_____________________________________________ Self-sufficient energy systems: a scenario analysis of key design decisions for a small-scale agricultural community

Self-sufficient energy systems:

a scenario analysis of key design decisions for a small-scale

agricultural community

_____________________________________________

University of Groningen – Newcastle University

Master’s Thesis: Dual Degree MSc. Operations Management

Date: 9th December, 2019

Author: Sander A. Kok

Student number: S2975068 / B180623804 Tel-number: +31 657238983

E-mail: sander.a.kok@gmail.com Address: Slachthuisstraat 145, 9713MH

Groningen, The Netherlands

ACKNOWLEDGEMENTS

I would like to express my gratitude and appreciation to all parties involved throughout the whole thesis process, without whom this final product would not have been possible. First of all, the sponsoring firm DNV-GL in Groningen which has offered me the opportunity to conduct this study and develop myself professionally. In particular, I wish to thank my on-site supervisor, Harm Vlap, for his support and guidance. Secondly, I appreciate all the time and effort spent by my academic supervisors, Dr. Small and Dr. Land, for reviewing my work, challenging me, and providing me with valuable feedback. Last but not least, I wish to acknowledge the community of Dearsum, being the inspiration to this study and for which I hope the results may prove practical in their future endeavours. It truly has been a journey of learning and discovery, with its ups and downs, but eventually resulting in an experience I can look back at with ample satisfaction.

ABSTRACT

Community initiatives for energy self-sufficiency are emergent, as they may benefit the individual community and contribute to the contemporary climate debate. This is also the case in the small-scale agricultural community Dearsum. Their aspiration of becoming energy self-sufficient based on the renewable energy sources - biomass from manure and wind energy - acts as the inspiration to this study. The main challenge is to integrate all the varying design considerations into a self-sufficient and economically viable design. Guidance hereon is lacking in the energy- and sustainability literature. Therefore, this study aims to provide guidance to communities like Dearsum and fill the existing gap in the literature, by analysing the ‘key design decisions’ for energy self-sufficient systems. A simulation approach is conducted, aimed at matching the supply of the renewable energy sources to the demand of a small-scale community. Inherent with this operational challenge is the application of buffers for which a grid connection, and hydrogen as an energy carrier, are applied. Based on feasible configurations of the renewables and buffer options, twelve unique scenarios are developed. Heuristics further determine the infrastructure and priorities of each scenario. These scenarios are modelled in MS-Excel on an hourly basis for a single year, yielding a performance economically and in the degree of self-sufficiency. Further experiments, altering the proportions of supply and demand, are conducted. The results show that a well-informed decision must be made on the application of biomass and wind tailored to the context-dependent demand. Balance is key. Combinations of biomass and wind with a buffer work best. Buffering through the grid is preferable, with hydrogen showing future potential. All in all, this research provides foundations for designing energy self-sufficient systems.

Keywords:

TABLE OF CONTENT

PART I: INTRODUCTION ... 1

1.1 Research Aim ... 2

PART II: THEORETICAL FRAMEWORK ... 3

2.1 Small-scale Agricultural Communities ... 3

2.2 Energy Self-sufficiency ... 3

2.3 Renewable Energy Sources ... 4

2.3.1 biomass. ... 5 2.3.2 wind. ... 6 2.4 Buffer Options ... 7 2.4.1 hydrogen. ... 7 2.4.2 grid connection. ... 8 2.4.3 heat storage. ... 8 2.5 Theoretical Framework ... 8

PART III: METHODOLOGY ...10

3.1 Methods ...10

3.1.1 simulation. ...10

3.1.2 scenario analysis. ...10

3.2 Problem and Scope ...11

3.3 Scenarios Development ...12

3.3.1 Heuristics. ...12

Scenario 11: biomass, wind and hydrogen. ...13

Scenario 12: biomass, wind, hydrogen and grid. ...14

3.4 Data Collection ...15

PART IV: MODEL FORMULATION ...16

4.1 Simulation Model ...16

4.2 Model Inputs ...16

4.2.1 demand profiles. ...16

4.2.2 supply profiles. ...18

4.3 Assumptions & Parameters ...19

4.3.1 assumptions. ...19

4.3.2 parameters. ...19

4.4 Model Rationale ...21

4.5 Model Outputs ...23

PART V: RESULTS ...26

5.1 Biomass Results ...27

5.2 Wind Results ...28

5.3 Biomass & Wind Results ...30

PART VI: DISCUSSION ...33

6.1 Implications ...33

6.1.1 biomass implications...33

6.1.2 wind implications. ...33

6.1.3 hydrogen implications...34

6.1.4 grid connection implications. ...34

6.2 Limitations ...34

PART VII: CONCLUSION ...35

REFERENCES ...36

APPENDICES ...39

Appendix A: Remaining Scenarios ...39

Scenario 1: biomass...39

Scenario 2: wind. ...39

Scenario 3: biomass and wind. ...40

Scenario 4: biomass and grid. ...41

Scenario 5: wind and grid. ...41

Scenario 6: biomass and hydrogen. ...42

Scenario 7: wind and hydrogen. ...43

Scenario 8: biomass, hydrogen and grid. ...43

Scenario 9: wind, hydrogen and grid. ...44

Scenario 10: biomass, wind and grid. ...45

Appendix B: Assumptions ...46

Appendix C: Parameters ...48

Appendix D: Experimental Results ...49

D.1: Biomass scenarios. ...49

D.2: Wind scenarios. ...50

1

PART I: INTRODUCTION

The targets set in the Paris Climate Accord, signed by 195 countries, essentially entails that global warming must be kept under a 2°C rise and that an effort must be made to keep this rise under 1.5°C, compared to pre-industrial levels (UN, 2018). These ambitious targets are partially responsible for the drive towards self-sufficient renewable energy systems (Boon & Dieperink, 2014; Rae & Bradley, 2012). This concept is often referred to as energy self-sufficiency and can be described as the capability to cover energy needs using local energy sources (Engelken, Römer, Drescher & Welpe, 2016). This is particularly suitable for small-scale agricultural communities, as access to sufficient land and resources are available for renewable energy technologies (Nakata, Kubo & Lamont, 2005). Moreover, various factors exist that could explain the emergence of self-sufficient renewable energy systems. This includes, but is not limited to: increased environmental awareness, (fluctuating) energy prices, inconsistent policies, governmental incompetence, independency from energy suppliers, improved technological capabilities, social cohesion and a green image (Boon & Dieperink, 2014; Rae & Bradley, 2012). In addition, benefits that local communities could attain with the implementation of self-sufficient renewable energy systems include: a local source of income, local management and control of the system and a cheap and reliable energy supply (Walker, 2008). Within the Netherlands, this development has the potential of contributing up to forty percent of national electricity demand (Boon & Dieperink, 2014). This may prove especially relevant to the Netherlands, as the share of renewable energy in the final energy consumption is progressing slowly. Namely, a 6.6% share in 2017, whilst having a target set of 14% in 2020 (Eurostat, 2019). Hence, considering the climate debate, its consequent drive towards self-sufficient renewable energy systems and its societal benefits, makes it relevant to study this topic in the context of small-scale agricultural communities.

2 Academically, this research topic belongs to a relatively shallow literature base. Prior research has mostly been qualitative and has mainly focussed on exploring the drivers (Boon & Dieperink, 2014; Engelken et al., 2016; Walker, 2008), requirements (McKenna, 2018; Rae & Bradley, 2012), implications (Van Der Schoor & Scholtens, 2015) and framework development (Mckenna, Bertsch, Mainzer & Fichtner, 2018; Müller, Stämpfli, Dold & Hammer, 2011) for self-sufficient renewable energy systems. To a lesser extent, research has been done in quantitatively modelling and designing such systems (Nakata et al., 2005; Kanase-Patil, Saini & Sharma, 2010). These quantitative studies have been conducted to economically optimize energy systems for their specific contexts and related inputs. In contrast, this study aims to provide guidance on designing energy self-sufficient systems, whilst being economically viable. Basically, adhering to the trade-off between energy self-sufficiency and economic viability. Inherent herewith, is the operational challenge of matching the generally intermittent supply of RESs with the demand. However, guidance on designing such self-sufficient energy systems is lacking in the energy- and sustainability literature. Therefore, the academic novelty, or ‘gap’, lies in providing guidance by examining the KDDs for such systems from an operations perspective.

1.1 Research Aim

The aim of this study is to lay the foundations for designing feasible energy systems, applying a combination of wind and biomass. This entails analysing what the KDDs are in designing an energy system for a (nearly) self-sufficient community. In addition, this entails investigating how these KDDs influence the performance of these systems. To clarify, the performance measures consist of the degree of self-sufficiency (DSS) and the economic performance of these systems. The research question that is derived from this, is as follows:

‘What are the key design decisions for energy systems based on biomass and wind energy and how do they influence the degree of self-sufficiency and economic viability?’

3

PART II: THEORETICAL FRAMEWORK

As mentioned in the introduction, this study follows a simulation approach, orientated around the problem of Dearsum. That is, designing a self-sufficient energy system based on the combination of biomass and wind. Therefore, this section addresses the theoretical background on these two RESs and their corresponding considerations. However, the concepts ‘small-scale agricultural communities’ and ‘energy self-sufficiency’ are defined and addressed first. Furthermore, the means of energy storage is discussed. It must be noted that the literature is applied for background reasons, rather than developing a framework for empirical testing. Nevertheless, this section attempts to integrate the comprehensive aspects of energy self-sufficient systems into a theoretical framework.

2.1 Small-scale Agricultural Communities

Community is a broad and ambiguous concept. Therefore, it is necessary to clearly define what is meant thereby, regarding this study. Many definitions of community exist and are commonly associated with a sense of place, identity, localism and shared values (Rae & Bradley, 2012; Walker, 2008). In addition, Rae & Bradley (2012) state that community as a scale is highly appropriate to acts towards sustainability issues. Applying community as a level-of-scale is recurring in the context of energy self-sufficiency (Rae & Bradley, 2012; Van Der Schoor & Scholtens, 2015; Walker, 2008) and similarly in rural settings (Kanase-Patil et al., 2010; Nakata et al., 2005). Forrest & Wiek (2015) characterize ‘small-scale’ communities by a population of fewer than ten thousand people, in their sustainability study. Thus, within the context of this study, small-scale agricultural communities are defined as areas in rural/agricultural settings, with fewer than ten thousand local inhabitants and who have shared values to act towards their common goal of energy self-sufficiency.

2.2 Energy Self-sufficiency

4 As mentioned before, most prior research into energy self-sufficiency has been qualitative of nature. These studies revolved around the drivers (Boon & Dieperink, 2014; Engelken et al., 2016; Walker, 2008), requirements (McKenna, 2018; Rae & Bradley, 2012), implications (Van Der Schoor & Scholtens, 2015) and framework development (Mckenna et al., 2018; Müller et al., 2011) for energy self-sufficient systems. Quantitative research has also been done in modelling such systems (Kanase-Patil et al., 2010; Nakata et al., 2005). Hereby, Nakata et al. (2005) looked at economically optimizing the design of a renewable energy system in rural Japan, which included biomass and wind. Likewise, Kanase-Patil et al. (2010) looked at optimizing energy systems in rural India for four scenarios with differing RESs configurations. However, the primary aim of this study is to analyse the KDDs for energy self-sufficiency, whilst adhering to economic viability. Therefore, these studies, although contextually different, may provide valuable insights.

2.3 Renewable Energy Sources

There exists a limited set of RESs and applications in the world as we know it. Panwar, Kaushik & Kothari (2011) define RESs as “resources which can be used to produce energy again and again”. In addition, they state that RESs have the potential to provide energy with (near) zero emissions of greenhouse gases and air pollutants. Table 2.1 provides an overview of these RESs and their applications (Panwar et al., 2011). According to Liserre, Sauter & Hung (2016), the most exploited of these are hydro-, solar- and wind energy.

Table 2.1: Overview of RESs and their application

5

2.3.1 biomass.

Biomass is a plentiful and prominent RES, and is considered an attractive option for energy systems (Dong, Liu & Riffat, 2009). The underlying reason for this is that biomass has a stable energy supply compared to other RES, like wind and solar, which are intermittent of nature. Biomass consists of a variety of raw materials, including organic wastes from industry and households, dedicated crops and agricultural wastes like manure (Panwar et al., 2011). However, within the scope of this study manure from livestock is the designated form of biomass. The manure is fed into a biodigester installation, from which biogas is obtained through anaerobic digestion (Nakata et al., 2005; Panwar et al., 2011). Subsequently, biogas may be applied for various purposes. Firstly, biogas could be used to supply the heat demand directly (Panwar et al., 2011). In addition, biogas may be applied to generate both heat and electricity through a combined heat and power (CHP) engine (Dong et al., 2009; Panwar et al., 2011). According to Dong et al. (2009) applying biogas with a CHP-application is used extensively, and is best suited for small-scale systems. Lastly, biogas may be upgraded to natural gas quality (green gas), which enables the benefit of distribution through the existing natural gas infrastructure (Dong et al., 2009; Panwar et al., 2011). Figure 2.1, adapted from Panwar et al. (2011), visualises the relevant applications.

Figure 2.1: Overview of biomass applications

6

2.3.2 wind.

Wind energy is a widely used, mature, competitive and nearly carbon-neutral RES in contemporary life (Panwar et al., 2011). Figure 2.2 illustrates how electricity is generated from wind energy through wind turbines. In addition, Panwar et al. (2011) state that wind power may prove practical for small power needs in isolated areas, which is also relevant for small-scale agricultural communities aiming to become energy self-sufficient. An inherent issue with wind turbines is the NIMBY (not-in-my-backyard) syndrome, experienced by communities. Basically, this is the resistance of communities toward unwelcome developments in their neighbourhood (Dear, 1992). In accordance with Walker (2008), this indicates how relevant it is that the community is involved in the decision and benefits from the wind turbine. Rae & Bradley (2012) state that resistance to wind turbines is less prevalent if it is known that it benefits the local community. Various articles imply that wind power is of utmost importance and beneficial for energy self-sufficient communities (Li, Birmele, Shaich & Konold, 2013; Nakata et al., 2005) In addition, Walker (2008) states that a community-owned wind turbine is economically viable, generates local income and electricity and maintains local control. Thus, wind energy is a relevant RES to be considered in the context of small-scale agricultural communities.

However, the application of wind energy comes with a set of considerations. Firstly, it has to be considered that the availability of the wind resource is intermittent of nature (Nakata et al.,2005; Taylor, 2009). This could lead to flexibility issues for energy systems (Nakata et al., 2005). For this reason, Panwar et al. (2011) suggest that wind should be used in conjunction with other energy sources. This supports the combination with biomass, which may be used to cover energy deficits. Conversely, storage is a consideration to cover excess energy and supply deficits (Rae & Bradley, 2012). Secondly, the size and capacity of a wind turbine is a design decision (Nakata et al., 2005; Taylor, 2009). This depends on the energy needs and the wind resource availability (Taylor, 2009). Furthermore, the economic performance of a wind turbine has to be considered (Nakata et al., 2005; Taylor, 2009). Lastly, policies pose a consideration as they may incentivize (Taylor, 2009) or even prohibit local wind turbines, as is the case in the Netherlands (Van Der Schoor & Scholtens, 2015).

7 2.4 Buffer Options

Buffering is an important consideration in the design of energy systems that apply intermittent renewable energy sources, like wind (Rae & Bradley, 2012). This aims to divert energy demand and -supply mismatches. For one, this can be done by storing the energy. Energy storage involves capturing and storing energy when there is an energy surplus, and for usage when there is an energy deficit (Rae & Bradley, 2012). Typically, batteries are used for energy storage. However, converting energy to the energy carrier hydrogen (H2) is a storage option with multiple additional applications (Rae & Bradley, 2012). Furthermore, the decision of having a grid connection rather than being stand-alone could act as a means of energy storage (Nakata et al., 2005; Rae & Bradley, 2012). These will be discussed next.

2.4.1 hydrogen.

“Hydrogen is expected to play a key role in the world’s energy future by replacing fossil fuels” (Panwar et al., 2011). Namely, converting hydrogen to energy is clean and generates no pollutants, except water (Dunn, 2002). However, even though hydrogen is the most common element on Earth, it does not exist naturally, and therefore, must be created (Panwar et al., 2011). Figure 2.3, adapted from Ulleberg, Nakken & Eté (2010), shows the structure of the production, storage and utilization of hydrogen as an energy carrier. This image shows wind energy as input, but also other RESs like biomass or solar may act as input to produce hydrogen through electrolysis of water (Dunn, 2002; Panwar et

8

2.4.2 grid connection.

Having a grid connection, or not, may prove an important design decision for energy systems. Although, applying a grid connection does not constitute absolute energy self-sufficiency (Engelken et al., 2016), it may be beneficial. This includes the possibility to use the grid as a buffer for covering energy supply deficits and for offloading excess energy (Nakata et al., 2005; Rae & Bradley, 2012). Basically, it is a means of balancing energy supply with demand. Herewith, the grid may complement the application of RESs by mitigating the intermittent nature hereof (Nakata et al., 2005). Moreover, a grid connection may prove financially cheaper than the other storage options, and selling excess energy could generate an additional income source (Rae & Bradley, 2012). On the other hand, not having a grid connection has its benefits, as outlined in the introduction. Therefore, it is important to consider the design decision of applying a grid connection or not, and weighing the benefits hereof.

2.4.3 heat storage.

Aside from the two main buffer options, a heating grid and hot water storage tanks (HWSTs) are considered as storage medium for heat. The HWSTs are commonly used for tackling mismatches in supply and demand (Celador, Odriozola & Sala, 2011). Herewith, electricity is converted to thermal energy and stored in cylindrical tanks at the residential level. Important consideration hereof, is the storage size (Celador et al., 2011). In addition, a heating grid is considered for transportation and storage of heat, which may receive feed-in from heat through biogas or hydrogen (Nastasi & Lo Basso, 2016).

2.5 Theoretical Framework

This chapter has outlined the theoretical background on energy self-sufficient systems, based on the applicable RESs and buffer options. However, it is key to integrate these elements and its considerations into a theoretical framework. Therefore, table 2.2 summarizes the discussed considerations associated with the implementation of the RESs and buffer options. Recurring themes are the various applications, size and capacity, and the associated costs hereof.

9 In addition, a conceptual model of a small-scale agricultural community has been developed to give an impression of the interplay between the RESs and the buffering options. Figure 2.4 visualises this.

10

PART III: METHODOLOGY

3.1 Methods

3.1.1 simulation.

As this study aims to lay the foundations for designing energy self-sufficient systems based on the renewable energy sources (RESs) biomass and wind, and the buffer options grid connection and hydrogen, simulation is performed as the main method. “Simulation is the process of designing a model of a real or imagined system and conducting experiments with that model” (Roger, 1999). According to Smith (1999), simulation is applied in many scientific and practical fields, including the design of future systems, and allows for the analysis of a system’s behaviour without requiring a real-life system. Simulations are therefore appropriate for analysing the operations of future, potentially self-sufficient, energy systems based on biomass and wind energy. The simulation development process outlined by Smith (1999) is adapted in conjunction with scenario analysis, as explained below, and used to further structure this study.

3.1.2 scenario analysis.

In conjunction with the described simulation method above, a scenario analysis is conducted. Scenarios are “a set of hypothetical events set in the future constructed to clarify a possible chain of causal events as well as their decision points” (Amer, Daim & Jetter, 2013). Moreover, Amer et al. (2013) state that scenarios help with holistic future planning, uncertainties and identifying decisions. Additionally, scenario analysis is a means of simplifying large amounts of data and captures different states in which several variables can change at a time, and through which it is possible to identify patterns among the endless number of outcomes (Schoemaker, 1995). Hence, performing scenario analysis provides a powerful tool for examining a range of future possibilities, which is argued to be challenging in the sustainability context (Swart, Raskin & Robinson, 2004). Consequently, this makes the application of scenario analysis fitting for this study. Namely, as outlined in the theoretical framework, many possibilities for designing self-sufficient energy systems exist with the RESs biomass and wind, the buffer options grid and hydrogen, and all other related components and routes. The steps of scenario analysis as discussed by Schoemaker (1995) partially coincide with the steps by Smith (1999). Therefore, the steps are adapted, as shown in figure 3.1, and adhered to in further discussions.

11 3.2 Problem and Scope

The first step in the adapted simulation process is to define the problem and scope. As mentioned before, the problem originates from the real-life village Dearsum, which acts as the underlying case for this research. This small-scale agricultural community aspires to become self-sufficient in their own energy supply. Herewith, they act as a pilot for other communities with the same purpose. This community consists of approximately 125 inhabitants, including four farmers with roughly 750 dairy cows, located on a surface of 445 hectares (AlleCijfers, 2019). Dearsum had prior experience with some means of energy self-sufficiency, including a wind turbine and a biodigester (manure mono-fermentation). These had become outdated and decommissioned ever since. Nowadays, their ambition of becoming energy self-sufficient has revived. This is mainly motivated by their awareness of climate change and the Dutch Government’s decision to phase out the natural gas supply. However, it is important to note that their ambition of becoming energy self-sufficient is not unconditional. Namely, self-sufficiency should come with the least amount of effort or costs for the inhabitants, and with a (nearly) full energy supply.

12 3.3 Scenarios Development

The second step in the adapted simulation process is the development of scenarios. For this purpose, inspiration is acquired from the study of Kanase-Patil et al. (2010). They determined their scenarios based on different configurations of the RESs applicable within the context of their study. Within the scope of this study the configurations are based on the RESs (Biomass, Wind) and buffer options (Grid, Hydrogen). Hence, applying this method as a full-factorial design, leads to sixteen (2^4) possible combinations. However, several configurations may be omitted beforehand, as they are unrealistic or unfeasible. Firstly, a configuration where none of the four elements are applied is unrealistic, as no energy would be supplied to the community. Additionally, a scenario with merely a grid connection, even though possible and resembling the current state, is omitted, as no internal energy is produced. Also, a configuration with merely hydrogen is impossible, as hydrogen needs a source from which to be produced. Lastly, a scenario applying hydrogen and a grid connection is unfeasible and unrealistic. In this case hydrogen would be produced from externally produced energy, which encompasses efficiency losses and a significant rise in costs, with no additional income compared to simply drawing energy from the grid. Consequently, the twelve remaining scenarios are visualised in table 3.1 below.

Table 3.1: configuration-based scenarios

3.3.1 Heuristics.

13

Scenario 11: biomass, wind and hydrogen.

This is an off-grid scenario applying both RESs wind and biomass for the generation of energy and buffers through the application of hydrogen. Therefore, applying the heuristics yields the infrastructure visualised in figure 3.2. (1) biomass (manure) digestion is used to produce biogas, which is applied for heat demand directly through a biogas boiler. Simultaneously, a wind turbine supplies directly to the electricity demand. (2) Any excess biogas is fed into a CHP-engine, producing electricity and heat. The heat is assumed lost in this case but could be used for powering the digestion installation or for other purposes. The generated electricity may be used to cover electricity supply shortages from the wind energy. Concurrently, wind energy overages are converted to hydrogen through a P2G electrolyser. (3) Likewise, overages of electricity from the CHP-engine are converted to hydrogen, which is stored in a H2-tank. (4) Hydrogen is used to supply any unsatisfied heat demand through a hydrogen boiler, meaning households require two specific boilers. (5) Lastly, unsatisfied electricity demand is covered by hydrogen through a G2P fuel cell.

14

Scenario 12: biomass, wind, hydrogen and grid.

This scenario applies a full configuration of the RESs, biomass and wind, and the buffer options, hydrogen and a grid connection. In this case hydrogen is not used for buffering, but together with CO2 converted to green gas through methanation (Vo et al., 2018). Following the heuristics yields the infrastructure visualised in figure 3.3. (1) Biogas is upgraded to green gas and used to supply the heat demand directly. Concurrently, generated wind energy supplies the electricity demand directly. (2) Overages of green gas are fed into the natural gas grid. Additionally, overages of electricity are converted to hydrogen through an electrolyser and combined with CO2 in a specialised bioreactor to produce green gas, which is fed into the gas grid. (3) If any constraints of the electrolyser, bioreactor or CO2 captured from upgrading to green gas, are met, remaining electricity is fed into the electricity grid, which acts as a buffer for supplying unsatisfied demand later in time. (4) Lastly, any overages of green gas and electricity are sold off at the end of the year. It must be noted, that in this case an exception is made for following the heuristics. Otherwise, this scenario would have applied biogas directly for heat, rather than upgrading first, requiring additional investments for dual-boilers. In addition, following the heuristics would have led to this scenario becoming scenario 10, as electricity overages would be offloaded to the grid, rather than being converted to hydrogen.

15 3.4 Data Collection

The next step in the adapted simulation process is to collect the required input data (Roger, 1999). The data collected and applied within this study consists primarily of secondary data. Table 3.2 provides an overview of the types of data collected and the sources from which it is obtained. The demand profiles of both gas- and electricity consumption of Dearsum in 2017, has been provided by DNV-GL in Groningen, which is the responsible consulting party to the energy project of Dearsum. DNV-GL has previously executed collection and -analysis of relevant data. The available secondary data is concerned with the context-specific parameters of Dearsum, energy demand, cost estimations and energetic values and -efficiencies. In addition, DNV-GL has provided valuable insights regarding energy systems in general. Furthermore, a consultant of Energy United, an energy service provider, has shared valuable insights into the business case of biomass and -gas installations. The various other data are acquired from sources that are openly available.

16

PART IV: MODEL FORMULATION

4.1 Simulation Model

Now that the first steps of the adapted simulation process have been discussed, the next step is to construct the simulation model. This includes the input data, assumptions and parameters, mathematical formulations or rationale and output data, which may be processed by a computer software program. For this study the program MS-Excel has been applied to model the twelve unique scenarios, conform the pre-determined infrastructures and priority rulesets. MS-Excel is a well-known and widely used spreadsheet software. It provides a clear and structured manner of simulating, through referencing to cells and applying mathematical formulas and conditions. Herewith, a unique spreadsheet is constructed for each scenario and coupled to an input sheet containing all the input data and parameters. Each spreadsheet shows the applied RESs, components, related parameters, the hourly energy flows and the final performance.

By sequentially following the defined priority rules of a scenario, the energy flows may be constructed. This is done for all the 8760 hours of a single year. For each hour the energy flows in kilowatt hours (kWh) from supply to demand are simulated. Naturally, the flows start at the initial energy production of the RESs, flows through the relevant components, incurring efficiency losses on the way, up until the supplied demand at the end of that hour. Summing the hourly flows results in a differing technical performance, income, capital expenditures (CAPEX) and operational expenditures (OPEX) per scenario. This is then translated to a certain performance for each scenario, economically and in the degree of self-sufficiency (DSS), by which the scenarios may be compared individually and with each other.

4.2 Model Inputs

In essence, the inputs of the simulation model consist of the hourly energy demand and supply profiles. This includes the demand profiles for both electricity and heat, which represent the demand pattern for standard Dutch households, and the supply profiles for wind and biomass energy. The demand and wind energy profiles are taken from the year 2017, whilst the biomass supply profile is consistent across multiple years. The reason for applying merely 2017 is that the data is unavailable for more recent years, due to changes in privacy regulation.

4.2.1 demand profiles.

17

Figure 4.1: heat demand profile

On the other hand, the demand profile for electricity, depicted in figure 4.2, shows less volatility in variation over the full year, compared to the heat demand. However, there seems to be variation on the daily level. This may be attributed to day and night, where during the days and evenings more electricity is demanded than during the night, when people typically are asleep. In addition, there appears to be an outlier in the electricity demand profile, where the demand is exactly zero. Even though this might be an error in the demand data, it is decided not to eliminate this. It may have been caused by a blackout and the effect of this datapoint on the system’s overall performance is negligible.

Figure 4.2: electricity demand profile

jan feb mrt apr mei jun jul aug sep okt nov dec

Ho u rly d em an d

Heat Demand Profile (2017)

Gas consumption

jan feb mrt apr mei jun jul aug sep okt nov dec

Ho u rly d em an d

Electricity Demand Profile (2017)

18

4.2.2 supply profiles.

The supply profile of biomass energy, based on the input of manure, is assumed to be generally constant over the year. It is assumed that with the same number of cows the production of manure will not differ more or less over the year. However, as depicted in figure 4.3, the grazing season has been taken into account. Herein, the cows graze in the fields and all the manure is considered lost. This has an assumed duration of four months in which the cows graze for six hours per day. Hence, resulting in a 25% loss of manure per day during those four months, which approximates an 8.33% annual loss.

Figure 4.3: biomass energy supply profiles

In contrast, the profile of wind energy in figure 4.4 shows the exact opposite of the biomass energy supply profile. Herein, unpredictable peaks and troughs are seen, ranging from large peaks to null, throughout the year. Naturally, this entails that to match demand with supply, tactics have to be applied. Within the context of this study, this includes buffering when supply overages exist to cover for when shortages arise. Other tactics do exist, but these lie outside the scope of this research.

Figure 4.4: wind energy supply profile

jan feb mrt apr mei jun jul aug sep okt nov dec

Ho u rly s u p p ly

Biomass Energy Supply Profile

With grazing Without grazing

jan feb mrt apr mei jun jul aug sep okt nov dec

Ho u rly s u p p ly

Wind Energy Supply Profile (2017)

19 4.3 Assumptions & Parameters

Various assumptions and parameters are applied in the simulated models, as is inherent to simulation. The most important of which will be presented. All the remaining assumptions and parameters, including the following, are presented in Appendix B and C, respectively.

4.3.1 assumptions.

General:

➢ The reference year is 2017.

➢ Energy calculations are exclusively performed in kWh. ➢ Financial calculations are performed in Euros (€). ➢ Interests and net present value unconsidered. ➢ Asset-lifecycles and downtime unconsidered.

➢ If the energy system is overloaded, the initial production is reduced to equilibrium. Context-specific:

➢ No distinction is made between the demand profiles of houses and farms. ➢ Biomass and wind energy are the only applicable RESs.

➢ Households own a natural gas boiler. Renewable energy sources:

➢ Average wind speed ≥8.0 m/s for Dearsum (RVO, 2017).

➢ There exists no policy or resistance against the size of the wind turbine. ➢ Biomass is purely revolved around manure mono-fermentation.

➢ A central biodigester is applied, rather than multiple decentral biodigesters. Hydrogen:

➢ The ratio H2 to CO2 for the Sabatier-reaction is 4:1 (m3). ➢ Hydrogen is not applied for mobility applications. Grid connection:

➢ Energy may (dis)charged from and to the grid unconstrained.

➢ Netting arrangement is applied for selling/buying kWh at the end of the year.

4.3.2 parameters.

20 Regarding the conversion efficiencies, the relevant percentage is multiplied with the energy flow in kWh that is to be converted, resulting in the converted energy flow. Similarly, the storage efficiencies are applied to calculate the storage amount in time t, by multiplying the relevant efficiency with the storage amount in kWh at time t-1.

The other parameters presented in Appendix C consist of parameters on prices, RESs characteristics, energetic values and the costs of components. Concerning the prices, it is important to note that the selling price per kWh, for both electricity and gas, is deduced by taking half the buying price. This lower selling price accounts for the assumed netting arrangement not being applicable in contextual settings besides the Netherlands. Also, this measure is relevant for the Netherlands as the netting arrangement will be phased out in coming years. Moreover, the buying price is what the end-consumers pay and what the external energy supplier receives, and vice versa for the selling price. The subsidy prices are multiplied with the theoretical capacity (kWh) of the related component. This equals the component capacity (kW) times the full load hours. Regarding the applied cost-related parameters, it is important to note that the CAPEX equals the initial investment, whilst the OPEX encompasses the annual costs for operations and maintenance. These values are linked to the capacity of the components. However, certain capacities will be altered for further analysis. Naturally, this entails a different CAPEX and OPEX for the altered unit. In order to estimate these costs, either the capacity is multiplied by the price per unit capacity, or the six-tenth rule from Whitesides (2012) is adopted. This rule estimates the new cost by multiplying the known cost with the ratio between the new and old capacity to the power six-tenths, as explained below. This rule will be applied for calculating both the new CAPEX as well as OPEX.

𝐶2 = 𝐶1 ∗ (𝑆2 𝑆1)

0.6

(1)

Where: C1 the known cost (€) of capacity S1 (kW)

21 4.4 Model Rationale

The simulation model rationale will be explained on an hourly basis, following the indices, parameters, variables and components listed in table 4.4. The indices consist of the eighteen components applied across all the scenarios, and the time horizon. Moreover, the parameters consist of the energy generation, -demand, -efficiencies and capacities. The variables in this model are the hourly energy flows between the components and the energy storage levels.

Table 4.4: indices, parameters, variables and components

22 (1) Firstly, it is checked whether the produced energy is able to supply its appropriate demand. More specifically, biogas for heat demand and wind energy for electricity demand.

𝑿𝟏,𝟖,𝒕 = 𝐻𝑡 IF (𝐸𝐶8· 𝐵𝑡 ≥ 𝐻𝑡) 𝑿𝟐,𝟏𝟖,𝒕 = 𝐸𝑡 IF (𝑊𝑡≥ 𝐸𝑡) = 𝐵𝑡 IF (𝐸𝐶8· 𝐵𝑡 < 𝐻𝑡) = 𝑊𝑡 IF (𝑊𝑡< 𝐸𝑡)

(2) Secondly, overages of biogas are fed into a CHP-engine, which is used to cover electricity shortages, whilst the generated heat is assumed lost. Also, overages of electricity are converted to hydrogen. 𝑿𝟏,𝟏𝟖,𝒕 = 𝐸𝑡− 𝑊𝑡 IF (𝐸𝑡− 𝑊𝑡 ≤ 𝐸𝐶4· (𝐵𝑡− 𝐻𝑡) ≤ 𝐺4) = 𝐸𝐶4· (𝐵𝑡− 𝐻𝑡) IF (𝐸𝐶4· (𝐵𝑡− 𝐻𝑡) < 𝐸𝑡− 𝑊𝑡 ≤ 𝐺4) = 𝐺4 IF (𝐸𝐶4· (𝐵𝑡− 𝐻𝑡) ≥ 𝐸𝑡− 𝑊𝑡 > 𝐺4) Heat Loss = (1 − 𝐸𝐶4) · (𝐵𝑡− 𝐻𝑡) 𝑿𝟐,𝟏𝟑,𝒕 = 𝐸𝐶5· (𝑊𝑡− 𝐸𝑡) IF (𝐸𝐶5· (𝑊𝑡− 𝐸𝑡) ≤ 𝐺5) = 𝐺5 IF (𝐸𝐶5· (𝑊𝑡− 𝐸𝑡) > 𝐺5)

(3) Thirdly, the overages of electricity from the CHP-engine are converted to hydrogen and stored in a hydrogen tank.

𝑿𝟒,𝟏𝟑,𝒕 = 𝐸𝐶5· (𝐵𝑡− 𝑋1,8,𝑡− 𝑋1,18,𝑡) IF (𝐸𝐶5· (𝐵𝑡− 𝑋1,8,𝑡− 𝑋1,18,𝑡) ≤ 𝐺5) = 𝐺5 IF (𝐸𝐶5· (𝐵𝑡− 𝑋1,8,𝑡− 𝑋1,18,𝑡) > 𝐺5)

Furthermore, the storage 𝑌_13,𝑡 is calculated by summing the inventory of 𝑌_13,𝑡−1 times the storage efficiency 𝐸𝑆₁₃, and the incoming hydrogen flows of 𝑋_2,13,𝑡 and 𝑋_4,13,𝑡 times the conversion efficiency 𝐸𝐶13. However, if this surpasses the storage capacity of 𝑆13, then the

inventory will be capped, and it is assumed that the initial energy production is reduced to fit the hydrogen storage capacity.

𝒀𝟏𝟑,𝒕

23 𝑿𝟏𝟑,𝟏𝟏,𝒕+𝟏

= 𝐻𝑡+1−𝑋1,8,𝑡 IF (𝐸𝑆13·𝑌13,𝑡+𝐸𝐶13·𝑋2,13,𝑡+1≥ 𝐻𝑡+1−𝑋1,8,𝑡) = 𝐸𝑆13·𝑌13,𝑡+𝐸𝐶13·𝑋2,13,𝑡+1 IF (𝐸𝑆13·𝑌13,𝑡+𝐸𝐶13·𝑋2,13,𝑡+1< 𝐻𝑡+1−𝑋1,8,𝑡)

(5) Lastly, if at the next time period the heat demand 𝐻_𝑡+1 is satisfied and the electricity demand 𝐸_𝑡+1 is not satisfied from wind energy 𝑊_𝑡+1 and 𝑋_1,18,𝑡+1, then any available hydrogen storage 𝑌_13,𝑡 may be applied through the G2P-application 𝐺₆, to supply the electricity demand. Note that no additional hydrogen production is available, or there would not exist an electricity supply shortage. 𝑿𝟏𝟑,𝟏𝟖,𝒕+𝟏 = 𝐸𝑡+1− 𝑊𝑡+1− 𝑋1,18,𝑡+1 IF (𝐸𝑡+1− 𝑊𝑡+1− 𝑋1,18,𝑡+1 ≤ 𝐸𝑆13· 𝐸𝐶6· 𝑌13,𝑡≤ 𝐺6) = 𝐸𝑆13· 𝐸𝐶6· 𝑌13,𝑡 IF (𝐸𝑆13· 𝐸𝐶6· 𝑌13,𝑡< 𝐸𝑡+1− 𝑊𝑡+1− 𝑋1,18,𝑡+1≤ 𝐺6) =𝐺6 IF (𝐸𝑡+1− 𝑊𝑡+1− 𝑋1,18,𝑡+1 > 𝐺6) IF (𝐸𝑆13· 𝐸𝐶6· 𝑌13,𝑡> 𝐺6) 4.5 Model Outputs

Based on the previously explained inputs, assumptions, parameters and rationale, each modelled scenario generates the energy flows on an hourly basis for the year 2017. This includes the hourly energy production of electricity and heat, the overages or shortages, the converted energy flows, the stored energy and lastly, the amount of energy consumed in kWh.

➢ Energy production ➢ Energy overages ➢ Energy shortages ➢ Energy conversions ➢ Energy inventory sizes ➢ Energy consumed

Consequently, the hourly outputs of the 8760 hours of the year are summed and translated to the annual technical performance, as illustrated in table 4.5. In turn, the technical performance is used to derive the economic performance and the DSS.

24 As is visible in this case, both overages and shortages exist, caused by mismatches in demand and supply on an hourly basis. However, as this scenario applies a grid connection all the demand is supplied even though shortages exist. Moreover, sometimes energy deficits have to be replenished from the external grid, rather than through the buffering effect of the grid, affecting the DSS. Also, the amount of energy sold off at the end of the year is indicated. If a scenario does not apply a grid connection, the shortages indicate the missed demand and will have an effect on the final consumption.

Overall, the technical performance leads to a certain DSS, which is calculated, conform the definition by Engelken et al. (2016), as the proportion of locally generated energy to the local energy demand. However, the addition is made that locally generated energy must also be consumed at the right time. Therefore, the following formulae, 2 and 3, are applied for calculating the DSS for on- and off-grid scenarios, respectively. The DSS is calculated for electricity and heat individually, and totalled.

DSS = (𝐶𝑜𝑛𝑠𝑢𝑚𝑒𝑑 − 𝐷𝑒𝑓𝑖𝑐𝑖𝑡 𝑓𝑟𝑜𝑚 𝑔𝑟𝑖𝑑) / 𝐷𝑒𝑚𝑎𝑛𝑑·100% (2)

DSS = 𝐶𝑜𝑛𝑠𝑢𝑚𝑒𝑑 / 𝐷𝑒𝑚𝑎𝑛𝑑·100% (3)

Furthermore, the economic performance is measured for each scenario. This is measured by the performance indicators: return on investment (ROI) and the payback period (PBP) in years. The formulae for these, 4 and 5, are presented below.

ROI = 𝑁𝑒𝑡 𝑃𝑟𝑜𝑓𝑖𝑡 / 𝑇𝑜𝑡𝑎𝑙 𝐼𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡·100% (4)

= (𝐼𝑛𝑐𝑜𝑚𝑒 𝐸. + 𝐼𝑛𝑐𝑜𝑚𝑒 𝐻. − 𝑂𝑝𝑒𝑥) / 𝐶𝑎𝑝𝑒𝑥·100%

25 4.6 Experimental Design

The experimental design is the next step in the adapted simulation process. As has become clear so far, the simulation model basically acts as a calculation script for measuring the performance of twelve unique scenarios. Initially, this is done for the applied base values, which consists of 750 cows, 1MW wind turbine and the actual demand of Dearsum. However, for the purpose of verification and validation of the results, experiments are performed. The design hereof is based on altering the front- and back end circumstances of the system, that is: the energy supply and -demand side. Table 4.6 shows that the experimental design is revolved around three varying capacities for both supply inputs. Namely, the number of cows and the wind turbine are halved and doubled compared to the base values. Conversely, to check whether upscaling affects the performance, the demand side is doubled.

Table 4.6: experimental design

All the possible combinations of these distinctions lead to eighteen (3x3x2) cases for each scenario. However, some scenarios apply only biomass or wind, rather than a combination hereof. Hence, leading to six cases (3x2). After altering the inputs, the required capacities of the applied components 𝐺_𝑖, are deduced. This is done by checking the minimum required capacity for reaching 100% DSS, or if not, the capacity that avoids production losses. This is a trial-and-error process in which the capacities are initially set to infinity and incrementally decreased until the threshold of not being fully self-sufficient. In case this is impossible, the capacities are decreased until the point that avoids production losses. Likewise, the required storage capacities of components 𝑆𝑖 are deduced. However, as the simulations run for a single

year and a cyclical solution is sought covering multiple years, a starting inventory is set for 𝑆𝑖.

26

PART V: RESULTS

This chapter covers the steps ‘analyse data’ and ‘document results’ of the adapted simulation process, under the assumption that the preceding steps have been executed. Basically, the derived findings from the simulated scenarios and conducted experiments are presented and discussed. Herewith, the results for the base values (750 cows, 1MW wind turbine, 1x demand) will be compared with the experimental results, which are presented in Appendix D. Table 5.1 presents the performance measures for the scenarios in which the base values are applied. Herein, blue indicates the ROI and the PBP from top-to-bottom and green indicates the degree of self-sufficiency (DSS) for electricity, heat and in total, from top-to-bottom, respectively.

Table 5.1: performance for base values

Table 5.2 presents the required outlet capacities 𝐺𝑖 and storage capacities 𝑆𝑖 for achieving as

high as possible DSS, applying the base values. For all the experimental capacity requirements refer to Appendix D. These values provide additional insights into the mechanics of the system for various scenarios and inputs. The ‘S’ represents the scenarios and ‘C’ the components.

27 5.1 Biomass Results

As becomes visible from the performance measures in table 5.1, biomass does not optimally perform when applied exclusively. For the base values, biomass may supply 90% of the total energy demand through a CHP-engine self-sufficiently. However, due to the absence of a grid connection this implies that the other 10% are unsupplied and results in blackouts. In contrast, table 5.3 shows that experiment 5, applying overcapacity on the supply side, results in reaching a (nearly) full DSS. However, this implies that a disproportionate amount of overcapacity exists, leading to a system which is overloaded with electricity that has nowhere to go.

Table 5.3: scenario 1 experiment 5

Therefore, applying biomass in conjunction with a buffer option could reduce the issue of overcapacity and diminish the effect of blackouts if a maximum DSS is not achieved. For the base values, adding a grid connection somewhat improves the performance to a total DSS of 94%. Hereby, supply is guaranteed, but not with a maximum DSS. This is because the CHP-engine does not receive enough biogas for generating electricity overages for buffering purposes. Similarly to the above, table 5.4 shows that the experimental results for doubling the supply do lead to a maximum DSS. In this case, the overages of electricity are offloaded to the grid, which acts as a buffer, balancing the overall system. Moreover, an additional income source is generated through the grid, responsible for the increased ROI of 5.6% and PBP of maximum five years.

Table 5.4: scenario 4 experiment 5

28

Table 5.5: scenario 6 experiments 3,5,6

Similarly, this is the case for biomass in conjunction with both buffering options. For the base values this scenario is not fully self-sufficient. This is because the electricity grid is now used as buffer for the electricity demand, meaning more electricity overages are offloaded to the grid rather than converted to hydrogen. This results in the heat demand not being covered self-sufficiently (98%), but the electricity demand is. In contrast, doubling the supply side, as shown in table 5.6, once again results in a maximum DSS. Yet, similarly to scenario 6 (experiment 5) the role of hydrogen becomes obsolete. Even though the supply is guaranteed through the grid, these experiments, for the same reasons as the other biomass and hydrogen scenario, do not result in economically viable energy systems.

Table 5.6: scenario 8 experiment 5

5.2 Wind Results

The RES wind is unsuitable to individually supply both the electricity and heat demand. None of the experiments, including the base values (1MW, 1x), result in a maximum DSS. As table 5.1 depicts, wind is able to supply the electricity demand self-sufficiently for 80% and the heat demand for 58%. This implies that blackouts occur for the remaining percentages. Increasing the capacity of the wind turbine somewhat increases the DSS, but also decreases the economic viability. Taking into consideration that a short-term buffer is already applied for heat, in the form of hot water storage tanks (HWSTs), additional long-term buffers are needed to match the intermittent supply with the demand.

29 In contrast, applying wind with hydrogen somewhat improves the DSS for the base values, compared to applying no means of buffering. However, large supply deficits are still entrenched with this RES for all experiments, leading to blackouts. Even when applying the minimum capacity and storage requirements for mitigating production losses, demand is not able to be met. However, as depicted in table 5.7, experiment 5 is able to fully supply the demand self-sufficiently. Herein, a large wind turbine in combination with exorbitant capacities for the electrolyser and H2-storage are applied, resulting in a 100% DSS. Consequently, this does negatively affect the economic performance, rendering this scenario unviable.

Table 5.7: scenario 7 experiment 5

On the other hand, if the grid is added as a means of buffering hydrogen, the economic performance is positively affected with a high ROI and feasible PBP. Additionally, supply is guaranteed in case of an insufficient DSS. Regarding the experiments, all experiments are able to fully supply the electricity demand self-sufficiently, but not the DSS in totality. The exception hereof, is experiment 5, depicted in table 5.8. The overcapacity of the wind turbine and the ability to buffer and sell overages through the grid results in the full DSS and high economic performance. An interesting prerequisite herewith, is that the electrolyser capacity should not be set to the minimum required capacity for allowing all electricity overages to be converted to hydrogen. Otherwise, no electricity will be buffered in the electrical grid, leading to a lower DSS. However, it must be noted that this is a hypothetical scenario in which the gas grid may be utilised for hydrogen transport and storage. Nonetheless, it may become reality in the future and thus, could give guidance on energy systems applying wind in conjunction with a hydrogen grid infrastructure.

30 5.3 Biomass & Wind Results

Combining biomass and wind for the supply of the demand, generally results in a significantly higher performance compared to applying these RESs individually. Firstly, applying both RESs without a means of buffering, results in a 99% total DSS for the base values. However, as no grid connection is available the shortages result in blackouts. Yet, this is a significant increase compared to the individual RESs without buffers. This is due to the initial application of biogas and wind energy for the heat- and electricity demand, respectively. Interestingly, any overages of gas or electricity in this case, are applied to cover the supply shortages on the opposite demand side, resulting in a higher DSS. The HWSTs allow for final electricity overages to be buffered on the short-term and eventually lost as heat, creating equilibrium in this system. Within the experiments, a maximum DSS is achieved when the number of cows is doubled for all wind turbine sizes and a single demand, as depicted in table 5.9. However, the HWSTs are rendered obsolete here, meaning the system has overcapacity. All in all, combining biomass and wind results in a (nearly) self-sufficient energy system, whilst adhering to a positive economic performance.

Table 5.9: scenario 3 experiments 13,15,17

Furthermore, applying a grid connection as a means of buffering the RESs wind and biomass, results in a fully self-sufficient system for the base values and economically viable scenarios for all experiments. In addition, supply is guaranteed when shortages arise. However, to attain a full DSS, the right combination of wind and biomass inputs are required compared to the demand. Namely, at least 750 cows and a 500kW wind turbine for the single demand and at least 1500 cows and a 1MW wind turbine for double the demand. Table 5.10 visualises this. The reason for this is that biomass and wind are purely used for the heat and electricity demand, respectively. Hence, they do not cover the intermittent nature of the opposite side.

31 On the other hand, applying hydrogen as a buffer in conjunction with the RESs biomass and wind, results in many off-grid scenarios, which reach a 100% DSS and are generally economically viable. These experiments apply varying combinations of the supply side and demand side. This is shown in table 5.11. Interestingly though, it is key to balance the capacities on the supply side compared to the demand side. Otherwise, hydrogen may be overutilized or underutilized. The experiments with undercapacity for one RES and overcapacity for the other, compared to the demand, tend to have high H2-storage requirements. For example, see experiments 3, 12 and 14. In contrast, all experiments with 1500 cows and a single demand, regardless of the wind capacity, renders hydrogen obsolete. This is due to excessive supply compared to the demand, resulting in an overloaded system. Therefore, applying the right combination of supply inputs in comparison with the demand, results in a more balanced and viable system. The base values (experiment 9), and twice these values (experiment 18), are prime examples hereof.

Table 5.11: scenario 11 experiments 3,5,7,9,11-18

32

33

PART VI: DISCUSSION

6.1 Implications

In general it may be concluded from the results that a well-informed and balanced selection of the RESs, applications and storage options is paramount for achieving energy self-sufficiency, whilst adhering to economic viability. This corroborates with the studies of Nakata et al. (2005) and Rae & Bradley (2012). Generally, the combination of both RESs biomass and wind, together with any form of buffering and the right applications results in energy self-sufficient and economically viable systems.

6.1.1 biomass implications.

The results show that biomass energy from manure fermentation has an essential role on energy systems, both economically and in the degree of self-sufficiency. It is shown that biomass does not work optimally when applied exclusively, and works best in conjunction with wind energy, as is suggested by Dong et al. (2009). Namely, biomass exclusively is unable to fully supply the demand, unless large overcapacity of supply inputs is applied through a CHP-engine. However, this results in congestion of the system, necessitating the need for a buffer. Herewith, any combination with a heating grid or hydrogen is considered unviable due to the high application costs and storage requirements. In contrast, the grid as a buffer does improve the DSS, especially if biomass is upgraded to green gas. This is in line with the suggestions of Dong et al. (2009). On the other hand, the results disagree with Nakata et al. (2005) whom state that the economic viability is a major consideration of biomass applications. However, the results have consistently shown that biomass is economically viable for certain scenarios, if applied correctly. This is in line with the study of Kanase-Patil et al. (2010), which reports a high degree of biomass application. All in all, the results support the statement by Dong et al. (2009) that biomass has become more viable and attractive for energy self-sufficient systems nowadays.

6.1.2 wind implications.

34

6.1.3 hydrogen implications.

In general, hydrogen applications benefit energy self-sufficiency, but it comes at a cost. In conjunction with purely biomass, hydrogen yields a negative economic performance. Likewise, this is the case for hydrogen with purely wind. In accordance with Dutta (2014) this is due to the exorbitant amount of required H2-storage. However, the results have shown that applying a balanced combination of the supply inputs, corresponding with the demand, reduces the required H2-storage. In addition, the hypothetical future hydrogen grid infrastructure, as outlined by Nastasi & Lo Basso (2016), improves the economic viability. Furthermore, applying hydrogen for methanation purposes, as outlined by Vo et al. (2018), has its implications. This process may benefit the DSS on the heat side and reduce CO2 output, but also reduces the economic performance. Thus, it is an economic and environmental trade-off, in which it is must be decided how much CO2-reduction is worth.

6.1.4 grid connection implications.

Although applying a grid connection does not constitute absolute self-sufficiency conform Engelken et al. (2016), the results indicate that it is beneficial in all cases. A connection to the grid yields scenarios with the highest economic performance as well as the DSS. Consequently, three out of four of the fully self-sufficient and economically viable scenarios apply a grid connection. This is no coincidence, as the grid is considered a buffer option with unlimited storage capacity for covering supply deficits and offloading excess energy (Nakata et al., 2005; Rae & Bradley, 2012). Additionally, conform Rae & Bradley (2012) the grid offers an additional income source, further enhancing the economic performance. However, it must be addressed that the assumption of unlimited storage capacity and the applied netting arrangement may not be generalizable to all contexts, including the Netherlands in the foreseeable future. Notwithstanding, this has partially been taken into account by assigning different prices for buying and selling energy.

6.2 Limitations

35

PART VII: CONCLUSION

This study has set out to lay the foundations for designing self-sufficient energy systems, by answering the research question: ‘What are the key design decisions for energy systems based on biomass and wind energy and how do they influence the degree of self-sufficiency and economic viability?’ To this end, twelve unique scenarios based on simple heuristics and varying configurations of biomass and wind, have been constructed and simulated over the timespan of a year. Hereby, the small-scale agricultural community Dearsum acted as inspiration and real-life case.

The results have shown that a well-informed decision must be made on which renewable energy sources to apply and made to fit with the context-dependent demand. Hereby, the combination of biomass and wind complement each other. The balance between these two renewable energy sources and the related demand is key. Elsewise, this may result in large capacity requirements and unviable systems. In addition, adding a buffering option generally improves the degree of self-sufficiency, but works best for the combination of biomass and wind. A connection to the grid has shown to be an advantageous decision for alleviating mismatches in demand and supply and generating an additional income source. Hydrogen has shown to play a vital role in attaining absolute self-sufficiency and reducing CO2-output, but generally comes at high costs. A hydrogen grid infrastructure has shown potential for enabling viable self-sufficient energy systems in the future.

36

REFERENCES

AlleCijfers. (2019). ‘Informatie buurt Dearsum’. Available at: https://allecijfers.nl/buurt/

dearsum-sudwest-fryslan/. (Accessed: 28th May 2019).

Amer, M. & Daim, T.U. & Jetter, A. (2013). A review of scenario planning. Futures, 46, 23- 40.

Beurskens, L. & Mast, E. (2017). Kostenonderzoek windenergie SDE+ 2018, Cost research report. ECN-N--17-011, ECN, DNV-GL, Amsterdam.

Boon, F.P. & Dieperink, C. (2014). Local civil society based renewable energy organisations in the Netherlands: Exploring the factors that stimulate their emergence and development. Energy Policy, 69, 297-307.

CBS. (2019a). ‘Aardgas en elektriciteit, gemiddelde prijzen van eindverbruikers’. Available at: https://opendata.cbs.nl/statline/#/CBS/nl/dataset/81309NED/table?fromstatweb.

(Accessed: 12th September 2019).

CBS. (2019b). ‘Windenergie op land; productie en capaciteit per provincie’. Available at: https://opendata.cbs.nl/statline/#/CBS/nl/dataset/70960ned/table?fromstatweb. (Accessed: 3rd September 2019).

Celador, A.C. & Odriozola, M. & Sala, J.M. (2011). Implications of the modelling of stratified hot water storage tanks in the simulation of CHP plants. Energy Conversion and Management, 52(8-9), 3018-3026.

Dear, M. (1992). Understanding and Overcoming the NIMBY Syndrome. Journal of the American Planning Association, 58(3), 288-300.

Dong, L. & Liu, H. & Riffat, S. (2009). Development of small-scale and micro-scale biomass fuelled CHP systems – A literature review. Applied Thermal Engineering, 29, 2119-2126.

Dunn, S. (2002). Hydrogen futures: toward a sustainable energy system. International Journal of Hydrogen Energy, 27, 235-264.

Dutta, S. (2014). A review on production, storage of hydrogen and its utilization as an energy resource. Journal of Industrial and Engineering Chemistry, 20, 1148-1156.

Engelken, M. & Römer, B. & Drescher, M. & Welpe, I. (2016). Transforming the energy system: Why municipalities strive for energy self-sufficiency. Energy Policy, 98, 365-377.

Energy Star. (2008). ‘Residential Water Heaters: Final Criteria Analysis’. Available at: https://www.energystar.gov/ia/partners/prod_development/new_specs/downloads/

water_heaters/waterheateranalysis_final.pdf. (Accessed: 19th November 2019).

Eurostat. (2019). ‘Share of renewable energy in gross final energy consumption’. Available at: https://ec.europa.eu/eurostat/databrowser/view/t2020_31/default/table?lang=en.

(Accessed: 1st _{August 2019).}

Forrest, N. & Wiek, A. (2015). Success factors and strategies for sustainability transitions of small-scale communities – Evidence from a cross-case analysis. Environmental Innovations and Societal Transitions, 17, 22-40.

InterClima. (2017). ‘Prijslijst buffervaten 2017 / 2018’. Available at: https://www.interclima.nl

37 Jensen, J.O. & Vestbø, A.P. & Li, Q. & Bjerrum, N.J. (2007). The energy efficiency of onboard

hydrogen storage. Journal of Alloys and Compounds, 446-447(2007), 723-728.

Kanase-Patil, A.B. & Saini, R.P. & Sharma, M.P. (2010). Integrated renewable energy systems for off grid rural electrification. Renewable Energy, 35(6), 1342-1349.

Lako, P. (2010). ‘IEA ETSAP - Technology Brief E04 - Combined Heat and Power’. Available

at: https://iea-etsap.org/E-TechDS/PDF/E04-CHP-GS-gct_ADfinal.pdf. (Accessed:

19th November 2019).

Li, L.W. & Birmele, J. & Shaich, H. & Konold, W. (2013). Transitioning to community-owned renewable energy: Lessons from Germany. Procedia Environmental Sciences, 17, 719-728.

Liserre, M. & Sauter, T. & Hung, J.Y. (2010). Future Energy Systems: Integrating Renewable Energy Sources into the Smart Power Grid Through Industrial Electronics, IEEE Industrial Electronics Magazine, 4(1), 18-37.

McKenna, R. (2018). The double-edged sword of decentralized energy autonomy. Energy Policy, 113, 747-750.

McKenna, R. & Bertsch, V. & Mainzer, K. & Fichtner, W. (2018). Combining local preferences with multi-criteria decision analysis and linear optimization to develop feasible energy concepts in small communities. European Journal of Operational Research, 268(3), 1092-1110.

Müller, M.O. & Stämpfli, A. & Dold, U. & Hammer, T. (2011). Energy autarky: a conceptual framework for sustainable regional development. Energy Policy, 39(10), 5800-5810. Nakata, T. & Kubo, K. & Lamont, A. (2005). Design for renewable energy systems with

application to rural areas in Japan. Energy Policy, 33(2), 209-219.

Nastasi, B. & Lo Basso, G. (2016). Hydrogen to link heat and electricity in the transition towards future Smart Energy Systems. Energy, 110, 5-22.

Nussbaumer, T. & Thalmann, S. (2014). Status Report on District Heating Systems in IEA Countries. Swiss Federal Office of Energy, and Verenum, Zürich.

Panwar, N.L. & Kaushik, S.C. & Kothari, S. (2011). Role of renewable energy sources in environmental protection: A review. Renewable and Sustainable Reviews, 15(3), 1513-1524,

Rae, C. & Bradley, F. (2012). Energy autonomy in sustainable communities – A review of key issues. Renewable and Sustainable Energy Reviews, 16, 6497-6506.

RVO. (2009). Van Biogas Naar Groen Gas Opwaarderingstechnieken en Leveranciers. Available at: https://www.rvo.nl/sites/default/files/bijlagen/Van%20Biogas%20naar%

20Groen%20Gas%20-%20Opwaarderingstechnieken%20en%20leveranciers.pdf.

(Accessed: 19th November, 2019).

RVO. (2017). ‘Windsnelheid per gemeente SDE+ december 2017’. Available at: https://www. rvo.nl/sites/default/files/2017/01/Windsnelheid%20per%20gemeente%20SDE%20juli

%202017.pdf. (Accessed: 9th September 2019).

RVO. (2019). ‘RVO Brochure SDE Voorjaar 2019.’ Available at: https://english.rvo.nl/sites/

default/files/2019/04/RVO%20Brochure%20SDE%20Voorjaar%202019.pdf.