Exploring a viable stand-alone ﬂexible multi-commodity energy system in the North of the Netherlands

(1)

flexible multi-commodity energy

system in the North of the

Netherlands

Pieter Minnee

A thesis presented for the degrees of

MSc. Technology & Operations Management

and

MSc. Operations & Supply Chain Management

University of Groningen - Supervisor: prof. dr. ir. J.C.Wortmann Newcastle University - Supervisor: prof. dr. A. Small, Msc

Netherlands and United Kingdom Date: 9-12-2019

Petrus Driessenstraat 4a, 9714CB Groningen p.h.minnee@student.rug.nl - P.Minnee2@newcastle.ac.uk

S2741474 - B80588606

(2)

(3)

commodity energy system (MES) in the North of

the Netherlands

Pieter Minnee

Abstract

To meet climate goals of 2030, an energy transition in the north of the Netherlands is required. Meeting the climate goals requires an increase of renewable energy. Converting green electricity to hydrogen is seen as a solution to counter renewable intermittency. The North of the Netherlands exhibits capabilities of transforming the current gas infrastructure into a flexible multi-commodity energy system using hydrogen and electricity and is therefore used as the area of focus of this study. This study focuses on creating viable business models using different forms of flexibility to counter intermittency. Based on different investment decisions and simulation results using different parameters, business cases were setup. From here the busi-ness models were validated through an expert panel in terms of viability regarding technical, stakeholder, scale-ability and evolve-ability. Circumstances under which a flexible multi-commodity energy system is viable are mentioned, together with limitations and topics for future research.

(4)

Preface

After five years of enjoyful studying, I present to you my final dissertation. This thesis is the final chapter of my Master’s degree at the University of Groningen and Newcastle University.

First of all, I would like to thank my supervisors for their useful feedback the past half year. It has been a struggle sometimes to get structure and logic in the thesis, but you provided me with useful ideas helping me throughout the process.

Furthermore I am grateful for the support I gained from employees at Groningen Seaports. They were very helpful with data and delivering capabilities to look at this complex issue with all the involved stakeholders. Henk, thank you for supervising me throughout this thesis and also for introducing me to the right persons within the industrial cluster. The ”zout aan de broek” experience was a nice opportunity to physically see all the processes in Delfzijl and Eemshaven.

Next, I would like to thank Austin D’Souza for providing me with the model logic and the proper framework to tackle this challenge. Christian van Someren for help-ing me out with the simulation and the experts for their time evaluathelp-ing the business model elements.

I hope my thesis will eventually solve a little piece of the ”climate puzzle”. This by giving insight into the value of flexibility from where out policy makers could design a viable, sustainable future.

At last, I would like to thank my family and friends for giving me unconditional support during the whole process.

Pieter Minnee

(5)

List of Figures

1 Business model design framework for viability (D’Souza et al., 2018) . 17 2 Research design for BM’s (Bouw et al., 2015) . . . 23

3 Basic business model concept 1: MES hydrogen-electricity . . . 28

(7)

5 BM concept 3: MES hydrogen-electricity + Balancing market . . . . 30 6 BM concept 4: MES hydrogen-electricity + Balancing market + buffer 31 7 Basic Procurement policy based on a price threshold (U) (APX) . . . 35 8 Expected average production and consumption pattern industrial

clus-ter in kg/hour (2022). . . 36 9 Retrieved from TenneT (2018); APX (2018) . . . 38 10 Mulder et al. (2019); Wijk (2017) linear hydrogen cost (Euro/kg) and

the corresponding electricity price (e/MW) . . . 39 11 Expected scaling of electrolysis capacity in Delfzijl (Industrietafel,

2019) . . . 48 12 Potential line packing routes connecting industrial clusters and the

salt cavern (Expert 1 and case company representative). . . 50 13 Possible e3 value model with external aggregator and PRP - Value

derived by external aggregator should decrease to increase viability for the whole BE. Fuel cell investment was seen not viable to sell electricity back to the market. These actors should therefore either disappear (electricity producers via fuel cell) or acquire less value (External Aggregator) . . . 64 14 Preliminary Information architecture with PRP and External

aggre-gator (BM4) . . . 65 15 Unique characteristics of the Northern Netherlands to exploit

flexibil-ity using the old gas infrastructure (red lines) to transport and store hydrogen. Using renewable flows from neighbouring countries, hydro-gen as an energy carrier can be formed with electrolysis in Delfzijl or Eemshaven. - Groningen Seaports . . . 67 16 Output Solver Microsoft VBA (BM2) . . . 74 17 Fuel cell option break-even requirements to sell back electricity to

external markets . . . 78 18 Short term hydrogen cost price estimation based on cost saved from

flexibility . . . 79 19 Possible evolve-ability options for the industrial cluster according to

Kats (2018) . . . 83

List of Tables

1 The particular steps obtaining a viable BM (D’Souza et al., 2015) . . 18 2 Summary of evolve-ability options for a hydrogen-electricity MES . . 20 3 Summarised BM evaluation criteria from the literature . . . 21 4 Summarized stakeholder analysis based on (D’Souza et al., 2018) . . . 26 5 Basic BM elements using flexibility . . . 27 6 Overview of experts interviewed for brainstorming and validation in

this research . . . 33 7 Description of the variables - based on Kroniger and Madlener (2014). 37 8 Prices external electricity markets (Euro/MWh) (Nordpoolgroup 2018)

38

9 Explanation hydrogen price model variables based on Wijk (2017) . . 40 10 Potential of using flexibility to decrease the cost price (Wijk, 2017)

(8)

11 Business case business model 2 (Day-ahead trading with extra storage

flexibility) . . . 42

12 Worst case, most likely case and best case scenarios for business model 2 43 13 Business case business model 4 (Balancing market with extra storage flexibility . . . 44

14 Worst case, most likely case and best case scenarios for business model 4 45 15 Possible value distribution (BM 2)* . . . 47

16 *Based on consumption by hydrogen industrial consumers, excess ca-pacity is used as starting point to store and discharge hydrogen as flexibility. . . 49

17 Line packing opportunities in the North of The Netherlands . . . 50

18 Preliminary business case summary (BM2) for scaling viability . . . . 51

19 Results of Evolve-ability of the MES . . . 52

20 Viability rating of the evaluation criteria (table 3 . . . 53

21 Possible value distribution (BM 4 best case) with less value for exter-nal aggregator . . . 65

22 BM design elements creating the design choices for the BM concepts (D’Souza et al., 2018) . . . 66

23 Characteristics Hydrogen route . . . 70

24 Electrolyser specifications . . . 71

25 Description of the variables used in equation 7 and 8 . . . 73

26 External electricity markets, based on D’Souza et al. (2018) . . . 73

27 Data collection according to the ethical standards (Karlsson (2016);Easterby-Smith et al. (2012)). Performed during data collection in this thesis. 75 28 Possible Business case scaling 2025 . . . 80

29 Possible Business case scaling 2035 . . . 81

(9)

1 Introduction

Abbreviations

APX Amsterdam Power Exchange - day-ahead market (NL) AQR Analytical Quantitative Research

BE Business Ecosystem

BMDFV Business Model Design Framework for Viability CAPEX Capital Expenditures

DSO Distribution System Operator MES Multi-commodity Energy System OPEX Operational Expenditures

PRP Programme Responsible Party TSO Transmission System Operator

Recent debates have raised awareness for members of the European Union to agree on new climate interventions. The Climate agreement in the Netherlands has set a target to reduce greenhouse gas emissions by 49% for 2030, compared to 1990 (In-dustrietafel, 2019). Industries in the North of the Netherlands should also cooperate to achieve this national target. The “Klimaattafel” (Climate table) is an initiative from the Northern provinces in the Netherlands concerning 31 industry type organ-isations that aim to reduce carbon dioxide (CO2) emissions. To achieve this goal,

new processing techniques and circular processes within the industries have to be implemented. According to Mohlin et al. (2018) , the cost of renewables will only decrease over time, allowing researchers and policy makers to investigate their role for in the near future. The challenges of renewables are considered rather more complex than their fossil counterparts (Alanne and Saari, 2006). When technologies and scalability tend to advance over time, infrastructure and services needed for conversions should be adjusted. Therefore energy systems containing high renew-able shares should remain flexible (Alanne and Saari, 2006);(Steinke et al., 2013). To make the situation even more complicated, the renewables are not only demand driven, but also supply driven (Alanne and Saari, 2006). Due to the higher inter-mittency of renewables compared to fossil energy, it is essential for the renewable energy system to be flexible in mitigating disturbances (Ulbig and Andersson, 2015). Therefore, the art of balancing these rapid changes in renewable generation and fore-cast errors could be seen as a flexible capability of an energy system (Kondziella and Bruckner, 2016). Balancing is seen as the capability of an energy system to adjust itself from potential shortcomings or abundances for a commodity by regulating the flow between supply and demand (Koirala et al., 2016). According to Mathiesen et al. (2015) flexibility can be created by balancing energy via storage or conver-sions. Here, synergy can be established between different energy sectors by trading in commodities (Shi et al., 2016).

(10)

longer periods of intermittencies (Moseley and Garche, 2014). Hydrogen (H2) as

an energy carrier can be seen as a solution for the long-term intermittencies when for instance low wind speeds or solar insolation are encountered for longer periods of time (Moseley and Garche, 2014);(eu2018, 2018). Next to storage flexibility, hydrogen also enhances energy systems in delivering conversion flexibility (O’Malley and Kroposki, 2017).

Conversion flexibility is seen as improving the energy efficiency of a MES by using different types of energy that can be converted to adjust for certain sup-ply and demand requirements needed either internally or externally of the system (Blaauwbroek et al., 2015).

O’Malley and Kroposki (2017) describe that energy system integration is often discussed, however using conversion flexibility to sell or trade abundant energy in an internal system, back to the external electricity market or other energy sectors is yet to be done for hydrogen.

The current problem is that approximately 95% of the hydrogen production is produced from fossil sources, which is considered “grey” (Hosseini and Wahid, 2016). To reach the climate goals of 2030, the amount of ‘green’ hydrogen (renewable energy sources) production should increase. To be considered green, a system of certificates has been implemented in Europe, using track-and-trace technology (Mulder et al., 2019).

To increase the amount of hydrogen generation that can be considered ‘green’, to electrolyze is seen as a suitable option (Vivas et al., 2018). With the process of electrolyzation, hydrogen can be created using water and abundant energy. This does require incentives for highly flexible power plants, storage and demand-side responses. The energy system should also remain reliable, robust, resilient and affordable for all parties involved (Ulbig and Andersson, 2015).

The area of focus for this energy transition study will be industries stationed in the industrial cluster in Delfzijl and the stakeholders of Groningen Seaports. Groningen Seaports invested in a new hydrogen backbone of four kilometres, able to transport hydrogen among the industrial hydrogen consumers around Delfzijl. A business model exploring viability options on how to use the hydrogen backbone is interesting for the focal company and its stakeholders (D’Souza et al., 2018). Schol-ars currently lack consensus on what type on business model ontology is considered the optimal way of performing business since they often focus on different business perspectives each (D’Souza et al., 2015);(Taran et al., 2015). To solve this issue, an approach is used by continuously testing different building blocks, having an iterative approach with different scenarios (see Literature Review section).

The industrial cluster in Delfzijl exhibits capabilities to trade commodities in-ternally, but also externally with electricity. With the increase of hydrogen demand (Mansilla et al., 2012), this thesis will address using hydrogen as the tool to create flexibility in the industrial park to counter fluctuations in supply among renewables. This by using storage and conversion opportunities to deliver flexibility.

(11)

deliver value for all stakeholders involved. Therefore a value stream concerning a trading system of hydrogen conversion should be investigated and modelled to discover under which circumstances, a flexible energy system might be viable for Groningen Seaports and its stakeholders. As the facilitator, Groningen Seaports should setup services that are operationally flexible for mitigating fluctuations in the energy system. Makarov et al. (2009) characterises operational flexibility of an energy system as power, energy, ramp-rate and ramp duration. Eriksson and Gray (2017) argue that involving all stakeholders into the proposed new energy system is needed to be operationally flexible.

D’Souza et al. (2018) performed research based on multiple commodities energy systems (MES) taking in consideration involved stakeholders, however did not in-clude hydrogen. In order to couple MES with each other, conditions under which possible volumes and prices commodities can be exchanged viably should be explored (Biegel et al., 2014). Using conversion and storage as source of flexibility, hydrogen can be used as a flexible tool to trade energy internally between consumers and producers (Koirala et al., 2016). Hydrogen is also capable to be used different en-ergy systems and could be coupled with other commodities on the external market (Krieger et al., 2016). Krieger et al. (2016) describe the opportunities of using heat exchange systems as a possibly energy source to provide electrolysis and vice-versa using energy from the fuel cell to generate heat when needed. The coupling of sev-eral energy systems using conversions to enhance flexibility is interesting and should be further investigated (Mathiesen et al., 2015).

The literature leaves a gap on what might be the best practice to create such a flexible hydrogen system within the energy provision. Investigating the roles in the value stream and how stakeholders interact could be interesting (D’Souza et al., 2018). A viable business model (BM) therefore has to be created. To do so, this thesis uses the business model design framework for viability (BMDFV) (D’Souza et al., 2018). The transition to a multi-commodity energy-hydrogen energy sys-tem with the interplay of stakeholders and value exchange relationships should be modelled. Using the BMDFV, a quantitative approach is used to provide multiple business models. Each business model investigates possible opportunities for in-volved stakeholders to invest in new coupling of energy systems, in order to mitigate disturbances and create flexibility. The research is performed on a concrete case in the industrial cluster, which exhibits all the features needed to research the MES. This allows the possibility to answer the following research question:

“Under which market circumstances is it viable for the involved stake-holders to invest in a hydrogen multi-commodity energy system? Taking into account the value streams and interactions between stakeholders.”

(12)

2 Literature review

This literature review consists of five parts explaining the necessary components of a flexible MES in currently known in the literature.

At first, the energy market is explained. Based on D’Souza et al. (2018) extra value could be derived ’coupling’ the internal market with external electricity market under the right conditions. This could increase the viability of the MES.

Secondly, the conversion flexibility is explained in terms of electrolysis and fuel cell options to provide green commodity exchanges.

This is followed by storage capabilities of a hydrogen-electricity energy system. Storage is seen as the default way to increase flexibility in a MES with high con-centrations of renewables (Alizadeh et al., 2016). Therefore this tool is further elaborated in this literature background.

From here scaling and evolve-ability possibilities for other commodities that might be capable of joining the MES are discussed. This to elaborate the cer-tain ’coupling’ possibilities of the MES which might add viability (Mathiesen et al., 2015).

In the fourth part viability is explained in terms of the Business Model Design Framework for Viability (BMDFV) (D’Souza et al., 2015). The principle of allowing iterations in different business model perspectives to eventually create alignment from different perspectives allows the designer eventually to create a more coherent business case for all parties involved (D’Souza et al., 2015). This could suite the intentions from the focal company(s) to attract investments from the BE to make the MES more viable. The BMDFV is therefore chosen in this study and further elaborated.

The last part consists of the criteria on which the business model concepts will be validated in order to be considered viable.

2.1 The energy market

In order for the industrial park to exploit flexibility of the upcoming hydrogen route, the business ecosystem should allow all stakeholders to be connected within the internal energy market (D’Souza et al., 2018). Interactions between stakeholders and the value exchange relations between them should be coordinated and further extended with extra stakeholders when considering trading on the external electricity markets. (D’Souza et al., 2018).

The internal energy market

(13)

business ecosystem (Van der Burg et al., 2019). An internal aggregator could also allow capacities of commodities to be traded based on different time slots of hydrogen use by consumers. In exchange of providing flexible production schedules, consumers could on their turn get compensation in the form of more affordable renewables (Bouw et al., 2015). By facilitating in curtailment reduction services, the aggregator increases flexibility of the MES (Liu et al., 2015). Hence increasing its capabilities to react to the energy market fluctuations (Liu et al., 2015).This is especially useful in energy systems with renewable mixes (PV, Wind) above 20-30% (Huber et al., 2014). The need for flexibility and aggregator’s also increases when energy systems have to be scaled and permissible use of energy is increased (Steinke et al., 2013).

The external electricity market

The conventional external electricity market is recognised as a few player market, in which prices do change every minute, however major suppliers depict the prices based on a margin(Forward market) (Grimsrud et al., 2014). Biegel et al. (2014) stated that there exist three different energy markets which contain renewables: day-ahead, intraday and an ancillary service market (balancing). To use the external electricity markets, conversion flexibility is needed between hydrogen and electricity. In order to trade electricity on the energy markets, the hydrogen first has to be converted into electricity or vice-versa (Schmidt et al., 2017).

During periods of high intermittency, a MES could benefit from conversion flex-ibility on the external electricity market, reacting to certain deficits or abundances in the different markets. (Merkert et al., 2015)

2.1.1 Day-ahead market

On this market one can submit bids in certain blocks of 1 hour in the day-ahead and physically receive electricity the day after. Due to their stability, the volatility of price changes could be considered rather small (Sioshansi et al., 2009). Due to their stability however, the energy market is mainly using day-ahead biddings in their system (Biegel et al., 2014). In the past the day-ahead market was considered sufficient for performing all trades, due to their relative slow price fluctuations due to a low share of renewables. (Philipsen et al., 2019). In the near future, the share of renewable energy in the mix on the energy market is likely to increase (Mansilla et al., 2012);(Mohlin et al., 2018). This results in price estimations on the day-ahead market that turn out to be inaccurate (Philipsen et al., 2019). This since biddings are often based on erroneous assumptions about wind power and solar insulation the day after and could therefore lead to either redundant commodities or shortages the day after.

2.1.2 Intraday market

(14)

2.1.3 Balancing market

The renewable character of intermittency (Ulbig and Andersson, 2015) causes certain price fluctuations within seconds (Biegel et al., 2014). This requires a service which mitigates price fluctuations of the biddings till the last second, hence the ancillary service market (balancing market) (Biegel et al., 2014). On the balancing market, the Transmission System Operators (TSO) facilitate Programme Responsible Parties (PRP) to trade on a market which allows an extra trade position next to the day-ahead and intraday markets. The possibility of purchasing electricity at low or even negative prices (e/MWh) on the balancing market and later on sell during high price periods to the same market, could be lucrative and therefore taken into consideration in this research (M¨oller et al., 2011).

Transmission System Operators - The main task for TSO’s is to guarantee balance on the grid, therefore preventing blackouts by reducing congestions and shortages caused by intermittency or other irregularities (Schermeyer et al., 2018). Congestion of the electricity grid has been tripled since 2014 due to the upscaling of renewables (Schermeyer et al., 2018). During periods of high wind power or solar insolation, renewable energy has to be curtailed in order to keep the grid balanced at a frequency of 50 Herz. In order to maximise output and minimise curtailment from renewables, TSO’s ask Programme Responsible Parties (PRP) to provide balancing services in exchange for monetary value (Gerard et al., 2016).

Programme Responsible Party - The PRP is responsible for submitting e-programmes to the TSO. These e-programmes specify the amount of electricity the PRP’s customer, hence the industrial cluster in this study, expects to take or put back on the grid based on their production schedules. At the end of the day, the actual consumption from the PRP’s customer is measured and submitted to the TSO (TenneT, 2019). Metering information about actual consumption is sent first to a Distribution System Operator (DSO) before it is redirected to a TSO. Not every party is allowed to fulfill the PRP’s function. Due to legal constraints every party acting as a PRP has to be licensed in order to trade on the balancing market (Gerard et al., 2016). It should be notified that compared to the European markets, the Dutch balancing market is considered an open-market with no price influences from TSO’s (Appendix 6.2 Expert 2 ). The purchased electricity from this market could be used in an electrolyser to gain additional hydrogen, reducing its dependency on current suppliers of hydrogen to the route. Redundant hydrogen could vice-versa put in a fuel cell to be sold on the electricity markets when lucrative. Status regarding flexibility, prices and buy or sell orders desired should be performed by an external aggregator on an external trade platform.

External aggregator - The external aggregator could use the information from the internal energy control system to negotiate with PRP’s about possible trades (D’Souza et al., 2018). Trading on the balancing market requires an energy sys-tem that is capable of being even more flexible compared to the day-ahead market (Klaassen et al., 2018). The business ecosystem should therefore remain flexible in order to react to more frequent balance activities (Klaassen et al., 2018).

(15)

opportunities of using flexibility to attract value from these fluctuations is interesting and methods to exploit flexibility are discussed below.

2.2 Conversion flexibility

The first option to exploit flexibility is using conversion flexibility using electrolysis. Here water and (renewable) electricity are used to form oxygen and green hydrogen. See equation 1

2H2O(l)+e−=⇒ O2(g)+ 2H2(g) (1)

Schmidt et al. (2017) stated that three types of electrolysis could be used for conversion of (renewable) electricity to hydrogen. The best suitable electrolysing type currently on the market to create flexibility is the PEM (Polymer Electrolyte Membrane) electrolysis technique (Buttler and Spliethoff, 2018). This electrolyser offers fast ramp-up – ramp down rates and the production rate can be varied more easily compared to the more mature electrolysers like the alkaline types (Buttler and Spliethoff, 2018). Prices on the balancing market could change within four seconds, therefore flexible electrolysers are preferred when trading on the balancing market (Bouw et al., 2015). PEM does require higher investment costs, while the more developed technology of an alkaline electrolyser is capable to perform balancing activities as well considering most trades (Buttler and Spliethoff, 2018). The third type, Solid oxide electrolysis (SOE), has a potential of 100% efficiency, however this technology is not fully developed yet. Therefore SOE is not included in this research. Steam Methane Reforming - Another option to produce hydrogen is Steam Methane Reforming (Mulder et al., 2019). Reacting methane (CH4) with steam

(H2O(g)) to form grey hydrogen and carbon dioxide. See equation 2.

CH4(g)+ 2H2O(g) =⇒ 4H2(g)+ CO2(g) (2)

Grey SMR is needed in this research since it is capable of producing twice af-fordable hydrogen than the electrolysis method (Mulder et al., 2019). Furthermore, electrolysis is not capable of satisfying all hydrogen demand till 2050 (Wijk, 2017). Therefore, ’grey’ hydrogen from SMR or blue hydrogen, which is could still be ob-tained as a side product from chlorine production, is still needed in a MES including hydrogen (Mulder et al., 2019). For this reason grey and blue hydrogen sources have to be included in this research (Wijk, 2017).

Fuel cell - The capability to provide electricity back to the grid could be done using a fuel cell (see equation 3).

O2(g)+ 2H2(g) =⇒ 2H2O(l)+e− (3)

(16)

2.3 Storage flexibility

Hydrogen tank

Next to supply of energy from the balancing activities facilitated by an aggrega-tor, excess of potential hydrogen could be stored, if beneficial to increase flexibility. (Clegg and Mancarella, 2016). Kondziella and Bruckner (2016) performed literature review on the flexibility requirements of renewable electricity systems. Hydrogen en-ergy systems often use storage as a default way to gain flexibility. This since research tends to be focused on future grid expansions and avoiding PV/wind curtailment by maximising all possible gains by converting as much electricity in hydrogen (Bertsch et al., 2016). The hydrogen obtained from water and electricity could be used to be stored in a certain medium sized barrels capable of handling between 200 and 500 bar of pressure. During the research, several storage tank capacities will be used to evaluate viability. Based on their CapEx and OpEx, the best suitable tank will be selected for the scenarios and taken into the BM. While selecting the best suitable storage tank, demand of electricity, hydrogen capacity of the route and potential supply of energy to the tank should be computed as well (Clegg and Mancarella, 2016). Production rates, CapEx, OpEx of the electrolyser facilitating storage and release (fuel cell) of hydrogen can be found in the appendix (3.3).

Gray et al. (2011) argues that using hydrogen storage capabilities is a better choice compared to li-ion batteries in terms of higher capacity and less leakage over longer periods of time. Even though it might have disadvantages regarding conver-sion losses of electrolysis and fuel cell(s), hydrogen storage is still a more ecological choice since this method has better safety features and a smaller carbon footprint compared to the li-ion batteries (Gray et al., 2011). Storage also assists in maintain-ing a continuous supply, based on fluctuatmaintain-ing electricity generators. Furthermore it indirectly solves congestion problems faced by TSO’s by storing redundant elec-tricity from the balancing market in the form of hydrogen (Kroniger and Madlener, 2014). Therefore, storage aids in stabilising supply and maintaining a buffer in the system when supply in the multi-commodity system might be unreliable due to ei-ther longer periods of high electricity prices or downtime of electrolysers (Kroniger and Madlener, 2014).

Line packing

Next to internal supply of hydrogen directly from the electrolyser or via a storage tank, line packing could also be used. The volume of H2(g) maintained in a pipeline

(17)

capacity is further elaborated. Line-packing should be further optimised when fo-cusing on longer pipelines due to differences in sizes of pipelines (R´ıos-Mercado and Borraz-S´anchez, 2015).

2.4 BMDFV

(18)

(19)

Table 1: The particular steps obtaining a viable BM (D’Souza et al., 2015)

BMDFV Description Outcome

Step 1 Design brief business idea Explore coupling opportunities of commodities in a flexible MES

Step 2

Four perspectives of viability on which possible BM concepts should create alignment

(Casadesus-Masanell and Ricart, 2011) using different building blocks

Using essential elements from the basic model (figure 3), additional building blocks

could enhance more alignment and viability for the different business perspectives

Step 3

Evaluation of the business model concepts (Section 2.6).

Attention is paid to

evaluating the capabilities of parties to implement the business model(s).

The different perspectives will evaluate the business model concepts

based on certain evaluation criteria important for the MES

Step 4

If not considered viable in step 3, The business model should be critically accessed if the BM needs adjustments or is not considered viable under the circumstances given

Adjustments could be done by performing sensitivity analysis under which circumstances it might be

viable and analyse if these adjustments are achievable in a reasonable

time-frame

Appendix 2, (Table 22) presents the different business model design elements to create design choices for the BM building blocks (D’Souza et al., 2015).

2.4.1 Business model design perspectives

Technonology perspective - The technology perspective in a MES is interpreted as the processing capabilities of the BM concept to viably convert and transport commodities (Eriksson and Gray, 2017). The technology perspective could further be decomposed into the Information Service architecture (IS) and the Physical ar-chitecture layers (D’Souza et al., 2018).

From the IS layer, the BM concepts should be viable of performing continuous information interactions about production schedules, biddings and programmes re-quested by the aggregator(s) in the MES (Mourshed et al., 2015). In Appendix 1.2 a possible IS is provided for the upcoming MES.

The Physical architecture layer is considered the lay-out of the infrastructure needed to perform commodity exchanges in the BM concepts.

The techno-economic feasibility is often used to measure viability (Eriksson and Gray, 2017). This is considered to be the return of investment on the considered hydrogen route and services and should be evaluated for insights about potential investments (Eriksson and Gray, 2017). Evaluation only of techno-economic viability is often only sufficient for mono-commodity energy systems (D’Souza et al., 2018), however might not hold for MES (D’Souza et al., 2018). This since MES BM’s are often more complex and require alignment from different perspectives to be considered viable (Casadesus-Masanell and Ricart, 2011);(D’Souza et al., 2018).

(20)

and is vital for the BM to work (D’Souza et al., 2018). The focal company is often responsible for the content, structure, and governance of transactions within a MES and therefore a crucial actor for receiving value in a BM (Hellstr¨om et al., 2015).

Service/product perspective - Describes the value that is created for the consumer by providing a clear conceptualisation of the service concept (D’Souza et al., 2018). Based on consumer requirements and constraints, they set stage for further designing the BM on what might be necessary elements or potential break-points in the BM concepts (Casadesus-Masanell and Ricart, 2011).

Business ecosystem perspective- Polatidis and Haralambopoulos (2007) ar-gue that social aspects should also be incorporated in a proper renewable energy system. It is imperative for a MES that a focal company involves their stakeholders properly in the business ecosystem for renewables. (Polatidis and Haralambopoulos, 2007). Industrial Symbiosis is acknowledged to reduce the waste of industrial parks and improve value for the actors involved, not only the focal company. (Mortensen and Kørnøv, 2018). This requires the business ecosystem to be aligned with the other three perspectives (service-product, technological perspective and focal actor perspective) to make the BM viable for a MES (See step 2 figure 1). In order to obtain a viable BM from the BMDFV, the design choices should be based on deci-sions that stimulate storage and conversion capabilities, in order to create a flexible BM for a potential MES. This means setting up a viable business model by creat-ing a value for the long-term stakeholders of the MES. D’Souza et al. (2018). This also holds for the proposed hydrogen backbone in the industrial park. Due to the interdependency of all stakeholders in the possible new MES, all parties parties in the business ecosystem (BE) should obtain enough value from the MES to create support for the investment. In appendix 1.1, a possible outlook with value streams is shown (Figure 13).

2.4.2 Design choices

Business design elements- To reach the goal of the Climate Agreement, which has to be viable for all stakeholders to gain momentum for the investment, the BMDFV approach is suitable (D’Souza et al., 2015). If the proposed BMDFV is to be implemented, it is imperative that all stakeholders have to put in effort to make it viable (Van der Burg et al., 2019). This is therefore seen as the general guideline for each BM (D’Souza et al., 2018). Revenue streams have to be fairly distributed, services should be viable, value should be captured and parties have to deliver the amount of flexibility or hydrogen agreed on through multilateral contracts. Further-more it is expected that legislation in the future will require safe operations during the commodity exchange. In Appendix 2, table 20 elaborates on specific aspects of the business design elements.

2.5 Future scaling and evolve-ability

2.5.1 Scaling viability

(21)

to exhausting fossil energy resources. In the future, depleted salt caverns might show potential in storage capacities when a longer route is in place, delivering high system flexibility to the MES. (Michalski et al., 2017). Having relatively low invest-ment costs compared to hydride storage tanks, they might be seen as an alternative to the current hydrogen tanks. The caverns however are often large and require large amounts of hydrogen to effectively store hydrogen, hence high demand and supply are needed to be techno-economical viable (Michalski et al., 2017). Next to the conversion losses through transport over longer distances , safety hazards will also increase (Dodds et al., 2015). Although the new hydrogen route’s capacity is increased with 15 to 20 times, the risk level is also increased (Rusin and Stolecka, 2015). When taking into consideration the pipeline transporting hydrogen to civilian areas, safety valves and compressors should be placed in the pipeline to reduce the risk level to an acceptable level (Dodds and Demoullin, 2013). Furthermore older steel pipelines have to be replaced since the steel is susceptible to embrittlement (Dodds and Demoullin, 2013).

2.5.2 Evolve-ability viability

The proposed MES should be validated for which parts might be coupled to other MES systems like heat and biomass (Shi et al., 2016). Following the ‘merit order’, next to P ower − to − H2 flexibility, the MES might benefit with short term P ower −

to − heat options (Finck et al., 2018) and for the long term conversion to syngas to produce ’green’ plastics (Kats, 2018). From the literature, the following evolving options are elaborated in table 2.

Table 2: Summary of evolve-ability options for a hydrogen-electricity MES

Commodities Description Role MES Source

Heat-to-District

Using the abundance of

heat (steam) from processes in the industrial park to deliver heat via a turbine to civilian areas

Heat for civilian

areas Lygnerud and Werner (2018)

Heat-to-Power

Using excess heat (steam) to produce electricity via a turbine and a generator

Feedstock for

industrial processes Bouw et al. (2015) H2 → CH4

”Methanisation” to

produce ”green’ biogas Replace grey CH4 Agneessens et al. (2018) O2 as feedstock

Side product from electrolysis used as further feedstock

Can be sold to lower hydrogen cost price

Kato et al. (2005); Wijk (2017) O2 + Biomass +

H2 → syngas

Building blocks to create the multi-functional ’syngas’ to be further used in industries

Multi-functional gas to be used for different industrial processes

Kats (2018)

(22)

2.6 Evaluation criteria BM

Based on step three of the BMDFV (Figure 1), Mansilla et al. (2012) and Shi et al. (2016) evaluation criteria from the literature have been summarised in table 3.

Table 3: Summarised BM evaluation criteria from the literature

Evaluation criteria for the BM concepts Challenges BM concepts have to overcome

Viability of the proposed services

Viability of the flex-package based on research Van de Burg et al.,(2019) and Thesis Radstaak (2019), should be capable

of delivering value, performing flexible trades and have a fast response-rate.

Viability for Groningen Seaports (focal company)

BM should allow Groningen Seaports to perform pivotal role (D’Souza et al., 2018) establishing the BE. Challenge remains to facilitate services and trades taking into consideration legal constraints enforced by legislators.

Viability for

the Business Eco System (Stakeholders among the industrial park –

connected to the route)

Challenge to distribute value between

stakeholders (figure 1). All parties should be involved and extract value from the

BM (D’Souza et al., 2015) - Value distribution among the

stakeholders in case of a positive NPV

Technological Viability

Challenges are the technological

capabilities of the industrial park to provide flexibility described in the literature and scenario’s. Furthermore energy markets might lack volatility or volumes (MW)

to effectively use conversion flexibility. (Bouw et al., 2015)

Scaling opportunities

Challenges regarding transforming old infrastructure, salt cavern interactions and complexity of

potential new stakeholders in the BE

Evolve-ability regarding other MES

Challenges in the MES

to ’couple’ more commodities effectively. + techno-economic viability

(Shi et al., 2016)

Overall rating of the viability of the BM

Challenge remains on how

(23)

3 Methodology

At first will the reason why Analytical Quantitative Research in combination with BMDFV was chosen is elaborated. Secondly, the steps for data analysis are men-tioned. This is followed by the case description elaborating on the stakeholders of the upcoming hydrogen backbone. These are mentioned to gain more insight on which parties are involved in the upcoming model. In the fourth section, possible business model concepts using flexibility are elaborated based on capabilities of the industrial cluster and literature. In the final section, validity and limitations of an AQR approach are discussed and methods used to increase validity and robustness of this research are mentioned.

3.1 AQR method

(24)

Figure 2: Research design for BM’s (Bouw et al., 2015)

Wieringa (2014) proposes evaluating the business models using modelling simu-lations of the upcoming MES. As a research approach for AQR, a deductive research is the most suitable for this situation (Bell et al., 2018). Based on theory in the theoretical background, the four concepts are setup which might solve proposed problems faced in the industrial park. The goal of an exploratory AQR study is to provide new critical success factors for a hydrogen MES to eventually after several iterations ‘pass’ step three (Figure 1 of the BMDFV (D’Souza et al., 2015). There-fore this research design shows much similarity with the research design framework from Bouw et al. (2015) seen in figure 2.

Limitations of AQR might be the practicability of the model in reality. This since the proposed model is based on future forecasts of demand and supply of the backbone and predicted technology that might be available in the near future. This research was aware that assumptions and data from the present could be different in the future, and the results obtained were discussed in the discussion section. After a potential positive the Techno-economic viability, the gains should be fairly dis-tributed. Based on semi-structured interviews, experts will be interviewed under which conditions they think stakeholders ought the MES viable. If accumulated requirements of the BE tend to outperform the positive gains from the BM, con-cessions on wealth distribution have to be made. This will enter a new phase of negotiations. It is therefore key for the internal aggregator to link all providers and consumers flexibility and maintaining both sides of the spectrum. Furthermore the aggregator should guarantee availability of hydrogen supply of a certain level, considering capacity constraints of flexibility (from consumers)(Van der Burg et al., 2019).

3.2 Data analysis

Data is gathered by performing quantitative observations on the day-ahead market and balancing market, determining price differences and analyse supply and demand of hydrogen for the involved stakeholders.

(25)

delivering requirements needed for a flexible MES. The upcoming MES was mod-elled using Microsoft VBA. The model should simulate meeting the energy demand from the hydrogen consumers while maximising the total NPV (net present value) of the investments. Microsoft VBA is capable of easily switching between different scenario’s by ticking boxes. Therefore suitable to test different BM building blocks. This while also being able to simulate certain price changes and perform the sensi-tivity analysis. In the Results section, the model logic modified from Bouw et al. (2015) is explained further. Viability in other perspectives should be included in decision variables, parameters and constraints. During the modelling process, the data was analysed in certain steps.

Data analysis steps in this research

(26)

the BMDFV. Due to extra CapEx/OpEx regarding these extra investments, scaling might not be a lucrative endeavour when potential gains tend to be lower. Further-more line-packing will also play a Further-more significant role during scaling (Dodds and Demoullin, 2013). During validation, the BM’s should be evaluated on the ability to evolve as well to see if more commodities can be ‘coupled’ to support each other (Mathiesen et al., 2015).

The proposed BM should be validated for which commodities might also con-tribute for other flexible MES systems like heat for example (Shi et al., 2016). Following the ‘merit order’, next to Power-to-H2 flexibility, the MES might benefit with short term power-to-heat and thermal energy storage for short term flexibility options Krieger et al. (2016). This process can done by using the side product steam (heat) from processes on the industrial park. Using respectively a steam turbine and generator to generate electricity (Finck et al., 2018).

3.3 Case description

As mentioned during the introduction, the case provided in the industrial park is ideal in exhibiting opportunities for flexibility regarding a MES. The large supply of renewables from Norwegian, Danish, Dutch and German wind farms in the fu-ture will contribute to an increase of possible hydrogen production (Wijk, 2017). Furthermore, the Slochteren gas field has contributed to an already existing gas industry in the northern Netherlands, making a switch to hydrogen more attractive (Wijk, 2017). The industrial park has capabilities to facilitate possible linkage with the external market, while being able to produce hydrogen due to the electrolyser available on the industrial park. Initiatives by end-users to use hydrogen as feed-stock has been risen and demand is especially high in Germany, which is in close proximity of industrial cluster (Wijk, 2017).

3.3.1 The hydrogen backbone

The complete 4 KM hydrogen backbone includes two parts. The old route was able to connect chlorine producers and has a hydrogen mobility fuelling station. The old route is seen as conventional, able to resist 2 bar of pressure. The new backbone, which can be considered state of the art is capable of handling 42 bar. Next to the extended capacity, Groningen Seaports is the first in the market to accurately mea-sure the presmea-sure in the pipelines. To compress the presmea-sure from the old pipeline, a compressor is used from a producer. In the pipeline itself, several compressors are included to obtain the preferred pressure for each of the stakeholders. See results section for a schematic overview (figure 8) of the hydrogen backbone.

3.3.2 Stakeholders

(27)

Table 4: Summarized stakeholder analysis based on (D’Souza et al., 2018)

Stakeholders Goals Value Proposition Roles and Responsibilities

Groningen Seaports

Attract new consumers to the route. Facilitate balancing services and maximise flexibility in the MES.

Exploit potential revenue obtained for customers from facilitating the route

Internal aggregator facilitate necessary information services. Accountable for consistent hydrogen transport

in the route.

Industrial Consumers Minimise energy costs

Additional revenue by costs saved from using the MES discount hydrogen prices

Provide flexibility to the route by maintaining flexible production schedules

Industrial Suppliers Maximise profit

Maximising profit by selling hydrogen to internal

market at lowest electricity prices

Optimise production and supply hydrogen when needed

Legislators Fair and safe platform Safe operations

Provide safety to the route by using metering services provided by Groningen Seaports Programme responsible party

Collect flexibility for optimal trade on external market

Flexible control load to respond within seconds

Balance Portfolio E- programmes to TSO External Aggregator: Collect flexibility on an external trading platform. Provide price status to the Internal system. Transmission grid

operator (TSO)

Maintain grid stability (transmission grid)

Help maintain balance – minimise congestion Transmission service System service -grid balance Distribution grid operator (DSO)

Maintain grid stability

(distribution grid) Grid balance Transmission service

3.3.3 Fuel cell investment

In the current industrial cluster, no fuel cell is in place yet. This therefore requires an analysis for an investment. (Appendix 8.1). If seen not viable to sell back to the external electricity markets, hydrogen is discharged from the buffer and given to consumers instead when needed. The simplified information architecture can be found in appendix 1.2.

3.4 Business model concepts

(28)

of either conversion or storage flexibility (Krieger et al., 2016). Based on different combinations of certain possible ’building blocks’, different ways of doing business can be done. Resulting in different business models (Jensen, 2014).

3.4.1 Business model concept 1

At first a basic business model with the essential blocks to perform flexible operations is elaborated (Table 5). This basic model uses conversion flexibility in the day-ahead and intraday market since these markets do not require a PRP. This could be viable since no share of revenue has to be delegated to a PRP (external aggregator) (Bouw et al., 2015). Furthermore, line packing alone is used as storage flexibility since it is considered viable in all business models. Since the base-case scenario already invested in the hydrogen pipeline, line-packing can be considered as relatively free storage flexibility (Dodds and Demoullin, 2013).

Table 5: Basic BM elements using flexibility

Situation: Action Producers

Actions industrial hydrogen Consumers Actions Aggregator High Prices Electricity markets

- Producers and prosumers produce electricity using a fuel cell (if viable) - Electricity is sold on the external electricity markets.

- less hydrogen is produced to deliver to industrial consumers Flexible capacity in schedules allow industrial consumers to either trade hydrogen internally when needed or ramp down operations -Aggregate flexibility obtained from consumers by facilitating trades of capacity

- Send all available flexibility for extra electricity conversion if viable

Low Prices Electricity markets

- Produce green hydrogen using electrolysation

- Electricity is bought on the external electricity markets - More hydrogen is produced to deliver to industrial consumers Flexible capacity in schedules allow industrial consumers to trade abundances internally or ramp up operations Aggregate flexibility obtained from industrial consumers by facilitating trade with abundances of capacity.

(29)

Figure 3: Basic business model concept 1: MES hydrogen-electricity

(30)

Figure 4: BM concept 2: MES hydrogen-electricity + buffer

(31)

(32)

To increase conversion and storage flexibility in the MES, extra trading options on the balancing market can be added and combined with extra storage to minimise curtailment. This results in maximum gains from flexibility. During peak moments congestions could be countered by using the extra storage capability (Clegg and Mancarella, 2016). During periods of low(er) electricity prices, the storage tank can be charged and during high electricity prices discharged to obtain maximum revenue (See figure 6).

(33)

3.5 Validity of the research

For sensitivity analysis, this research will apply three case scenario’s for the chosen business model concepts. From the most likely case, worst and best case scenarios are simulated based on reactions to optimal trade prices and flexibility assumptions from industrial consumers. All the data gathered will follow the ethical standards from Karlsson (2016) and Easterby-Smith et al. (2012) (Appendix 5). This to make sure this paper will be in line with the Data Protection Act (Carey, 2018). To validate parameters chosen in the model, several experts experienced in evaluating the certain criteria in table 3 should validate the proposed BM’s together with the business owner. During the modelling, the constraint of delivering enough ‘green’ hy-drogen should be noticed. This since a substantial part of the hyhy-drogen in the route has to be green eventually to meet the goals set by the Climate table. Furthermore, assumptions of price fluctuations of commodities should be validated regarding fu-ture prices. This requires a certain sensitivity-analysis based on possible fluctuations in the external electricity markets and internal hydrogen market regarding the ca-pability of the MES to respond properly to all possible fluctuations in the external electricity market. Based on AQR and the BM elements, the Research question,

(34)

Table 6: Overview of experts interviewed for brainstorming and validation in this research

Organisation Function Role research

Expert 1 Gasunie Business Developer

Brainstorming BM elements and balancing role 20 MW electrolyser specifications Expert 2 Nouryon

Producer industrial cluster

Renewable Development Manager Brainstorming BM elements and validation Techno-economic outcomes and BE perspective Expert 3 New Energy Coalition Project manager

Validation Evolve-ability and scaling of the BM concepts

Expert 4

University of applied sciences (Hanzehogeschool) Groningen

& Centre of Entrepreneurship

Main author of D’Souza et al. (2018) Expert in flexible MES

Validation model logic Validation overall viability Validation of assumptions, constraints and parameters business case

Expert 5 SI-advice - consultant

Expert

designing business model canvas and pipelines

Line-packing assumptions Focal actor perspective viability

Business owner Groningen Seaports Case company ICT-Expert

Validation focal actor perspective

Validation BE perspective Expert Focus Group Get There Leek Experts

ICT consultants

(35)

4 Results

At first, the model logic is explained. Secondly, data from the internal and external energy markets are presented. From there price models are explained and results are presented on how flexibility might decrease the green hydrogen cost price. Based on the outcomes of historical data, business case elements are discussed. From there, the business models were tested in the internal market for technical viability. After this the internal market was ’coupled’ with the external electricity markets. This model will evaluate the economic viability of the different concepts. Based on different ’building blocks’ and parameters based on previous research from scholars (Bouw et al. (2015),Buttler and Spliethoff (2018)), viability of the investment can be explained in return of investment (ROI) and net present value (NPV). Outcomes of the economic model could create different business cases for the viable business models. At last, the most likely case, best and worst case scenarios were generated for the viable BM’s in a sensitivity analysis.

4.1 Model logic

(36)

Figure 7: Basic Procurement policy based on a price threshold (U) (APX)

Based on certain conversion flexibility constraints of consumers (Appendix 6.2), excess capacity from the producer side is used as a starting point for the amount of hydrogen that can be bought or sold. The model follows a certain ’merit order’ from which source flexibility is used to buy or sell this excess capacity amount.

• 1 – At first, line-packing is used as a source of flexibility (50 KG).

• 2 – Extra storage flexibility is used in BM 2 and 4 using a barrel (1000 KG). Industrial Consumer flexibility for BM 1 and 3 is used by changing produc-tion schedules.

• 3 – Industrial consumer flexibility is used in BM 2 and 4 when the buffer is either full or empty due to longer periods of consecutive buying and selling, the amount requested for a trade is derived from flexibility provided by the hydrogen industrial consumers.

(37)

4.2 Analysis of the energy markets and prices

4.2.1 Internal hydrogen market

Based on expectations from Dool et al. (2016) (confidential) and the case description (Methodology), a schematic overview of the expected internal hydrogen market with average production and consumption patterns of hydrogen (kg/hour) can be seen in figure 8.

Figure 8: Expected average production and consumption pattern industrial cluster in kg/hour (2022).

The triangles in figure 8 represent the source of hydrogen in the industrial cluster. • Grey(steam methane reforming) from the prosumer.

• Blueas a side product from chlorine production (chlor-alkali - switches between grey and green feedstock - sometimes emitting greenhouse gasses (labelled as ’blue’ ).

• Greenfrom the electrolyser (60-200-1000MW).

(38)

4.2.2 Flexible external electricity markets (day-ahead and intraday market)

In the APX day-ahead market (Nordpool for the intraday in the Northern Nether-lands), biddings can be placed by parties without the need of a PRP. Due to its consistency in price patterns, perfect foresight is not needed in the day-ahead mar-ket (Sioshansi et al., 2009). In the upcoming model, the threshold(s) only allows profitable spot price trades and will be optimized with the constraint of limited flexibility available from the industrial hydrogen consumers. The potential revenue of the industrial park could be derived from the following equation (4). In this equation, based on (Kroniger and Madlener, 2014), electricity would be sold back to the external electricity market using a fuel cell if viable :

π = P (day − ahead) ∗ Ep+ P(day − ahead).Es∗ ηrt (4)

Where

ηrt= ηel∗ ηf c

Following the condition: ηrt >

Pdayhead,min

Pdayhead,max

Table 7: Description of the variables - based on Kroniger and Madlener (2014).

Symbol Description

π Potential revenue to be realised when trading on the day ahead-market Pdayahead Price day-ahead - (min = threshold to buy) (max = threshold to sell)

Ep Abundant electricity available not used to convert into hydrogen

Es Energy in storage (Line-packing + tank) released for trade

ηrt % of available electricity after a whole ‘roundtrip’ (Power-to-H2+ H2− to − power)

ηel Conversion rate electrolyser

ηf c Conversion rate fuel cell

Since trading on the Intra-day is similar, the economic model is assumed to work the same way as the day-ahead market, only with different price thresholds and vol-umes (See Appendix 4). The intraday market in the Netherlands is quite underused compared to other European countries (Bouw et al., 2015);(epex-spotmarket, 2019). This research is not taking into consideration this market due to the small amount of electricity available in the Netherlands (Bouw et al., 2015).

4.2.3 Balancing Market

(39)

Figure 9: Retrieved from TenneT (2018); APX (2018)

Table 8: Prices external electricity markets (Euro/MWh) (Nordpoolgroup 2018) Prices based on historical data and forecasts Upward Balancing market Downward Balancing market Day-ahead (2022) Intraday Mean 43,06 12,90 39,26 27,21 Standard deviation 61,20 31,15 8,15 22,34 Median 39,48 0,00 40,32 32,00 ≈Volume available per trade (MW) - - - 80,61

4.2.4 Hydrogen cost price

(40)

for sensitivity of the electricity price. In the near future however, the prices are considered to be closer to 3 or 4 euros per kg (Wijk, 2017). This means higher CapEx and OpEx should be used in the model for the short term determination of the hydrogen cost price.

Figure 10: Mulder et al. (2019); Wijk (2017) linear hydrogen cost (Euro/kg) and the corresponding electricity price (e/MW)

Long term hydrogen cost price:

Phydrogenlong =

Pel

17, 85 + Capex(0.4) + Opex(0.2) + T ransport(0.1)+

Interest − payback(0.3) + Other(0.2) − SaleO2(0.3)∗ = 2, 30 − 3, 50europerKG

(5)

Short term determination of the hydrogen cost price:

Phydrogenshort=

Pel

17, 85+ β1Capex(0.4) + β2Opex(0.2) + T ransport(0.1)+

Interest − payback(0.3) + Other(0.2) − SaleO2(0.3)∗ = 3, 5 − 4, 1europerKG

(41)

Table 9: Explanation hydrogen price model variables based on Wijk (2017)

Symbol Description

Pel Price of electricity obtained from either the day-ahead or balancing market

β1 Higher Capex due to less economies of scale (factor 2-3, Expert 2))

β2 Higher Opex due to less economies of scale and worse conversions (factor 2-3, Expert 3)

Sale O2

After electrolysis, O2 (oxygen) is also formed, This could be sold

and used for gasification, reducing the hydrogen cost price (to be included with evolve-ability validation of the MES ((Kato et al., 2005);(Wijk, 2017))

In order to establish inter linkage between the renewable electricity market and internal hydrogen market, data from table 8 and figure 10 allows the model to trade when prices tend to be lucrative. Resulting in the possible costs saved from using flexibility to decrease the hydrogen cost price (see table (10)). In Appendix 8.2, a table is shown with a sensitivity analysis of the cost price calculation.

Table 10: Potential of using flexibility to decrease the cost price (Wijk, 2017) of green hydrogen.

Cost price (kg) green hydrogen ≈ e4,- before flexibility

Day Ahead (BM 1+2) Day-ahead and

Balancing (BM 3 +4) Units Cost saved from

flexibility per kg ≈ 0,48 ≈ 1,01 e/KG Short term

% saved on cost price

hydrogen 12 25 %

New hydrogen cost price 3,52 2,99 e Long term

% saved on cost price

hydrogen 16 33 %

New hydrogen cost price 3,38 2,70 e

4.3 Business cases

4.3.1 Base case elements

(42)

4.3.2 Fuel cell investment

The ’roundtrip’ mentioned by Kroniger and Madlener (2014) (ηrt), experiences high

energy losses due to conversion losses in the electrolyser and fuel cell, the concept of selling electricity back to the market was seen as unviable (Appendix 8.1). This since CapEx of the fuel cell has to decrease from 2478 e/KW fuel cell (Baghaee et al., 2017) to 1200 e/KW and flexibility from the industrial cluster is required to be 15% = 3,6 hours a day to break-even. Furthermore other variables have to be set to ’best case’ settings, therefore the whole ”round-trip” of hydrogen in all business cases is considered unviable.

4.3.3 business model concept 1

Storage flexibility only from line-packing requires too much flexibility from industrial consumers (7,42%) to be considered technically viable to respond to the day-ahead market. Since processes in the industrial cluster are often continuous, this percentage has to be lower to be considered technically viable. This BM concept is therefore not used to be tested further further in a business case.

(43)

Table 11: Business case business model 2 (Day-ahead trading with extra storage flexibility)

Quantity Units Capex

Software costs (electronic trading

system) 100.000,00 e

Extra storage flexibility 1.000.000,00 e Hardware costs + infrastructure 100.000,00 e

TOTAL 1.200.000,00 e

Opex

Maintenance (4% Capex) 48.000,00 e Marginal costs of shifting schedules 34.340,00 e Personel (APX trades) 14.123,00 e Value for External Aggregator 0 e

TOTAL 96.463,00 e

Revenue

extra revenue from Day-Ahead 274.799,15 e/ yearly

revenue from Intraday 0 e

revenue from Balancing market 0 e

TOTAL 274.799,15 e

Other variables

Number ramp-up and downs 6868 marginal cost for ramping up

and down 5 e/cycle

Quantity shift (4,3% demand industrial consumers per hour) 79 (per hour) kg/hydrogen

Cost of capital 7,43 %

Tax (legislation) 25% %

Depreciation per year 10 %

Deprecation period (years) 10 years Reduction in green hydrogen cost price (kg) 0,48 e

For evaluating the business case in different scenario’s the following parameters have been modified for worst and best case scenarios:

Conversion rate electrolyser In the current business case, the electrolyser has a conversion rate of 65% (base-case). This has been modified for 60% and 70-75% for worst and best case scenario’s respectively (Clegg and Mancarella, 2016). Conversion losses also refer to the ability to react to trades, hence the industrial cluster is not able to respond in time to price changes (0% best case, 10% most likely and 20% worst case).

Costs of shifting consumption In the best case scenario, no marginal cost of switching consumption is incorporated per cycle. A cycle can be seen as as an act of deviating from scheduled consumption schedules and after reacting, switching back to the old schedule (Bouw et al., 2015).

(44)

level of investments needed is considered e100.000,- In the most likely case. If it is assumed that infrastructure is already in place based on the stakeholder capabilities, this investment could be lower.

Table 12: Worst case, most likely case and best case scenarios for business model 2

Variables Worst Case Most Likely Case Best Case Units NPV -282.839,80 28.035,98 547.980,97 e Marginal costs of shifting schedules 5 5 0 e/cycle Investment in hardware/infrastructure 100.000 100.000 0 e Revenue 229.653,57 274.799,15 327.141,84 e/year

Due to the technical constraints mentioned earlier in business model 1, Relying on line-packing alone to trade on the balancing market requires even more flexibility from the industrial hydrogen consumers (12,55%). Expert 2 argues that in the current industrial environment it is not viable to ask for that extra flexibility to trade on the external market. Business model 3 is therefore considered not viable in the scenarios in the near future since too much flexibility is required from industrial consumers to trade on the balancing market to obtain realistic results in a business case. Scaling viability might add more pipelines from which line-packing can be derived, hence could make BM 3 more viable (see discussion section).

4.3.6 business model concept 4:

(45)

Quantity Units Capex

Software costs (electronic trading

system) 150.000,00 e

Storage 1.000.000,00 e

hardware costs + infrastructure 500.000,00 e

TOTAL 1.650.000,00 e

Opex

Maintenance (22% Capex) 363.000,00 e Marginal costs of shifting schedules 212.406,00 e Personel (APX + Balance trades) 50.000,00 e External aggregator profit sharing

(Balancing market) 497.410,79 e

TOTAL 1.122.817,24 e

Revenue

revenue from Day-Ahead

(higher due to more flexibility + simulation outcomes**)

314.080 e/ yearly

revenue from Intraday 0 e

revenue from imbalance trading - balancing market 994.821 e

TOTAL 1.171.157 e

Other variables

Monetary value shared with

external aggregator / PRP 50* % Number ramp-up and downs 42481

marginal cost for ramping up

and down 5 euro/cycle

Threshold (U) - APX 5 %

Threshold (U) - Balancing Market 20 % % Extra conversion flexibility

required from consumers industrial cluster next to extra storage flexibility

7,11 %/hour

Quantity shift 149 (per hour) kg/hydrogen

Cost of capital 7,43 %

Tax (legislation) 25 %

Depreciation per year + economic 10 % Deprecation period (years) 10 years Reduction in green hydrogen cost price (kg) 1,01 e

Table 13: Business case business model 4 (Balancing market with extra storage flexibility

(46)

** Since trades on the balancing market are performed more often compared to the APX market, modelling is only performed based on prices of one month (January, 2018, see Figure 8) due to processing constraints in Microsoft Excel for simulating large datasets. Therefore the data from January 2018 is used as input for a whole year.

Table 14: Worst case, most likely case and best case scenarios for business model 4

Variables Worst Case Most Likely Case Best Case Units NPV -2.359.670,40 -842.866 2.045.115 e Marginal costs of shifting schedules 8 5 0 e/cycle Investment in hardware/infrastructure 650.000 650.000 600.000 e Revenue From APX 274.820 314.080 392.600 e/year Revenue from Balancing market 749.942 857.077 1.071.346 e/year Revenue sharing with external aggregator 50% 50% 40% % Revenue sharing with Business Eco-system 0 0 104.062 e/year

Exploring a viable stand-alone ﬂexible multi-commodity energy system in the North of the Netherlands

flexible multi-commodity energy

system in the North of the

Netherlands

Pieter Minnee

A thesis presented for the degrees of

MSc. Technology & Operations Management

and

MSc. Operations & Supply Chain Management

commodity energy system (MES) in the North of

the Netherlands

Pieter Minnee

Abstract

Preface

Contents

List of Figures

List of Tables

1

Introduction

2

Literature review

2.1

The energy market

2.2

Conversion flexibility

2.3

Storage flexibility

2.4

BMDFV

2.5

Future scaling and evolve-ability

2.6

Evaluation criteria BM

3

Methodology

3.1

AQR method

3.2

Data analysis

3.3

Case description

3.4

Business model concepts

3.5

Validity of the research

4

Results

4.1

Model logic

4.2

Analysis of the energy markets and prices

4.3

Business cases