi et al. selected a combination of tools, which are the LCA software SimaPro 8.02 integrated with the Ecoinvent 3.0 database and the ILCD 2011 Midpoint method developed by the Joint Research Centre of the European Commission, for the impact assessment. For the case study of a neighbourhood in Tuscany (Italy), CO2 emissions are estimated for different combinations of DH, geothermal HPs, biomass systems, material processes and transport options.

3.1.6. NIST – based

Chaouachi et al. [90] developed a conceptual architecture to describe and analyse functionalities and communication among different actors and domains, involved in an energy infrastructure. This architecture is based on standards released by the National Institute of Standards and Technology (NIST). The aim of the documents provided by NIST is to enable interoperability functions within the new

A co-simulation based framework for the analysis of integrated urban energy systems: Lessons from a Swedish case study

Doctoral Thesis in Energy Technology

A co-simulation based framework for

the analysis of integrated urban

energy systems

Lessons from a Swedish case study

MONICA ARNAUDO

A co-simulation based framework for

the analysis of integrated urban

energy systems

Lessons from a Swedish case study

MONICA ARNAUDO

Abstract

Sammanfattning

Preface

Acknowledgements

Nomenclature

Contents

1. Introduction

1.1. Research questions and objectives

1.2. Appended papers

1.3. Thesis structure

2. Integrated urban energy systems

2.1. Loading constraints on local electricity grids

2.2. Thermal mass in buildings

2.3. Vehicle-to-grid

2.4. Heat recovery

3. State of art

3.1. General overview of planning frameworks