Renewable Gasses in Future
Local Energy Systems
Modelling Hydrogen and Biogas Storage Requirements
to Balance Supply and Demand Mismatches
MS
CT
HESIST
ECHNOLOGY&
O
PERATIONSM
ANAGEMENTRenewable Gasses in Future
Local Energy Systems
Modelling Hydrogen and Biogas Storage Requirements
to Balance Supply and Demand Mismatches
Author: Supervisor:
ING.R.J.(RENÉ)SCHUPPERT DR.M.J.(MARTIN)LAND
S3509737 Co-Assessor:
RSCHUPPERT@HOTMAIL.COM PROF. DR. IR.J.C.(HANS)WORTMAN
24
J
UNE2019
MS
CT
HESISF
INALACKNOWLEDGEMENT
ABSTRACT
Purpose: The goal of this study is to analyse how renewable gasses can create flexibility for future local energy system’s households and buildings, to balance intermittent distributed renewable energy sources (DRES) supply and demand by determining local and national storage capacities under various transition pathways. Storage of renewable gasses (hydrogen and biogas), that can replace conventional fuels, is required to retain flexibility in energy systems, that is degraded by the increase of DRES.
Methodology: A simulation study is conducted on a future energy system of a Dutch municipality, in which supply from wind, solar pv, and biogas and demand from hybrid and all-electric heat pumps and electricity is modelled, with buffers to overcome daily and seasonal mismatches. Many experiments are modelled to capture the uncertainty of future energy systems and transition pathways on the required storage capacity of the system. Hourly supply and demand quantities have resulted in annual operational storage and transport decisions. Additional output data is collected from the model to describe the system.
Results: Storage capacity is required to balance daily and seasonal mismatches. Differences in supply and demand, ratio of supply and demand, and different year impact storage requirements. Moreover, a self-sufficient system behaves significantly different from a system with energy import and export. Matching indices are also found to vary among the modelled experiments.
Implications: Optimal local storage capacities can decrease total required storage capacity. Local storage is in the order of MWh (cylinder tankers) and national in the order of GWh (salt caverns). Optimising the distribution of wind and solar PV can decrease storage capacity, while variability in supply not always results in larger storage capacity. Scenario planning is found to be important to tackle such problem, due to uncertainties. Future studies should analyse the interconnection of multiple regions, different geographical settings, expand flexibility strategies used, and involve costs in decision-making.
CONTENT
1. Introduction ... 5
2. Theoretical background ... 6
2.1. Incentive to Increase DRES and Its Implications ... 6
2.2. The Role of Renewable Gasses in Future Local Energy Systems ... 8
2.3. Methodologies for Researching Energy Systems with Renewable Gasses ... 9
2.4. Summary of Critical Findings ... 10
3. Methodology ... 11
3.1. Research Model ... 11
3.2. Conceptual Model of the Local Energy System ... 12
3.3. Simulation Model ... 13
3.4. Model Input Data ... 14
3.5. Operational Rationale of the Simulation Model ... 20
3.6. Model Output Data ... 21
3.7. Experimental Design ... 21
4. Results ... 22
4.1. Analysing the Effects of Experimental Variables on Storage Capacities ( I ) ... 22
4.2. Analysing the Difference Between Local and National, and Only Local Storage ( II ) ... 29
4.3. Analysing Two Basic Mismatch Indices ( III ) ... 31
5. Discussion... 32
5.1. Implication of the Results ... 32
5.2. Limitations ... 35
5.3. Future Research ... 35
5.4. Conclusion... 36
References ... 37
Appendix 1 – Research model and steps taken ... 40
Appendix 2 - Summary of the NVDT scenarios (CE Delft, 2017) ... 41
Appendix 3 – Detailed rationale of the simulation model ... 42
Appendix 4 – Energy storage capacities for all experiments... 44
Appendix 5 – Ratio of solar pv and wind for scenario ‘optimal supply’ ... 45
1. INTRODUCTION
Energy storage is becoming increasingly urgent in the energy transition, as the growth in distributed renewable energy sources (DRES)is continuing (IEA, 2018). However, this growth of mainly wind and solar PV, implies more intermittent energy supply, i.e. non-dispatchable and fluctuating generation (Denholm & Hand, 2011), due to the intrinsic climate dependence of renewables (Parra et al., 2019). An increase in intermittent energy supply causes a decrease in flexibility, meaning the energy system is less able to respond to changes in energy generation and demand, creating daily and seasonal mismatches (Huber, Dimkova & Hamacher, 2014). Renewable gasses, like hydrogen and biogas, can store energy to balance this mismatch between renewable energy supply and demand and decarbonise energy systems.
Ferrero, Gamba, Lanzini, and Santarelli (2016) promote the installation of hydrogen-based energy storages, as it can convert surplus electricity into ‘green hydrogen’ and re-electrify hydrogen to satisfy demand. Besides hydrogen, biogas also allows for flexible energy storage, applicable to various demands to decarbonise the energy system (Sacher et al., 2019). Koirala, Van Oost, and Van der Windt (2018) argue that future energy systems will be a combination of centralised (local) and decentralised (national) energy storages working in synergy, to solve the energy balance conundrum. Hence, considering the interaction of local and national energy storages of hydrogen and biogas to retain flexibility, will be important for future energy systems.
Investigating the role of renewable gasses is particularly interesting for households and buildings in local energy systems, e.g. a municipality, as their energy demand and supply (prosumption) will be subject to severe changes (CE Delft, 2017). During the energy transition, the Dutch government is explicitly stimulating a bottom-up approach for households and utility buildings (VNG, 2018). Therefore, regions and municipalities must adopt national goals as guidelines, to design plans whereby installation of DRES and the required infrastructure must be specified. Storage of renewable gasses can support these plans by providing flexibility to the local energy systems, without the concern of curtailment. Curtailment is currently an issue, since local distribution grids often cannot cope with the increasing penetration of DRES (Meijer, 2019).
the influence of DRES on households and buildings has been presented by Tronchin, Manfren, and Nastasi (2018). However, knowledge regarding transportation, conversion, and storage requirements of renewable gasses in local energy systems is limited from an operations management perspective. This perspective is required to abstract out from the details and focus on explaining the system’s behaviour. Finally, system operators are still in development of comprehending system requirements for sectoral integration, i.e. connecting electricity and gas grids (Gigler & Weeda, 2018; Gasunie & Tennet, 2019). Hence, studying this subject is pioneering work with uncertainties. However, deepening knowledge has theoretical and practical relevance.
The predominance of wind, solar PV,and biogas energy, bottom-up approach for DRES,and the mentioned challenges, make it interesting to investigate the role of these three DRES in future local energy systems, as gathering knowledge regarding the effects of DRES on energy storage and future energy systems is in development. For this investigation, the following research question will be answered:
How should hydrogen and biogas storage components of a future local energy system’s households and buildings be designed and operated to balance DRES supply and demand mismatch?
The goal of this study is to analyse how renewable gasses can create flexibility for future local energy system’s households and buildings, to balance intermittent DRES supply and demand by determining local and national storage capacities under various transition pathways. The study’s contribution to literature lies in determining operational decision-making on where, when, and how much hydrogen and biogas to store, withdrawal, convert, and transport, while considering the previously mentioned opportunities and challenges.
The remainder of the paper is structured as follows. Section two elaborates on the theoretical background regarding the subject. Section three, methodology, describes the methods used to answer the research question and discusses the data used for analysis. Section four, presents the results of the study. Finally, the results are reflected on and concluded in section five, discussion.
2. THEORETICAL BACKGROUND
By means of the theoretical background, recent literature regarding the implications of DRES, role of renewable gasses in energy systems, and how previous studies tackled comparable problems is sought. At the end of this section, a summary is presented.
2.1. Incentive to Increase DRES and Its Implications
total power generation by 2040, with solar photovoltaics (PV) and wind energy being the main contributors (IEA, 2018). Besides wind and solar PV, bioenergy, considered by IEA (2018) as “the overlooked giant within renewable energy”, is and will remain a large renewable energy source. A prominent form of bioenergy during the current energy transition is biogas, which is a fuel that can be used for heating, transport, and generation of electricity (Sacher et al., 2019).
A large share of renewable energy generation comes from distributed renewable energy sources (DRES),that are dispersed within a country and realised on a local scale. According to Bozalakov et al. (2014), sustainability goals can only be met with a significant increase in number of DRES. To stimulate the increase in DRES in future energy systems, the Dutch government adopted a bottom-up approach in the energy transition (VNG, 2018). The bottom-up approach for DRES and predominance of wind, solar PV,and biogas energy, allow for an interesting topic to analyse role of these three DRES in future local energy systems. Moreover, comprehending the effects of DRES on future energy systems is in development. The three renewable energy sources are not only interesting for their increasing appearance and sustainable character, but mainly due to the challenges they bring to future energy systems.
Renewable energy supply is fundamentally different from conventional, fossil-based, energy generation. Kosmadakis, Karellas, and Kakaras (2013), Denholm and Hand (2011), and Stram (2016) discuss the characteristics of conventional and renewable energy systems. They conclude that the main advantage of conventional generation is its dispatchable generation capability, i.e., being able to adjust the electricity output to varying changes in demand. This is convenient for energy systems, since the necessity for storage is avoided. Renewable energy sources which are dispatchable, are hydropower and geothermal energy (Johansson et al., 2012). However, the EIA (2015) state, that these sources are restricted in growth and quantity, mainly, due to their geographic constraints.
To allow for more DRES and overcome its implications,flexibility is required to balance energy systems (Denholm & Hand, 2011; Huber, Dimkova & Hamacher, 2014; Lund et al., 2015; Koirala, Van Oost & Van der Windt, 2018). Flexibility can be accomplished by storing energy. Consequently, the required storage capacity, is an indication of imbalance in a system (Robledo et al., 2018). Including energy storage, Ferrero et al. (2016) present five methods to harness flexibility in future energy systems, namely: demand side flexibility; grid reinforcement; supply side flexibility; and smart grid concepts. Despite the effectiveness of energy storage method like hydropower, compressed air, and batteries in providing flexibility, they are either constrained by geography, capacity, or costs. Meanwhile, renewable gasses, like hydrogen and biogas, are geographically less constrained, can be transported in currently existing gas infrastructures, and can easily be stored in large volumes (Ferrero et al., 2016). Hence, hydrogen and biogas will be considered as energy carriers in this study, next to electricity.
Besides alterations in energy supply, future local energy systems will be subject to changes in demand. The bottom-up approach focusses on households and buildings of local energy systems, e.g. municipality, demanding energy for heating and electricity (VNG, 2018). CE Delft (2017) predicts that future households and buildings will largely be heated by either all-electric heat pumps or hybrid heat pumps. Hybrid heat pumps use both electricity and gas, like hydrogen and biogas. However, due to the novelty of these heating sources, its effect on energy systems are hardly known (Watson, Lomas & Buswell, 2019).
2.2. The Role of Renewable Gasses in Future Local Energy Systems
Renewable gasses can support in decarbonising future local energy systems, with storing energy to balance supply and demand. To investigate what role they can fulfil, hydrogen and biogas energy system components and characteristics must first be delineated.
2.2.1. Hydrogen
Hydrogen’s foremost roles during the energy transition are the integration of DRES and decarbonisation of energy systems (Acar & Dincer, 2019). Hydrogen can offer future energy systems daily and seasonal energy balance between supply and demand, by storing electricity from DRES, which is abundantly described in literature(Corbo, Migliardini & Veneri, 2011; Le Duigou et al., 2013; Gigler & Weeda, 2018).
systems or tankers. Meanwhile, it can be stored in cylinder tanker or underground reserves, like salt caverns, for daily and seasonal storage. By adopting such a system that provides the required flexibility to future energy systems, the mismatch between supply and demand can be balanced. Combining strengths of both the electricity grid and the gas grid, because of the conversion possibilities, a synergy is creased that is commonly referred to as ‘sectoral integration’ (Buttler & Spliethoff, 2018).
Gasunie and Tennet (2019), the Dutch gas and electricity TSOs, jointly developed the ‘Infrastructure Outlook 2050’. In the outlook, future pathways for high penetration of renewable energy, combined P2G hydrogen conversion, and energy storage are developed. Dutta (2014), provided an extensive review of hydrogen as a fuel, including its application in fuel cells and hydrogen fuelling engines. Gigler and Weeda (2018) present in their ‘Hydrogen Roadmap’, that hydrogen can perfectly be applied to fulfil heat and electricity demand of households and buildings. These reports are analysed, to grasp how hydrogen can be applied to households and buildings in future local energy systems.
Despite hydrogen’s potential in future energy systems, a few ‘key uncertainties’ remain (McDowall, 2014). Currently, a prominent barrier to large scale green hydrogen adoption, are its high production costs (Corbo, Migliardini & Veneri, 2011). Furthermore, connecting DRES with energy demand via hydrogen comes with high energy efficiency losses (Page & Krumdieck, 2009; Maroufmashat & Fowler, 2017). However, hydrogen energy systems are expected to operate at higher efficiencies in the future (Nakamura et al.,2015). Additionally, Acar & Dincer (2019) say, increasing hydrogen production systems’ capacities will reduce costs. Hence, despite the current challenges, hydrogen is expected to contribute in decarbonising future energy systems.
2.2.2. Biogas
Biogas is generated through a process of anaerobic digestion of biomass, derived from biological sources (Fardin, De Barros Jr. & Dias, 2018). Biomass is collected, transformed into biogas in a digester, and transported via pipelines or stored in buffers (Lauer & Thrän, 2018). Although biogas can be used to generate electricity through combustion engines or gas turbines, the fuel is also applicable to biogas hybrid heat pumps in households and buildings. Unlike hydrogen being integrated with the electricity grid, this study will focus on the role of biogas, solely applicable as fuel for biogas hybrid heat pumps.
2.3. Methodologies for Researching Energy Systems with Renewable Gasses
Both Welder et al. (2018) and Bennoua et al. (2015) applied an optimisation model to investigate the design and operation of a hydrogen energy system on a national scale. Kalinci, Hepbasli, and Dincer (2015) too adopted an optimisation model in which the optimum size of equipment is determined in a hydrogen energy system to minimise costs. Although an optimisation method could be similarly applied on local energy systems, to study the operational aspect and behaviour of a system, among multiple experimental variables, a simulation study is preferred (Blanco and Faaij, 2018). A simulation is, ‘an experimentation with a simplified imitation (on a computer) of an operations system as it progresses through time, for the purpose of better understanding and/or improving that system’ (Robinson, 2004, p. 7).
Krajačić et al. (2008) and Denholm and Hand (2011) both performed a simulation study to expand renewable energy integration and hydrogen into an energy system and balance hourly supply and demand for multiple scenarios. A closed system was researched and energy supply from biomass was neglected. Clegg and Mancarella (2015) analysed the operational impacts of multiple P2G processes on electricity and gas grids, by adopting a simulation modelling approach. By examining two networks, a closed system was analysed, whereas, this study focusses on the interaction of a local energy system with the national system.
From studies of McDowall (2014), Barton et al (2018), and Hong et al. (2019) can be concluded that the future of energy systems is highly uncertain, calling the need for analysing multiple transition pathways. This premise of ‘scenario planning’ will therefore be adopted to cover for the many possible future transition pathways.
2.4. Summary of Critical Findings
FIGURE 2.1.– SUMMARY OF THEORY AND POSITIONING
3. METHODOLOGY
This study considers the role of renewable gasses, providing flexibility in a local future energy system, to balance mismatching supply and demand for households and utility buildings. The research goal is to analyse operational decision-making into when, where, and how much electricity, biogas, and hydrogen to store and distribute. The conceptual local energy system is based on a municipality in The Netherlands: Midden-Drenthe. Why Midden-Drenthe? One of the motivations to perform this study is that more electricity grids will soon become insufficiently capacitated resulting in curtailment. The municipality Midden-Drenthe is currently refusing solar parks due to insufficient grid capacity (Meijer, 2019). Hence, the conceptual model is based on the municipality Midden-Drenthe in the year 2050. Electricity, biogas, and hydrogen are the energy carriers and DRES.
This section will guide you through the development and construction of the conceptual local future energy system and how it is investigated to accomplish the research goal. First the research model including its phases is outlined. Then, a conceptual model of the analysed system is presented. Next, the simulation model is explained. This is followed up by subsections describing the model inputs, outputs, rationale, assumptions, and experimental design.
3.1. Research Model
A phase-wise research model is developed to provide structure during the research. The phases are visualised in figure 3.1. and globally describe how the study is performed. A detailed explanation is of the research model and its phases can be found in appendix 1.
3.2. Conceptual Model of the Local Energy System
To guide this study in achieving its goal, a conceptual model of a local future energy system in constructed, as presented in figure 3.2. The upper segment of the model resembles the energy system of a municipality’s households and buildings (local), whereas the lower segment resembles the central (national) energy system. The system can be divided into supply components, demand components, and infrastructural components.
3.2.1. Supply Components
The supply components consist of biogas, wind, and solar PV, representing biogas plants, wind turbines, and solar PV panels respectively. Although more DRES are envisioned to play an important role in future energy systems (e.g., geothermal energy and hydro power), this study focusses on the influence of the presented sources of supply, because of their geographical independence. Wind turbines and solar PV panels generate electricity, whereas biogas plants generate biogas. 3.2.2. Demand Components
Demand components are divided into electricity, biogas, and hydrogen demand. Electricity is demanded by hydrogen- and biogas-hybrid heat pumps, all-electric heat pumps, and appliances and light. Biogas and hydrogen are demanded by the hydrogen-, and biogas-hybrid heat pumps. In this study, the three different heat pumps are projected to heat the households and buildings of the municipality. Heat pumps are selected as heat source to analyse their effect on an energy system, because of their increasingly prominent presence, as described in section two.
3.2.3. Infrastructural Components
The infrastructural components are divided into local and national components. The local infrastructural components are located within the municipality. A local electricity grid connects electricity supply from DRES and a G2P facility, with electricity demand. The hydrogen network
consists of a P2G electrolyser, hydrogen buffer, G2P facility, and pipelines. The biogas network contains a biogas buffer to store biogas and pipelines to connect supply and demand. National system components are the high-voltage grid, national biogas buffer, and national hydrogen buffer. These are connected to the local components by means of the high-voltage grid and the biogas and hydrogen pipelines respectively. Energy surpluses or deficits can be exported or imported from and to the national system unrestrictively. The national buffers and electricity grid are connected to, and must be shared with, all municipalities and are central storage locations. An example of a national hydrogen buffer is a depleted salt cavern (Gigler & Weeda, 2018; Gasunie & Tennet, 2019). However, in this study, the national grid and buffers are ‘black boxes’, i.e., supply and demand from other municipalities on a national scope are neglected.
The system is designed to balance supply and demand by means of flexibility, accomplished by energy storage, import, and export. Biogas plants are assumed to produce at a constant rate. However, biogas demand has daily and seasonal fluctuations. In times of surplus supply, biogas is stored in the buffer and withdrawn during times of deficit. When there is shortage or overload of biogas in the local system, biogas can be either imported from, or exported to the national buffer. The hydrogen network functions similarly. However, hydrogen must be produced from electricity, generated by wind turbines and solar PV panels, characterised by their intrinsic weather dependencies, resulting in daily, seasonal, and even yearly variations. Hence, to realise stable and reliable electricity and hydrogen supply, hydrogen can be stored in the local buffer to overcome the mismatch in supply and demand. From this buffer, hydrogen is supplied to the hybrid heat pumps, and/or reconverted into electricity via G2P. If the local energy system is unable to satisfy electricity demand, the national high-voltage grid will supply electricity. Although most system components are currently absent in the energy system at the presented scale, the high-voltage electricity grid is based on its current capacity, to optimally make use of what is already installed. Based on this conceptual model, a simulation model is created to analyse its behaviour.
3.3. Simulation Model
An answer must be given to how the formerly described system must be designed and operated to balance supply and demand under various experimental variables. For this, capacities of the system components must be determined, and answers must be given to when to store, how much to store, and where to store hydrogen and biogas.
system designs, scenarios, and policies (Robinson, 2004, p. 7). The flexibility of a simulation study helped in developing and constructing an appropriate simulation model for many experiments. Many experiments are required to thoroughly eliminate uncertainties of the future energy system, by capturing multiple likely future pathways. By doing this, results among different experimental variables can be compared, including what hydrogen and biogas buffer component capacities are required. The experiments vary according the following categories: energy supply; energy demand; ratio of supply and demand; and multiple yearly weather patterns.
The simulation framework in figure 3.3. is used as guidance through the simulation study and visualises the input data, simulation model, and output data. Input data are hourly supply and demand data for electricity, biogas, and hydrogen. Besides input data for supply and demand, system parameters are used as input data. These are constraints under which the system is designed and must operate. Detailed explanation of the data inputs is presented in section 3.4., model inputs. The input data is transformed into output data, needed to answer the research question and achieve its goal, via the simulation model. The simulation model represents the storage and withdrawal priority rules on an hourly basis. The simulation model is in essence a large sequence of ‘what if’ rules.
3.4. Model Input Data
As discussed in section 3.3., input data determines the quantitative system variables for which the simulation model is designed and must operate. The input data comprises hourly quantities of supply and demand, and the system parameters providing constraints to the simulation model. These three input components are explained and discussed in this section. But first, the scenarios used to determine supply and demand are explained, together with the proportions of supply and demand components.
3.4.1. Supply and Demand Scenarios
The input data for the simulation model of supply and demand is based on a report containing projected future energy system scenarios in 2050 for The Netherlands, created by CE Delft (2017), called Grid of the Future (Net voor de Toekomst, further referred to as: NVDT). In this report, four scenarios are constructed, each encompassing a different prospect of the future. The scenarios are
called: Regional, National, International, and Generic Steering (see appendix 2 for description). The projected supply and demand totals are extracted per scenario and converted into usable data for the simulation model, representing the described local future energy system of section 3.2. 3.4.1.1. Supply Scenarios
Regarding the supply components, a production mix of DRES is determined. The CE Delft (2017) report presents per scenario a production mix of biogas, wind on land, and solar PV, specifically for distributed renewable energy resources (DRES), as presented in table 3.1. In addition to the four supply scenarios, two extra supply scenarios are created and modelled, based on the scenario ‘national’ and only differ in ratio of wind and solar PV.
The fifth scenario, called ‘optimal supply’, applies the optimal ratio between installed capacity of wind turbines and solar PV panels, to minimise hydrogen storage capacity. Bennoua et al. (2015) conclude that an optimal ratio of wind and solar PV can reduce storage needs. By including this scenario, their conclusion can be validated and simultaneously optimise the system under investigation. The optimal supply ratio is calculated for each experiment to minimise required storage capacity.
The sixth scenario, ‘current supply’ extrapolates the current situation of the municipality Midden-Drenthe regarding DRES to 2050. Currently 14 MW solar PV and zero MW wind is installed in Midden-Drenthe. Therefore, this scenario has a production mix of 90% solar PV and 10% biogas.
TABLE 3.1.- PRODUCTION MIX PER SCENARIO (%)
Regional National International Generic Optimal Current
Biogas 5% 10% 10% 10% 10% 10%
Wind 35% 50% 45% 40% # 0%
Solar PV 60% 40% 45% 50% # 90%
3.4.1.2. Demand Scenarios
Like the supply scenarios, the CE Delft (2017) report presents ratio of electricity and heat demand components per scenario for households and buildings. First, electricity demand for appliances and light is elaborated on. Then, heat demand is explained.
The projected electricity demand for appliances and light of Midden-Drenthe in 2050 is determined quite straight forward. CE Delft projects a decrease of 25% of electricity use for appliances and light. The current electricity demand of households and buildings within Midden-Drenthe is extracted from Rijkswaterstaat (2017). Retracting 25% of that electricity demand results in a yearly electricity demand of 33 GWh for households and 48 GWh for buildings, identical for all six scenarios.
pump types. Other heating options are omitted, to focus on the effects of the increasing amount of heat pumps on the energy system. The ratios are presented in table 3.2. The two extra scenarios contain equal demand as scenario ‘national’ and are therefore left out of the table.
To determine heat energy demand per component, the ratios are multiplied by Midden-Drenthe’s heating demand for households and buildings in 2050. Although total demand is unknown for Midden-Drenthe in 2050, a ‘best guess’ can be made, by multiplying total heating demand of The Netherlands in 2050 (differs per scenario) and the current share of heat demand consumed by Midden-Drenthe of heat energy in The Netherlands (0,196%). The heat demand totals of The Netherlands, Midden-Drenthe, and per heating option are displayed in table 3.2.
TABLE 3.2.- DISTRIBUTION RATIOS OF HEATING OPTIONS PER SCENARIO (%)
Regional National International Generic
(%) GWh (%) GWh (%) GWh (%) GWh
National demand 104.277 107.722 115.667 113.611
Municipal demand 100% 204 100% 211 100% 227 100% 223
Hydrogen heat pump 0% 0 62% 130 35% 79 0% 0
Biogas heat pump 31% 63 27% 57 60% 136 99% 222
All-electric heat pump 69% 141 11% 24 5% 12 1% 1
3.4.2. DRES Supply Components
Hourly supply quantities are based on three supply components: biogas production by digesters and electricity generation by wind turbines and solar PV panels. Each experiment contains a different ratio of supply components. Both the supply and demand components are based on yearly production profiles, made available by N.V. Dutch Gasunie. These are hourly percentages as a portion of a total year, that must be multiplied by the total yearly production quantity to calculate the hourly energy supply1. Below, in figure 3.4., the biogas, wind, and solar PV supply profiles of 2012
are presented on an annual timescale and in figure 3.5. on a monthly scale, to emphasise the seasonal and daily fluctuations. Hourly percentages are shown in the figures rather than total energy supply in MWh, because total supply in MWh varies per experiment.
1Example: Hour 1 for wind has a portion of 0,0272%. If the yearly wind production is 200.000 MWh, then that hour 54,4
MWh of electricity is produced (0,000272*200.000=54,4). Similarly done for demand. 0,00%
0,02% 0,04% 0,06% 0,08%
Jan bFe Mar Apr May Jun Jul Aug pSe Oct Nov De
c Ho url y Po rtion (%) Month Solar PV Wind 2012 Biogas
FIGURE 3.5.- MONTHLY BIOGAS, WIND, AND SOLAR PV SUPPLY PROFILES OF 2012(PERCENTAGE PER HOUR)
To start with, the steadiest supply profile is discussed: biogas production. In this simulation study, based on Fardin et al. (2018), the assumption is made that the biodigester is continuously fed, and therefore, a continuous steady biogas output is produced. Hence, fluctuations across year are absent. The yearly full load hours of the biogas digester are therefore 8.760 hours. The portion of produced biogas per hour is approximately 0,0142%.
The supply profile of wind energy is more fluctuating, with peaks up to 0,0350%, and troughs of zero electricity generation. Electricity generation from wind varies across years. Hence, to consider these variations, different wind year profiles are modelled and used as input. Wind profiles of 2012, 2013, and 2017 are used. To express the variation among these years, their standard deviations can be compared, as displayed in table 3.3. Their average hourly portion is obviously identical. However, the wind year 2017 contains significantly more variation than the other years. The full load hours of wind turbines are approximately 3.000 per year (CE Delft, 2017: 71). Based solely on the wind profiles, a prediction can be made that scenarios of 2017 require larger storage capacities, due to the infamous operations law: an increase in variation requires an increase in storage capacity.
The supply profile of solar PV energy is even more fluctuating than the wind profile. Solar PV electricity can only be generated when the sun shines, which also happens to contain seasonal variability. The highest peak of the solar PV profile reaches 0,0839% with troughs at night at zero. Solar PV supply profiles are also modelled for the years 2012, 2013, and 2017. Their standard deviations are again displayed in table 3.3. According the standard deviations, the variations of the three solar PV year profiles differ slightly. However, compared to the standard deviations of the wind profiles, significantly more variation is present in the solar PV profiles. The full load hours confirm this statement, as solar PV full load hours are 1.000 hours per year (CE Delft, 2017: 71). This indicates that an energy system solely supplied by solar PV energy, should require larger buffer sizes, if demand is stable. However, it could be possible that a demand profile is better synchronised with the solar PV profile, hence requiring less storage.
TABLE 3.3.- COMPARISON OF YEARLY WEATHER PROFILE VARIATION
Wind Profile Solar PV Profile
Year 2012 2013 2017 2012 2013 2017
Average 0,0114% 0,0114% 0,0114% 0,0114% 0,0114% 0,0114%
Standard Deviation 0,0109% 0,0107% 0,0133% 0,0178% 0,0181% 0,0179%
3.4.3. Heat and Electricity Demand Components
Hourly demand quantities are based on the four demand components: hydrogen- and biogas-hybrid heat pumps; all-electric heat pumps; and electricity for appliances and light. Each scenario contains a different ratio of demand components, with yearly production profiles, i.e. hourly percentages. These demand profiles combined, accumulate to the demand profiles for the three energy carriers of the energy system under investigation, i.e. electricity, hydrogen, and biogas. Figure 3.6. (annual timescale) and figure 3.7. (monthly timescale) visualises the demand profiles for hybrid heat pumps (blue), all-electric heat pumps (red), and appliances and light (yellow).
The hydrogen- and biogas-hybrid heat pump profiles are identical, both represented by the blue line in figure 3.7. Both hybrid heat pump profiles comprise of four annual demand profiles, namely: gas and electricity demand for households; gas and electricity demand for utility buildings. Seasonal (figure 3.6.) and daily (figure 3.7.) fluctuations are present in the profile. Based on CE Delft (2017), households are assumed to consume 71% and buildings 29% of total heat demand. Furthermore, a fixed share of gas and electricity is demanded: 25% gas and 75% electricity. Hence, if 4 MWh is demanded for hour t, 1 MWh gas and 3 MWh electricity is demanded.
0,00% 0,02% 0,04% 0,06% 0,08%
Jan bFe Mar Apr May Jun Jul Aug pSe Oct Nov De
c Ho url y Po rtion (%) Month
Hybrid Heat Pumps All-E Heat Pumps "Electricity"
FIGURE 3.6.- YEARLY HYBRID HEAT PUMP (GAS AND ELECTRICITY) AND ELECTRICITY PROFILES (PERCENTAGE PER HOUR)
FIGURE 3.7.– MONTHLY HYBRID HEAT PUMP (GAS AND ELECTRICITY) AND ELECTRICITY PROFILES (PERCENTAGE PER HOUR)
3.4.4. System Parameters
The projected parameters for 2050 used in this simulation modelling study are as presented in table 3.4. and validated by Gasunie (2019).
TABLE 3.4.-SYSTEM PARAMETERS
Parameter Value
P2G efficiency 70%
G2P efficiency 70%
Transport efficiency for hydrogen and biogas pipelines to the national buffers 98%
Electricity grid capacity in MWh/hour to the national grid 105 MWh/hour
P2G efficiency is based on Bennoua et al. (2015) and Buttler and Spliethoff (2018). Conversion efficiency of G2P is high compared to current technologies, however it is predicted that fuel cells will develop in capacity and efficiency. The electricity grid capacity is based on the current installed high voltage electricity grid capacity dedicated for Midden-Drenthe (Rijkswaterstaat, 2017; Hoogspanningsnet.com, 2019).
The most important parameter, which is simultaneously a result, is the maximum local hydrogen and biogas storage capacities. If these are left unbound, the national storage components will not be used and all surplus or deficit of hydrogen and biogas will be stored and withdrawn from the local buffer. Hence, per experiment and per local buffer, the lowest local buffer capacity is calculated (that is why it is also a result). This is done by using Excel Solver. The optimum local storage capacity is calculated based on the lowest combined local and national storage capacity. For this reason, a pipeline transport efficiency is adopted (98%), to put a constraint on hydrogen and biogas import and export, and favour storing gasses locally.
3.4.5. Assumptions
The next assumptions are made during the simulation study:
o How energy is generated and where it comes from in the national system is omitted from the scope of this research.
o Electricity grid and pipeline transport efficiencies within the local energy system and all storage capacities of pipelines and electricity grids are neglected.
o Supply of biogas is assumed constant.
o Base gas, or cushion gas, of storage components is not considered.
3.5. Operational Rationale of the Simulation Model
With the inputs at hand, the rational of the simulation model can be described. In figure 3.8., the conceptual model is extended with numbers, representing priorities for direction of energy flow. The biogas network is described separately, as it is not connected to the electricity and hydrogen networks. The numbers indicate prioritisation of energy flow. A detailed operational rationale of the simulation model is presented in appendix 3.
Electricity and Hydrogen Priorities:
1. Locally generated electricity is transmitted to local electricity demand of the four demand components, until either demand is fulfilled, or supply is at its maximum.
2. Hydrogen must always be satisfied. This is initially done by the local hydrogen buffer. If demand remains, hydrogen will be imported from the national buffer.
3. Surplus electricity is stored as much as possible in the local hydrogen buffer via P2G. Electricity deficit is compensated for as much as possible via G2P from the local hydrogen buffer.
4. When the local hydrogen buffer is fully utilised and negative residual load remains, electricity is transmitted to the national high-voltage grid. When there is positive residual load, electricity is imported from the national grid to satisfy demand.
5. Finally, hydrogen can be imported for G2P when electricity deficit remains or exported via P2G when the local buffer and national electricity grid cannot store the DRES surplus. Biogas priorities:
I. Produced biogas is stored in the local buffer and demand is withdrawn from it.
II. If the local buffer is fully utilised when there is biogas surplus, biogas is stored in the national buffer. If demand is larger than supply and the local buffer cannot compensate for the deficit, biogas is imported from the national buffer to satisfy demand.
3.6. Model Output Data
The simulation model brings forth hourly output data per scenario, based on the input data over a one-year period. Output data is divided into two categories: hourly system quantities and operational system data. Outputs of the two categories are presented in table 3.5.
TABLE 3.5.– OUTPUT DATA
Hourly system quantities Operational system data
Hydrogen, biogas, and electricity flows Buffer capacities and utilisations
Storage behaviour Conversion and transport capacities and utilisations
P2G and G2P production Conversion and transport losses Annual imported and exported energy 3.7. Experimental Design
Based on the input data, the simulation model, and output data described above, results emerge. These must be analysed in an orderly fashion. Hence, this experimental design is constructed to provide structure and outline which analyses are performed.
3.7.1. Analysis I: Effects of Experimental Variables on Storage Capacities
The experimental variables of this simulation study are determined to investigate their effect on the main system parameter: storage capacity. However, additional outputs, as described in the previous section, will be compared to analyse the overall operational system performance. The experimental variables come in three categories: NVDT scenario (transition pathway); ratio of supply and demand; and weather year profile. Combined, they contain 72 experiments. An overview of the experiments is presented in table 3.6.
TABLE 3.6.– OVERVIEW OF MODELLED EXPERIMENTS
Ratio Supply and Demand
Weather Year Profile NVDTScenario 75% 100% 125% 150% 2012 2013 2017 2012 2013 2017 2012 2013 2017 2012 2013 2017 Regional 1 2 3 4 5 6 7 8 9 10 11 12 National 13 14 15 16 17 18 19 20 21 22 23 24 International 25 26 27 28 29 30 31 32 33 34 35 36 Generic Steering 37 38 39 40 41 42 43 44 45 46 47 48 Optimal Supply 49 50 51 52 53 54 55 56 57 58 59 60 Current Supply 61 62 63 64 65 66 67 68 69 70 71 72
3.7.2. Analysis II: Difference Between Local and National, and Only Local Storage
3.7.3. Analysis III: Two Basic Mismatch Indices (OEF and OEM)
Finally, two basic mismatch indices are analysed, based on Cao, Hasan and Sirén (2013): OEF (on-site energy fraction) and OEM (on-site energy matching). OEF indicates the proportion of demand directly covered by DRES. OEM indicates the proportion of total DRES supply that is directly supplied to demand. Subsequently, the OEF and OEM of experiments are compared to quantify and analyse mismatch in the system.
3.7.4. Simulation Method
The simulation study is performed in Microsoft Excel, as it enabled for high flexibility in modelling, easy access to change parameters and constraints, and the size of the model allowed for using this program. To simulate all experiments, a macro is written with VBA (visual basic for applications), the programming language of Excel. Generated hourly output data is automatically is retrieved by the simulation model. These results are presented and analysed in the following section.
4. RESULTS
Results from the analyses are presented in this section. At the beginning of each subsection, the findings are presented in bullet points, after which they are explained and discussed.
4.1. Analysing the Effects of Experimental Variables on Storage Capacities ( I )
This section elaborates on what impact the different experimental variable inputs have on hydrogen and biogas storage requirements. For this analysis, one ‘reference experiment’ is depicted and used to compare the various experimental variables with. Although the ‘regional’ NVDT scenario has most local focus, it is unfit to act as reference experiment, because it disregards hydrogen demand for hybrid-heat pumps. Hence, a ‘national’ NVDT scenario, with a supply and demand ratio of 100% and weather year profile 2013, is chosen as reference experiment: experiment 17. This scenario contains 10% biogas, 50% wind, and 40% solar PV supply. Demand consists of 62% hydrogen-hybrid heat pump, 27% biogas-hybrid heat pump, 11% all-electric heat pump, and equal demand for electrical appliances and light as the other NVDT scenarios. Required local and national storage capacities of all experiments are displayed in appendix 4. Note that vertical axes of figures with local buffers are presented on a scale of 600 MWh and national buffers 35 GWh.
4.1.1. Aggregated Results for All Experimental Scenarios
Before results for the three experimental variables are presented, two aggregated results will be highlighted, because they apply to all experiments.
For all experiments, hydrogen and biogas local storage capacity is significantly smaller than national storage capacity (on average 1% of national storage capacity). This results from the minimisation function in the simulation model, which searches the optimal local storage capacity, that minimises combined (local and national) storage capacity. To illustrate this, optimal local storage capacity for hydrogen of 0,23 GWh is found in the reference experiment. The sensitivity analysis in figure 4.1. demonstrates the effect of changing this optimal local storage capacity. When the optimal local storage capacity is increased or decreased, combined storage capacity increases.
Optimal local storage capacity results from a trade-off between efficiency losses for conversion and transport. If the local buffer is large, much demand can be satisfied locally, resulting in high conversion losses for P2G and G2P. This increases the local buffer, but only decreases the national buffer slightly. If the local buffer is small, more import and export from and to the national hydrogen buffer is required, resulting in increased transport losses. This increases the national buffer, while the local buffer only decreases slightly. By considering this trade-off, the model found the optimal local storage capacity for each experiment. Hence, optimising local storage capacity can decrease combined storage capacity. Because of this explanation, little attention is given to local storage capacity in the rest of the results.
o Energy balance is accomplished for all experiments with 100% supply demand ratio.
To balance an energy system, the mismatch between supply and demand is covered by storing energy. A balanced system is characterised by equal start and end quantities in its buffers. However, as will be presented later in figure 4.3. and 4.6., no national buffer has these equal quantities. The reason is that demand can be fulfilled from multiple local and national buffers. This study assumes that energy can always be imported from and exported to national buffers. It omits how that energy in the national scope is being treated and where it comes from. Hence, a deficit of one energy carrier or buffer can be compensated by another, as long as the accumulated buffers have equal starting and ending quantities. To demonstrate that the modelled energy system is balanced, and indirectly that simulation model functions appropriately, all annual energy storage
23,98 23,78 23,75 23,76 23,83 23,89 23,70 23,75 23,80 23,85 23,90 23,95 24,00 0,00 0,10 0,20 0,23 0,30 0,40 0,50 1,00 C o mbined st o rag e capa city ( G w h)
Local storage capacity (GWh)
28,45 24,30
capacity behaviours of the reference experiment are accumulated in figure 4.2. Equal start and end quantities (75 GWh) are displayed, indicating a balanced system.
FIGURE 4.2.– ANNUAL TOTAL CUMULATIVE ENERGY STORAGE BEHAVIOUR OF THE REFERENCE EXPERIMENT
4.1.2. Comparing the Six Transition Pathway Scenarios
This subsection analyses the six transition pathway scenarios, based on their required local and national storage capacities. The transition pathways are compared, based on their supply and demand characteristics. First, differences in demand of the four NVDT scenarios of CE Delft (2017) will be analysed, as they vary significantly in demand, while supply is rather equal. Then, differences in supply will be analysed, by analysing the ‘national’ scenario and the two extra scenarios, as they vary significantly in supply, while demand is equal. The main results of this subsection are:
o Scenario ‘current supply’ requires the largest and ‘generic steering’ the smallest hydrogen storage capacity.
o Scenario ‘generic steering’ requires largest and ‘national’ smallest biogas storage capacity. o Local hydrogen storage is absent in the scenarios ‘regional’ and ‘generic steering’, due to
zero hydrogen hybrid heat pumps.
o Supply from only solar PV increases required storage capacity, while optimising the ratio of wind and solar PV supply decreases required storage capacity.
4.1.2.1. Comparing the NVDT Scenarios Regarding Hydrogen Storage Capacity (Similar Supply, Different Demand) Hydrogen storage capacities of NVDT scenarios are presented in table 4.1, including the share of hydrogen-hybrid heat pumps (H-HHP) to satisfy heat demand. The largest and smallest required combined storage capacities are the ‘national’ (23,75 GWh) and ‘generic steering’ (0,50 GWh) scenarios respectively. Highest storage capacity of ‘national’ scenario results from its high share of hydrogen hybrid heat pumps for heating (62% compared to 35% of scenario ‘international’). Scenarios ‘regional’ and ‘generic steering’ require little combined storage capacity, due to the absence of hydrogen-hybrid heat pumps (‘regional’ contains mostly all-electric heat pumps and ‘generic steering’ mostly biogas-hybrid heat pumps). However, hydrogen storage capacity is required, meaning there is demand for hydrogen. This is a result of sectoral integration, i.e., electricity is converted into hydrogen and re-electrified for electricity demand, to overcome supply
0 20 40 60 80 100 120
Jan Feb Mar Apr May Jun Ju
l A u g Se p O ct N o v De c G Wh Months Biogas Buffer (local) Biogas Buffer (national) Hydrogen Buffer (local) Hydrogen Buffer (national) Electricity Buffer (national)
‘EQUAL START AND END QUANTITY’
and demand mismatches. Hence, some hydrogen storage capacity is required. These results indicate that required combined hydrogen storage capacity is mostly dictated by hydrogen-hybrid heat pumps and slightly by (re)converting hydrogen. Local storage capacities are absent at scenarios ‘regional’ and ‘generic steering’. This shows that local storage capacity is undesirable, when hydrogen demand is low or absent.
TABLE 4.1.– HYDROGEN STORAGE CAPACITIES FOR THE NVDT SCENARIOS
Scenario configurations Hydrogen storage capacities (GWh)
Experi-ment
NVDTscenario Ratio supply/ demand
Weather year
H-HHP
share
Local National Combined
5 Regional 100% 2013 0% - 2,74 2,74
17 National 100% 2013 62% 0,23 23,52 23,75
29 International 100% 2013 35% 0,23 11,67 11,90
41 Generic Steering 100% 2013 0% - 0,50 0,50
4.1.2.2. Comparing the ExtraScenarios Regarding Hydrogen Storage Capacity (Different Supply, Equal Demand) The extra NVDT scenarios (‘optimal supply’ and ‘current supply’) are compared to the ‘national’ scenario in table 4.2., because they contain different supply but equal demand. ‘Current supply’, containing only solar PV supply for electricity, requires the largest combined storage capacity: 46% increase. This means that a system solely supplied by solar PV electricity, requires a larger storage capacity. As previously expected in section 3.4.2.3., this is due to more variability in solar PV supply than in wind or a combination. Scenario ‘optimal supply’ requires the smallest storage capacity: 7% decrease. The simulation model sought the optimal ratio between wind and solar PV, which minimised storage capacity. This result shows that required storage capacities can be decreased by optimising the ratio of wind and solar PV, for experiment 53, 57% solar PV and 33% wind (biogas has a fixed 10%). However, the ratios vary per experiment, which are all displayed in appendix 5.
TABLE 4.2.- HYDROGEN STORAGE CAPACITIES FOR THE EXTRA SCENARIOS
Scenario configurations Hydrogen storage capacities (GWh)
Experi-ment
NVDTscenario Ratio supply/ demand
Weather year
Local National Combined
17 National 100% 2013 0,23 23,52 23,75
53 Optimal supply 100% 2013 0,17 21,82 21,99
65 Current supply 100% 2013 0,59 34,08 34,66
FIGURE 4.3.- ANNUAL LOCAL HYDROGEN STORAGE BEHAVIOURS OF THE TRANSITION PATHWAY SCENARIOS
FIGURE 4.4.– MONTHLY LOCAL HYDROGEN STORAGE BEHAVIOURS OF THE TRANSITION PATHWAY SCENARIOS
In figure 4.5., lines of the buffers descend during winter months (hydrogen is withdrawn, i.e., imported) and ascend during summer (hydrogen is injected, i.e., exported). Import is caused by an abundance of energy deficit and export by surplus. Even though demand and supply totals of the ‘national’, ‘optimal supply’, and ‘current supply’ scenarios are identical, only the ‘current supply’ scenario exhibits large hydrogen export (34 GW total). This finding can be explained by the difference in wind and solar PV ratio. High variability of solar PV supply in scenario ‘current supply’ causes extreme quantities of surplus in summer months, resulting in much exported hydrogen.
FIGURE 4.5.- ANNUAL NATIONAL HYDROGEN STORAGE BEHAVIOURS OF THE TRANSITION PATHWAY SCENARIOS
4.1.2.3. Comparing the NVDT Scenarios Regarding Biogas Storage Capacity (Similar Supply, Different Demand) Biogas storage capacities of the NVDT scenarios are presented in table 4.3, including the share of biogas-hybrid heat pumps (B-HHP) to satisfy heat demand. An annual visualisation is presented in figure 4.6. The extra scenarios, ‘optimal supply’ and ‘current supply’, are identical to scenario
‘national’ and therefore not presented. The four NVDT scenarios have similar supply but contain different shares of biogas-hybrid heat pump demand. Total demand is identical.
The largest and smallest required combined storage capacities are the ‘generic steering’ (32,10 GWh) and ‘national’ (8,19 GWh) scenarios respectively. These are also the scenarios containing the highest and lowest share of biogas-hybrid heat pumps. This result shows that biogas storage capacity rises proportionally to the share of biogas-hybrid heat pumps. This can be explained by the closed system of biogas with one supply and one demand component. Supply is steady and therefore buffers are mostly influenced by demand. Hydrogen buffers do not behave proportional to total demand. The first difference is that the hydrogen network is subject to sectoral integration, i.e., electricity can be converted into hydrogen, from which hydrogen can either be supply to hydrogen-hybrid heat pumps or re-electrified via G2P. Secondly, hydrogen production is more variable, due to variability in wind and solar pv supply. This shows that systems with sectoral integration and variable supply, imply less predictable buffer behaviours.
TABLE 4.3.- BIOGAS STORAGE CAPACITIES FOR THE NVDT SCENARIOS
Scenario configurations Biogas storage capacities (GWh)
Experi-ment
NVDTscenario Ratio supply/ demand
Weather year
B-HHP
share
Local National Combined
5 Regional 100% 2013 31% 0,26 8,96 9,22
17 National 100% 2013 27% 0,23 7,96 8,19
29 International 100% 2013 60% 0,56 19,05 19,61
41 Generic Steering 100% 2013 99% 0,91 31,19 32,10
FIGURE 4.6.- ANNUAL NATIONAL BIOGAS STORAGE BEHAVIOUR FOR THE NVDT SCENARIOS
4.1.3. Comparing the Four Supply and Demand Ratios
This subsection analyses the four supply and demand ratios (75%, 100%, 125%, and 150%), based on their required local and national storage capacities. The percentage indicates the amount of annual supply relative to annual demand. Hence, the system can be analysed under conditions of annual energy surplus and deficit. Biogas results are omitted, as similar results arose as the previous subsection: biogas storage capacity rises proportionally to the increase in supply and demand ratio. The ratios are analysed for scenario ‘national’ and ‘international’, because of surprising differences in results. The main results of this subsection are:
o Largest combined hydrogen buffer capacity at 75% ratio, in the ‘national’ scenario. o Largest combined hydrogen buffer capacity at 150% ratio, in the ‘international’ scenario. o Smallest combined hydrogen buffer capacity in both cases at 125% ratio.
o Electricity demand mostly being satisfied locally and hydrogen demand nationally. 4.1.3.1. Comparing the Supply and Demand Ratios Regarding Hydrogen Storage Capacity
Hydrogen storage capacities of supply and demand ratio experiments are presented in table 4.4. The largest required combined hydrogen storage capacity in scenario ‘national’ is at 75% ratio, while the largest storage capacity in scenario ‘international’ is at 150% ratio. Because supply of both scenarios is nearly similar, the difference is mainly caused by demand. Scenario ‘national’ contains 62% hydrogen-hybrid heat pumps in its heat demand, while scenario ‘international’ contains 35%.
TABLE 4.4.- HYDROGEN STORAGE CAPACITIES FOR THE SUPPLY AND DEMAND RATIOS
Scenario configurations Hydrogen storage capacities (GWh)
Experi-ment
NVDTscenario Ratio supply/ demand
Weather year
Local National Combined
14 National 75% 2013 0,05 28,93 28,98 17 National 100% 2013 0,23 23,52 23,75 20 National 125% 2013 0,53 15,76 15,57 23 National 150% 2013 0,01 19,41 19,42 26 International 75% 2013 0,06 16,26 16,32 29 International 100% 2013 0,23 11,67 11,90 32 International 125% 2013 0,43 10,40 10,83 35 International 150% 2013 0,11 23,99 24,10
FIGURE 4.7.– LARGEST AND SMALLEST ANNUAL NATIONAL HYDROGEN STORAGE BEHAVIOUR FOR THE SUPPLY/DEMAND RATIO EXPERIMENTS
4.1.4. Weather Year Profile
Finally, results of the different weather year profiles are presented, with the main finding: o More variability in supply profiles does not necessarily result in larger buffer capacities. The prediction in section 3.4.2., which said, “scenarios for 2017 should require a larger buffer capacity due to the highest variability in supply”, is proven to be invalid. As can be observed in table 4.5., the combined storage capacity of weather year profile 2017 is smallest (22,26 GWh). This can be explained by better synchronisation of the supply and demand patterns in 2017, than 2012 and 2013, resulting in a smaller combined storage capacity.
TABLE 4.5.- HYDROGEN STORAGE CAPACITIES FOR THE WEATHER YEAR PROFILES
Scenario configurations Hydrogen storage capacities (GWh)
Experi-ment
NVDTscenario Ratio supply/ demand
Weather year
Local National Combined
16 National 100% 2012 0,21 23,56 23,76
17 National 100% 2013 0,23 23,52 23,75
18 National 100% 2017 0,36 21,90 22,26
4.2. Analysing the Difference Between Local and National, and Only Local Storage ( II )
This analysis demonstrates the effects of national storage capacity, by comparing a system with and without a national energy system. The reference experiment (experiment 17) including national hydrogen, biogas, and electricity buffers, is compared with an additional experiment (experiment 73), neglecting national buffers, while all else is equal to experiment 17. The main result is:
o An energy system without national energy storage requires smaller total storage capacities (hydrogen, biogas, and electricity), but a larger combined hydrogen storage capacity. The results for hydrogen storage are presented in table 4.6. Local storage for a system with and without national buffer is 0,23 GWh and 103,48 GWh respectively. Combined storage requirement to balance the energy system is 23,75 GWh for the system with national buffer (scenario 17) and 103,48 GWh without national buffer (scenario 73). That is a difference of nearly 80 GWh. This difference is compensated by the national electricity grid in scenario 17. If import and export from the national electricity grid would come from an unrealistically large battery, its
capacity would require 100,70 GWh. Accumulating all local and national buffers of scenario 17, larger total storage capacity is required than only local storage (21 GWh more). This is because an ‘only local storage’ system requires more installed capacity, as will be explained later, and therefore can satisfy a larger share of demand directly from DRES. This results in less demand that must be satisfied from storage. However, hydrogen storage capacity of a system without national influence is significantly larger.
TABLE 4.6.- HYDROGEN STORAGE CAPACITIES WITH AND WITHOUT NATIONAL STORAGE CAPACITY
Scenario configurations Hydrogen/electricity storage capacities (GWh) Experi-ment NVDT scenario Ratio supply/ demand Weather year Storage locations Local hydrogen National hydrogen Combined hydrogen National electricity
17 National 100% 2013 Local & national
0,23 23,52 23,75 107,70
73 National 100% 2013 Local 103,48 - 103,48 -
Required biogas storage capacities for a balanced local energy system are presented in table 4.7. Local biogas buffers for scenarios 17 and 73 are 0,23 GWh and 8,11 GWh respectively. National buffer requirement of scenario 17 is 7,96 GWh. Combined biogas storage capacities result in a slightly larger required storage capacity in the system with national buffers (8,19 GWh).
TABLE 4.7.– BIOGAS STORAGE CAPACITIES WITH AND WITHOUT NATIONAL STORAGE CAPACITY
Scenario configurations Biogas storage capacities (GWh)
Experi-ment NVDT scenario Ratio supply/ demand Weather year
Storage locations Local National Combined
17 National 100% 2013 Local & national 0,23 7,96 8,19
73 National 100% 2013 Local 8,11 - 8,11
The ‘only local storage experiment’ requires more DRES supply from wind and solar PV for a balanced system (scenario 17: 299 GWh and scenario 73: 377 GWh) and therefore more installed capacity, as presented in table 4.8. More DRES supply is needed to compensate for the increase in conversion losses, due to an increase in electricity and hydrogen conversion to satisfy electricity demand locally. For experiment 17 and 73, P2G conversion losses are 64 GWh and 14 GWh and the G2P conversion losses are 35 GWh and 7 GWh respectively. Note that the difference between these conversion losses of the two experiments are equal to the difference in DRES supply. These results indicate that, a system with only local storage requires smaller total energy storage capacity, but more DRES supply is needed to compensate for the increase in conversion losses.
TABLE 4.8.– INSTALLED DRES CAPACITIES WITH AND WITHOUT NATIONAL STORAGE CAPACITY
Scenario configurations Installed DRES capacity (MW)
Experi-ment NVDT scenario Ratio supply/ demand Weather year
Storage locations Wind Solar PV Biogas
17 National 100% 2013 Local & national 55,4 133,9 1,7
73 National 100% 2013 Local 69,9 167,7 1,6
4.3. Analysing Two Basic Mismatch Indices ( III )
As described before, Cao, Hasan and Sirén (2013) present two commonly used indices to assess the matching between DRES supply and household and building demand, notated by OEF (on-site energy fraction: proportion of demand directly covered by DRES) and OEM (on-site energy matching: proportion of total DRES supply that is directly supplied to demand). Better matching indices are indicated by high values of OEF and OEM, with 0% ≤ OEF & OEM ≤ 100%. This study adopted these indices to quantify and analyse the mismatch problems of electricity and biogas in the local energy system. Hydrogen mismatch could not be calculated, as hydrogen is not produced directly from DRES.This section addresses the most interesting results, whereas the matching indices of all experiments can be found in appendix 6. The main results are:
o Increase in energy surplus results in higher OEF and lower OEM for electricity and biogas. o Scenario ‘optimal supply’ contains the best matching indices and ‘current supply’ worst.
Most notable differences in matching indices are found for variations in the energy system’s ratio of supply and demand for electricity, as presented in table 4.9. OEF increases, as the ratio of supply and demand rises, whereas OEM decreases. The logical explanation is that, the more DRES is installed, larger share of demand can be directly supplied by DRES (higher OEF). Simultaneously, even though the amount of DRES that is directly supplied to demand increases, the share of total supply from DRES that can directly be used for demand, decreases (lower OEM). The reversed effect takes place when the supply and demand ratio decreases. Equal effects are found for the biogas system. To further express the meaning of OEF and OEM and to visualise the mismatch in the researched energy system, the principle of the two indices are presented in figure 4.8, based on the reference experiment. The two lines represent electricity demand (dotted) and DRES supply (continuous). OEF equals the ratio of C to the combined areas of A and C. OEM equals the ratio of C to the combined areas of B and C.
TABLE 4.9.– MATCHING INDICES FOR ELECTRICITY MISMATCH FOR DIFFERENT SUPPLY AND DEMAND RATIOS
