Energy flexibility within industry: an industry perspective on flexibility creation

(1)

1

Energy flexibility within industry: an industry perspective on flexibility creation

Thomas Jonkman (s2976684)

A master thesis submitted to the Faculty of Economics and Business of the University of Groningen in order to obtain the degree of Master of Science in Technology and Operations Management

June 25th₂₀₁₈

Supervisors: Prof. dr. ir. J.C. Wortmann Ing. R.J.H. van der Burg, MSc

(2)

2

Abstract

Current energy systems are evolving due to an increasing amount of renewable energy technologies being integrated into the market. Although renewable energy technologies result in clear environmental benefits since less greenhouse gasses are emitted, the uncontrollable nature of renewable energy technologies causes an increasing discrepancy between energy supply and demand. Energy systems need to become more flexible in order to align energy consumption and generation patterns. Fortunately, the industry sector offers an unexploited potential of energy flexibility. However, flexibility within industry has hardly gained any attention from a business perspective, making it difficult to understand how flexibility is being created and integrated into industrial processes and what constraints exist that limit the creation of flexibility. Therefore, this research has studied the creation process of flexibility within industry and the associated constraints by means of an empirical case study methodology. Findings show that the creation process of flexibility within industry can be generalised into three phases. Firstly, industrial processes are primed for flexibility creation through the coupling of market information with internal processes and potential flexible assets are selected. Secondly, the assets are deployed and influenced by energy market information, product characteristics, product demand, and storage availability. Thirdly, the flexibility supplying party is being remunerated for the amount of flexibility (Mega Watts per hour(MWh)) that has been supplied. Findings also indicate that the creation process of flexibility is being limited by market regulations, psychological factors, technical factors, and commercial factors. By having studied flexibility within industry from a business perspective, this study adds novel insights into how flexibility is being integrated within industrial processes, the type of processes that can be activated in order to create flexibility, the market structure of industrial flexibility suppliers, and the necessary agreements that have to be in place in order to initialise flexibility creation.

Key words: Energy flexibility, Flexibility within industry, Flexibility, Demand-side management,

(3)

3

Acknowledgements

This thesis resembles the completion of my master program Technology and Operations Management which sets an official hallmark to the end of my time as a student, although I hope to never stop studying. It has been a pleasure researching the challenges that revolve around renewable technologies, a subject that always gave me the feeling of actually contributing to a pressing societal problem. However, studying this subject in such a context, would not be possible without the help of important people who have supported me during my master’s.

First of all I would like to extend my profound gratitude to my two supervisors, prof. dr. ir. J.C. Wortmann and ing. R.J.H. Van der Burg, MSc for their useful guidance and critical feedback throughout the process of writing my master thesis.

Moreover, I would like to thank Fabian Salzmann for our fruitful discussions, especially during the formulation phase of this research.

Furthermore, I could not have performed this research without the help of the people I interviewed. I would therefore like to extend my gratitude to those who assisted me in gaining knowledge about how flexibility is incorporated into their companies.

Finally, I would like to thank my family for always having supported me throughout my studies and my girlfriend, Lisa, for her support and useful feedback and lessons on writing an academic paper.

(4)

4

List of abbreviations

aFRR Automated frequency restoration reserve

APX Amsterdam Power Exchange

BRP Balance responsibility party

DSM Demand-side management

DSO Distribution system operator

FCR Frequency containment reserve

GTS Gas transport services

ICT Information communication technology

IFS Industrial flexibility supplier

mFRR Manual frequency restoration reserve

MW/h Mega Watt per hour

RES Renewable energy source

RR Replacement reserve

SSM Supply-side management

TBS Technical building systems

(7)

7

1. Introduction

Current power systems are shifting towards decarbonisation, clean energy generation, and more efficient energy consumption mechanisms (Villar et al., 2017). In terms of investments in renewable energy sources (RES), such as wind energy and solar power, the Netherlands experienced the fastest growth of all European countries as investments grew from EUR 1.54 billion in 2013 to EUR 5.30 billion in 2014 (REN21, 2015). Although the increase in RES brings clear benefits to the energy system as less greenhouse gasses are being emitted, the often stochastic and uncontrollable nature of RES causes an increasing variance in energy supply and demand (Hurtado et al., 2017). To cope with this serious challenge, the energy system has to become flexible, where flexibility of an energy system can be defined as the possibility of modifying generation and consumption patterns leading to an increase in stability of a power system while maintaining adequate system performance (Alizadeh et al., 2016; Eid et al., 2016).

When more RES are integrated into the energy market, the market experiences a discrepancy between energy supply and demand (Biegel et al., 2014). Therefore, grid frequency and voltage control of the energy system need to be managed effectively in order to secure a match between supply and demand of energy. Fortunately, the integration of RES can be supported by different sources of energy flexibility, namely demand-side management (DSM), supply-side management (SSM), energy storage, and energy conversion (Lund et al., 2015). When energy flexibility is being created, a new role within the existing energy market emerges: the flexibility supplier who exploits the sources of flexibility. In the energy network, the industry sector offers a substantial, unexploited potential of flexibility (Weeber et al., 2017). Also in terms of energy consumption in the Netherlands, the industry sector is the biggest consumer compared to households, the energy sector, transport, and agriculture (Provoost et al., 2014). Therefore, when flexibility is being created within industry by e.g. shifting energy consumption patterns of internal processes according to the energy market, the discrepancies between energy supply and demand can be minimised, showing the potential for industrial flexibility suppliers (IFS).

(8)

8 flexibility from a technical perspective (Weeber et al., 2017; Schultz, Sellmaier and Reinhart, 2015). Also, research about constraining factors that limit flexibility deployment have primarily been researched from a technical perspective, concerning the impact specific machine states have on flexibility creation (Simon et al., 2017). In order to fully deploy the potential of flexibility within industry effectively, it is essential to gain insights in the factors that constrain flexibility creation and limit the potential of flexibility sources within industry as a whole.

Therefore, this study aims to study the business side of flexibility creation within industry, focusing on the phases of energy flexibility creation within industry and the constraining factors that hinder this flexibility creation. While investigating the creation of flexibility, an operations management (OM) perspective shall be taken. This perspective suits this research since OM research aims to study the processes that are devoted to production and delivery of products and services, which are the processes that are key contributors to creating flexibility (Slack, Brandon-Jones and Johnston, 2017; Langer et al., 2014). Consequently, this research contributes to OM literature. Furthermore, by researching how the sources of flexibility are created within industrial firms, this study aims to contribute to energy flexibility literature. Moreover, the factors that constrain the flexibility creation process within industrial firms shall be investigated. In order to gain insights into the mentioned gaps and to contribute to both OM literature and energy flexibility literature, this study aims to answer the following research questions:

- How is energy flexibility being created within industry from a business perspective? - What are the constraints that limit the energy flexibility creation process?

These research questions are investigated empirically by means of a multiple case study research that aims to study the creation of energy flexibility in-depth. However, firstly, a literature review about the energy market, energy flexibility, and energy flexibility within industry was performed. By using an empirical approach, this research aims to contribute to practice by displaying how energy flexibility can be integrated into operational processes and therefore supporting the implementation of energy flexibility within industry as a whole.

(9)

9

2. Literature review

In order to answer the research questions posed in chapter 1, an in-depth investigation of the related topics of the current Dutch energy market and actors is required. Firstly, the energy market will be explained. Secondly, the notion of flexibility is described. Thirdly, flexibility will be analysed from an industry perspective, revealing the remaining gaps and added value of this research.

2.1. The changing energy network

The modern energy system can be defined as a commodity based system where technologies, institutions, and consumers are intertwined and aim to produce, transfer, and consume energy (Verzijlbergh et al., 2014; Kirschen, 2003). Different actors involved in modern energy systems are the transmission system operator (TSO, responsible for stability of the transmission system), distribution system operators (DSO, responsible for the power distribution (Hansen et al., 2013)), wholesale or retail suppliers of energy (called balance responsibility parties (BRP)), wholesale entities (responsible for energy generation), regulators, and consumers (Villar et al., 2018). Figure 2.1 visualises the traditional energy system incorporated with these market actors.

Figure 2.1: The current energy market

Due to a decline in costs and advances in technologies of decentralised electricity generation such as photovoltaics and wind energy generation, a new actor, called the prosumer, has entered the energy market (Parag and Sovacool, 2016). This prosumer actively participates in the energy system by both acquiring energy from and generating and selling energy to the market. Moreover, the awareness of climate issues are a cause for consumers to pursue energy neutrality (i.e. consuming as much energy as generating energy). Steps towards decarbonisation have been taken by the instalment of heat pumps and services promoting the use of electric vehicles (Biegel et al., 2014). Although this shift achieves obvious environmental benefits, a redesign of the energy market is necessary (Villar et al., 2018). Traditional energy systems are mainly based on the production of energy from centralised fossil fuel power plants. Because the generators from these plants are directly linked to the grid, the energy production is flexible and supports frequency changes in the energy system (Kundur, 1994). RES

Energy producers

and BRP

DSO Consumers

TSO

(10)

10 however are unable to perform in the same flexible way as centralised power plants do and have a highly fluctuating character: they promptly increase or decrease in production depending on the weather (Biegel et al., 2014). Therefore, as RES are increasingly introduced the market and creating more variability and uncertainty, new approaches need to be defined to balance the energy grid and to harmonise energy supply and demand.

Flexibility services, such as energy storage measures, have proven to support the evolving energy system by creating ways to balance the power grid (Eid et al., 2016). Flexibility services can be defined as ‘services that deploy aggregated flexibility sources in such a way to fulfil a specific need of a party in the electricity system (Van der Burg et al., 2018). Hence, these flexibility services can be seen as a result of flexibility sources aggregated from individual flexibility suppliers and thereafter delivered to energy market participants who are responsible for balancing the energy market. This assimilation process is performed by an aggregator, that acts as an intermediary between the flexibility supplier and parties such as the BRP, TSO (Navid and Rosenwald, 2013), and DSO (Zhang and Kezunovic, 2016). Resulting from this line of reasoning, Figure 2.2 depicts the changing energy system (adopted from van der Burg et al., 2018).

Figure 2.2: The energy market incorporated with flexibility

(11)

11 Due to fluctuations in the market, a new market, called the imbalance market, arises where market actors can trade energy every 15 minutes in order to create a balance between energy supply and demand. So-called frequency reserve is being traded by market actors who can spare capacity and is characterised by primary reserve (Frequency Containment Reserve (FCR)), secondary reserve (Frequency Restoration Reserve (FRR)), and tertiary reserve (Replacement Reserve (RR)) according to level of urgency. Specific information about these types of reserves can be found in Appendix I.

2.2. Flexibility within the energy system

When one refers to flexibility being offered to the energy market, operational flexibility is commonly used as it refers to deploying and managing flexibility devices during operations. As defined by Ulbig and Andersson (2015), operational flexibility is “[…] the technical ability of a power system unit to modulate electrical power infeed and/or power outfeed from the grid over time.” Flexibility can be distinguished based on four different sources: DSM, SSM, energy storage, and energy conversion (Lund et al., 2015).

(12)

12 the available flexibility are often neglected, making it hard to grasp how DSM can be integrated into industry.

The second flexibility source, SSM, refers to measures and technologies that modify the output of power generating units to create balance in the grid when e.g. a high amount of net load comes from RES. An interesting, much discussed SSM technique is that of curtailment (Lund et al., 2013; Conti, Greco and Raiti, 2009). Curtailment in the sense of energy flexibility means limiting the power output of an energy generating unit and can be used when there is an oversupply of energy. An example of power curtailment is running a wind turbine on half capacity instead of fully turning it down. SSM is also described as infeed flexibility as it refers to altering electricity that is being fed into the grid.

The third flexibility source, energy storage, is used to balance temporary mismatches between supply and demand of electricity. Energy storage technologies include compressed air storage and batteries (Lund et al., 2015). An already established storage technology that is being used to provide flexibility services is a 100 MWh battery placed in Australia by Tesla Motors, which is able to provide flexibility to potentially 30.000 households (McGuirk, 2018). One of the main benefits of energy storage compared to solar- and wind power, is that energy storage can provide baseload power production. However, energy storage often requires high capital costs, making it difficult for IFS to integrate energy storage as a source of flexibility within industrial processes (Denholm et al., 2010). Energy storage is known to alter electrical infeed and outfeed at the same time.

The fourth flexibility source, energy conversion, is used when there is an oversupply of renewable energy. Renewable energy can, for example, be converted into hydrogen, compressed air or chemical energy (batteries) (Lund et al., 2015). Although quite similar to storage as a source of flexibility, from a business perspective it differs in the sense that conversion is limited to either impacting the energy infeed or energy outfeed of the system, where storage can alter both at the same time (Van der Burg et al., 2018).

(13)

13 Lastly, the location of flexibility on the grid is an attribute since flexibility demand might be location dependent.

As a result of flexibility creation, markets exist where aggregators can aggregate flexibility from different suppliers. The result of such a market perspective enables flexibility to be traded as measurable goods. Van der Burg et al. (2018) therefore propose flexibility as discrete flex-packages that are being traded on the energy market. The following framework (Figure 2.3) depicts such a flex-package.

Figure 2.3: Energy flex-packages

2.3. Flexibility within industries

The increasing dynamics and complexity within markets due to globalisation, increasing network speed and transfer of information, and swift emergence of new technologies, have required operations and manufacturing functions to be more flexible (Wiendahl et al., 2007). Flexibility in this context is defined as the ability to adjust a certain system of processes in order to cope with a changing market environment without penalising itself in costs, performance, and time of effort (Sethi and Sethi, 1990; Grassl, Vikdahl, Reinhart, 2013). Energy flexibility covers part of this definition as it is related to coping with the dynamic energy supply and demand within the energy market. The effective application of energy flexibility has been investigated in various industrial settings, showing the increasing importance of energy flexibility within industry. The following examples of energy flexibility within industry illustrate this increasing emphasis.

Keller and Reinhart (2016), Langer et al. (2014), and Keller, Braunreuther and Reinhart (2016) have proposed the integration of energy demand and supply parameters within Enterprise Resource Planning (ERP) systems that support production planning processes. In common production processes, primary equipment (i.e. resources with direct linkage to the production process) are often responsible for 66% to 87% of the factory’s power need. Therefore, Keller et al. (2016) proposed a production planning process that is co-determined by present energy values of both on-site energy generation

Flexibility supplier DSM, SSM, energy storage and conversion

(14)

14 sources (e.g. photovoltaics) and the energy market in order to create energy flexibility (Keller et al., 2016). Although the study performed by Keller et al. (2016) show the successful adoption of energy market data in production control systems, no empirical evidence has been obtained with regards to the creation process of flexibility, making it difficult to understand how flexibility can be fully integrated into operational processes.

As a second example, Simon et al. (2017) have proposed a systematic approach into different machine states within a manufacturing environment and their possibility to adapt to fluctuating energy levels. Simon et al. (2017) present strategies to enhance load adaption of machines, minimise technical constraints, and respond to interactions between work-stations providing flexibility contributing to a flexible energy system. Simon et al. (2017) propose the evaluation of flexibility potential within industry by looking at specific machine states. However, no further constraints have been researched that influence flexibility creation as a whole, making it difficult to grasp what other, non-technical factors hinder flexibility creation.

As a third example, Weeber et al. (2017) have proposed a more factory-wide view on the adoption of flexibility measures and propose that technical building systems (TBS), auxiliary processes and other non-value adding processes within a factory are most supportive of supplying flexibility within industrial environments. Additionally, Kuhlmann and Bauernhansl (2015) propose a methodology used during the early stages of factory planning for creating an energy-agile production system that reduces production capacity during peak hours.

As a fourth example, Schultz et al. (2015) have made significant efforts in investigating the synchronisation of demand and supply of energy by using energy-oriented production control by means of a technical simulation study. The authors found that when electric energy is treated as production capacity measure, deviations in energy consumption can potentially be reduced by adoption of an oriented order release protocol. However, the integration of such an energy-oriented control measure has not yet been researched empirically, making it difficult to understand what steps are necessary in order to integrate flexibility within operational processes.

(15)

15

Table 2.1: Overview of contemporary literature about energy flexibility within industry

K el le r an d R ei n h ar t, 2 0 1 6 La n ge r et al ., 2 0 14 K el le r et al. , 2 01 6 Si mo n et al. , 2 01 7 W ee b er et al. , 2 0 17 K u h lman n an d B au er n h an sl , 2 0 15 Sc h u lt z e t al ., 2 01 5

Planning and scheduling x x x x x x

Inventory control x x x x

Inclusion of energy parameters in production system

x x x x

On-site energy generation x

(preliminary) costs benefits shown x

Process improvement perspective x x

Internal (In) vs external view (Ex)* In In In In In Ex Ex

Factory planning and design x x

*In: includes only production processes, Ex: includes other business processes and environmental factors

As can be observed from Table 2.1, IFS have mainly been studied from a technical, internal approach, focusing mainly on the integration of planning and scheduling parameters in production processes. No empirical research has been performed focusing on the creation process of flexibility, making it difficult to understand how flexibility is being integrated within industrial processes and what the constraints are that limit the potential flexibility.

2.4. Aim of this study

(16)

(17)

17

3. Methodology

In current literature, the deployment of flexibility within industry has mainly been investigated from a technical, quantitative approach (Table 2.1), which has led to a lack of knowledge about how flexibility is specifically being created and integrated within industrial processes. To fully understand how flexibility is being created within industry, an empirical study through multiple case study research was performed. Case study research is most suitable for this research because case studies aim to gain insights into unfamiliar situations in order to develop or enhance theories (Meredith, 1998). Furthermore, by applying case study research, specific and underlying, context-related information can be gathered from practical cases (Delattre et al., 2009; Sandelowski, Docherty and Emden, 1997). The reason why this qualitative approach was preferred above quantitative research, was because of the exploratory nature of this research. While quantitative research aims to gain insights in the relationship between variables in a population (Hopkins, 2008), qualitative studies use a limited amount of cases that are explored in a highly detailed manner, which was most suitable for this research (Voss, Tsikriktsis and Frohlich, 2002).

3.1. Research setting

Multiple organisations were chosen that were investigated in-depth. These organisations all had different roles in the context of flexibility, i.e. exploiting different sources of flexibility, differ in position in the flexibility energy market (Figure 2.2), or exploit the same flexibility source in a different way. However, the organisations have the common characteristic that they are all directly involved in creating flexibility within industry. All organisations that were involved in this case study research were already deploying flexibility sources within their processes or were in the final phase before starting to create flexibility. Further stakeholders of this case study research are academia interested in energy flexibility literature.

3.2. Data collection

(18)

18

Table 3.1: Researched cases

Case number Type of party Role of interviewee Flexibility source Case A Multinational chemicals and paint

provider (IFS 1)

Innovation technologist

DSM

Case B Multinational chemicals provider (IFS 2)

Technology

innovation manager

DSM

Case C Cooling house (IFS 3) and aggregator (aggregator 1)

Quality advisor and account manager

DSM

Case D Gas and hydrogen storage facilitator (IFS 4)

Technical manager Energy storage, conversion

Case E Aggregator (aggregator 2) Business analyst DSM, SSM

The main focus of this study was to acquire data from IFS, but in order to acquire more comprehensive, diverse data concerning IFS, two aggregators of industrial flexibility were included in this case study, resulting in four IFS and two aggregators. Interviews were conducted with employees of the organisations that were directly involved in the creation process of flexibility. In order to conduct the interviews most effectively and to acquire all relevant data related to the gaps in the literature about energy flexibility within industry, a pre-determined interview protocol was formulated that consisted of the following parts. The full interview protocol is described in Appendix III.

1. The first steps necessary to activate flexibility creation and the devices that are activated for flexibility creation;

2. The impact flexibility creation has on overall production processes and other business processes;

3. What departments need to work collectively in order to create flexibility within industry; 4. The specific flexibility sources that are exploited and what constraints exist that limit the

exploitation of these flexibility sources;

5. What and how market actors are involved in the flexibility creation process.

(19)

19 that the right information was gathered. Further actions that were taken to ensure validity and reliability are provided in Appendix IV and were based on the work of Yin (1994, p. 33).

3.3. Data analysis

First, data from the semi-structured interviews was recorded and transcribed. Subsequently, a chain of evidence was visualised according to a coding scheme. This was developed in order to understand what has been said and how it translates into energy flexibility creation within industry. For this process, inductive coding was used which represents the creation of new theory. By using coding as a means to analyse data, a deep reflection and interpretation of the data can be won (Miles, Huberman and Saldana, 2014). Furthermore, the use of codes in this qualitative analysis enabled the conversion of primary data to be turned into information that was used for interpretation. The process of analysing the data that was collected during the interviews was performed by using ATLAS Ti, which enables extraction of relevant data related to the research topic. After this, a chain of evidence was visible that flowed from the primary data, into the coding scheme, into valuable interpretations (Appendix V). Furthermore, applying inductive coding for different case studies enabled a cross-case approach which contributed to more generalised conclusions. The following structure of coding was performed.

1. First cycle coding was performed in which segments of the raw, transcribed data were summarised. In this step, the initial data was reduced into fragments, sentences, or paragraphs that had the potential to support or answer the research questions.

2. Second cycle coding was carried out. In this stage, the summaries, or first-order codes were grouped into themes related to energy flexibility, and the exchange of energy flexibility between the IFS and the aggregator. Examples of possible second-order codes are financial constraints that hinder the flexibility creation or contractual forms that are necessary for flexibility activation.

3. Third cycle coding was performed from which second-order codes were grouped together in pools that were linked to the main concepts. During this stage, some first-order codes were duplicated when they belonged to different themes (or third-order codes).

4. A within-case analysis was explored and thereafter explained. Within this step, results related to the research questions were described according to relevant passages from appendix III, the coding scheme.

(20)

20

4. Findings

This chapter will give a detailed description of the creation of flexibility for IFS. As mentioned in the previous chapter, first, a within-case analysis was performed. This means that, for case A until E, the creation of flexibility is described. When describing the cases, a visualisation of how the flexible device is activated is provided. Furthermore, all constraining factors are visualised in a table that shows the impact of a factor on the flexibility attributes. Secondly, a cross-case analysis was performed in which differences per case are explained. Third and lastly, this chapter concludes with a summary that conceptualises the creation of flexibility across multiple industries.

4.1. Case A – Industrial Flexibility Supplier

Case A represents a multinational chemicals producer (€14.75 billion turnover) which supplies DSM (Figure 4.1). IFS 1 produces chlorine for industrial usage through electrolysis and this primary process is primed for flexible use by switching down the 200 MW electrolysis machine by a maximum of 25%. Outfeed flexibility is provided after IFS 1 determines an energy consumption strategy according to product demand, storage capacity, and the energy prices of the imbalance market without interference of an

aggregator. IFS 1 is able to provide FRR within seconds and around 10 times per day. However, ensuring the delivery of chlorine to downstream processes is the leading determinant in providing flexibility and the concerning amount of flexibility.

4.1.1. Pre-deployment phase

The process of creating flexibility in case A follows a few consecutive phases. In the first phase, a dedicated energy department is in place that constantly monitors the imbalance market and internal energy consumption levels. When a deviation in energy price is measured at the imbalance market, the IFS determines an operational position: “[…] you follow the signal that the energy market (TSO) makes public, and you can see whether there is a shortage or abundance of energy. Then you can choose to act [IFS 1]”. During this phase, everything is performed automatically and the process of electrolysis follows the market signal. No interference of a party is involved in this so-called manual FRR (mFRR) flexibility creation, as being said by IFS 1: “We just follow a signal from the APX and we act upon this signal, whether we are actively (automatically) or passively (manually) producing flexibility. Not many parties have to be involved in this [IFS 1].” In the future however, IFS 1 is aiming to supply

(21)

21 automated FRR (aFRR), where a pre-determined amount of flexibility is contractually agreed upon between the TSO and IFS 1.

4.1.2. Flexibility deployment phase

In phase 2, an actual call for flexibility is received and the actual device is being switched down to a capacity level which is determined in the pre-deployment phase. There are a few factors influencing the flexibility deployment, the biggest being commercial factors as mentioned in case A: “Our customers are most important when you look at chlorine. We want to keep serving our customers and [...] when our customers want chlorine and they are willing to pay for it, we will produce, even if the energy prices are very high[IFS 1].” Other factors influencing this flexibility deployment are as follows: - Technical factors: “There are technical challenges […] because gas gets produced at the electrolysis process. When the gas cannot be released because of the pipes are too narrow, you can blow up [IFS 1].”

- Product factors: “You only have a limited chlorine storage capacity. The society is not very keen on large chlorine storage containers [IFS 1].”

- Process quality: “All processes follow the electrolysis, so in this sense you can create [flexibility] relatively easy [IFS 1].”

In the deployment phase, a continuous cycle of monitoring the imbalance market and adjusting the flexible asset occurs according to the developed energy strategy.

Table 4.1: Factors and constraints of case A Factor Constraint Influences

Commercial Customer demand - Energy provision capability (MWh), available time: decreased when high demand

Technical Available equipment - Ramp rate capability (MW/h): limits due to quick gas release

Product Unable to store product

- Energy provision capability (MWh), available time: limits due to storage cap

4.1.3. After-deployment phase

(22)

22 power in MWh that has been delivered and the pre-determined amount of flexibility that the TSO and the IFS 1 have agreed on contractually.

4.2. Case B – Industrial flexibility supplier

Case B also includes a multinational chemicals producer IFS 2 (€58,2 billion revenue) who supplies DSM to the same type of market as IFS 1. IFS 2 also supplies chlorine for industrial usage through electrolysis (Figure 4.2). This primary process that consumes approximately 500 MW of electricity is used for flexibility creation by turning down the machine by a minimum of 40 and maximum of 80 MW according to market data. IFS 2 constantly bases its energy consumption strategy on product demand, storage

capacity, and the energy market prices from which the amount of flexible power is derived. IFS 2 provides outfeed flexibility in the form of mFRR for time periods of a few minutes directly to the APXs imbalance market without interference of an aggregating party.

4.2.1. Pre-deployment phase

IFS 2 follows a few consecutive phases when flexibility is being created. In the first phase, a dedicated energy department is in control of a process control system that automatically calculates the economic consideration between producing chlorine and the amount of chlorine that is lost when electricity consumption is being reduced: “A process control system is connected to an enterprise business system which calculates the optimal production rate. This signal is being sent to the factory which translates into a reduction or increase in capacity [IFS 2].” The process control system is coupled to the enterprise resource system that sends a signal directly to the electrolysis machines which are in turn lowered in energy consumption. Only mFRR is being supplied as, due to high quality standards, IFS 2 wants to be in full control of their chlorine factory. Hence, no pre-determined contract that includes a flexibility band width is set between IFS 2 and the TSO.

4.2.2. Flexibility deployment phase

In the deployment phase, the flexible device is being set to a level that is determined by the process control algorithm. This algorithm constantly monitors the energy market and momentarily deploys the flexible assets. The deployment of flexibility is determined by a few factors, among which the most prominent factor is a commercial factor: “You need a certain buffer when providing flexibility. […] you

(23)

23 cannot keep doing the same thing, otherwise your downstream factories will run out of raw materials [IFS 2].” Other factors influencing the flexibility creation are for example:

- Product factors: “A chlorine factory is not a candy factory. We want to be in control at all times [IFS 2].”

- Technical factors: “We have a big plug to arrange this demand-side procedure but some companies are not willing to invest in this [IFS 2].”

- Market factors: “The amount of hours of imbalance where the electricity price deviates from the average energy price is limited, around 150 hours higher and 150 hours lower on a total of 8700 hours. This makes it difficult to invest in assets that can provide flexibility [IFS 2].” And: “The physical, international connection could dampen the (flexibility) problem [IFS 2].”

Table 4.2: Factors and constraints of case B Factors Constraints Influences

Commercial Customer demand

- Energy provision capability (MWh), available time: decreased when high demand

Technical Network connection*

- Power provision capability (MW): limited when a smaller plug is available

- Electrification of assets: necessary when smaller plug is available but can be costly

Market Low flexibility demand

- Electrification of assets: less need for investing in electrification of processes when low flexibility demand

International connections*

- Local flexibility generation (MWh): decreased when transfer between nations is more mature

Product Product safety

- Energy provision capability (MWh), available time: decreased due to high quality standards

*Not an issue for IFS 2 but known as an overall issue of flexibility creation

4.2.3. After-deployment phase

(24)

24 IFS 2 is being remunerated by the TSO for the amount of MWh of flexibility that has been supplied to the market.

4.3. Case C – Industrial flexibility supplier & aggregator

Case C consists of IFS 3 (€293.3 million

revenue), a firm that provides heat and cooling treatment services for flowers and is considering the implementation of a software system that enables the provision of DSM (Figure 4.3). This is provided by aggregator 1, which is also included in case C. Aggregator 1 provides software systems that enable firms to

analyse their production processes and energy consumption levels. Through this system, IFS 3 is able to provide a flexible band width of energy that aggregator 1 controls and deploys. This flexible band width is based on an energy consumption strategy which is influenced by energy market prices and product characteristics (see 4.3.2.). IFS 3 will be able to connect 125 large industrial cooling assets to this software system which will enable them to decrease 600 MW/h of electricity at times of an energy price peak. Hence, DSM is being assimilated by the aggregator and IFS 3 is able to save 10% of the energy costs of their primary process.

4.3.1. Pre-deployment phase

During this first phase, the aggregating party lays contact with the IFS and makes sure the software the aggregating party offers is compatible with the systems of the IFS. Thereafter, the software is installed at the IFS and data about production volumes, internal energy consumption levels, and energy market data is collected. Furthermore, the margins in which the company chooses to be flexible are determined by the customer (flexibility band width): “Every day the customer can enter a position […] in the different energy markets. With another module we can then adjust the imbalance and make the right choices. […] it is all performed automatically [Aggregator 1].” Moreover, in this stage a contract is also set between the aggregator and the IFS. In this contract, variable fees are set that are being paid per MWh of electricity that is being traded via the portal and a fixed fee for using the software system.

4.3.2. Flexibility deployment phase

In the deployment phase, the flexible assets are being deployed and demand-side flexibility is being created according to the systems of the aggregator (who collects energy market data). The main factor

(25)

25 influencing this flexibility deployment is product characteristics. As being said by IFS 3: “A flower bulb is not […] a piece of dead meat in which case you are much more flexible with your cooling system [IFS 3].” This factor influences the energy provision capability (MWh) and the available time of flexibility as the total amount of flexible power and time are limited due to the product being restricted to certain temperatures. No other factors, such as market or regulatory factors, play a role, as mentioned by the IFS 3: “It is not at all exciting. It is just a matter of switching on and off [IFS 3].”

4.3.3. After-deployment phase

The flexibility supply goes from the IFS 3, to aggregator 1, to the Dutch TSO. Remuneration takes place between the IFS 3 and aggregator 1 for the use of the portal and between aggregator 1 and the Dutch TSO for guaranteeing flexibility in an imbalanced market. Here, aggregator 1 is supplying mainly secondary reserve to the market and is in a pilot phase of supplying tertiary reserve (emergency power). IFS 3 benefits from deploying flexibility by internal energy cost savings on its primary cooling process. No penalties are incurred by aggregator 1 to IFS 3 when IFS 3 is unable to supply flexibility: “Customers only pay the price of an imbalanced market; we do not incur specific penalties [Aggregator 1].”

4.4. Case D – Industrial flexibility supplier

Case D includes IFS 4, a fast-cycle gas and

energy storage company that provides flexibility as its core activity (Figure 4.4). IFS 4 measures real-time energy market data provided by the GTS1_{and whenever}

a customer or the GTS wants to (periodically) store gas, IFS 4 injects their storage facilities with a maximum of 13.2

GWh/h and withdraws from their facilities with a maximum of 18 GWh/h. For example, when a customer experiences a gas abundance in the market, it can store its gas against a storage fee and withdraw its gas again when there is a shortage, hence balancing the gas market. In addition to already providing storage services for gas, IFS 4 will provide conversion services in the future through the creation of hydrogen from wind and solar energy, which is now in the pilot phase.

1_{Gas Transport Services, serves the same role as the TSO but operates on the gas market}

(26)

26

4.4.1. Pre-deployment phase

In the first phase, gas market data is being analysed and interpreted by the IFS 4. When a customer of IFS 4 wants to store gas, contracts are first being set between IFS 4 and the customer in question: “Contracts vary from short to long term, from 1 to 10 years. They include agreements concerning how much gas is entering, how fast it is entering and how fast the gas is flowing out of the storage [IFS 4].” All systems that are in place automatically arrange the whole process of deploying the flexible assets, including the injection and withdrawal of gas to and from the storage facilities.

4.4.2. Flexibility deployment phase

In phase 2, flexibility is being created according to the GTS its gas price rates. The pressure of the gas network is being measured and the IFS 4 acts upon these fluctuations in time periods of hours. Whilst it is technically possible to inject or send out the maximum amount of gas within 15 minutes, contracts have to be in place 2 hours before sending to or injecting gas from the grid since this is required by market regulations. The main factor influencing the flexibility deployment is related to the gas market itself: “Our biggest challenge is that there are a lot of gas storage facilities on the market. There is also a lot of gas available so the margins on storage are at a minimum. It is a difficult time for us [IFS 4].” Other factors that influence flexibility deployment are the following:

- Regulatory factors: “Contracts with market actors have to be in place 2 hours before injecting or sending gas so that the TSO can verify your action [IFS 4].”

- Technical factors: “[the market] is less fragmented. […] there is a limited amount of entry points where you can acquire gas [IFS 4].” And: “We are limited to injecting 1.1 million of cubic meter of gas […] because of our compressing machines [IFS 4].”

- Psychological factors: “Our license to operate to the community is highly important as gas is not the most popular product, especially not in the area of Groningen [IFS 4].”

- Trust factors: “It is highly important to guarantee delivery to our customers […] if we do not deliver gas, it costs us around 1% of our yearly revenue [IFS 4].”

4.4.3. After-deployment phase

(27)

27

Table 4.3: Factors and constraints of case D Factors Constraints Influence

Market Abundance of gas storage

- Energy provision capability (MWh), available time: lower since low demand for storage

Regulatory Contract requirements

- Swiftness of flexibility deployment: limited due to 2 hours contract time

Technical Network connection - Ramp rate capability (MW/h): limited due to capacity of compressors

Injection and withdrawal capacity

- Ramp rate capability (MW/h), Energy provision capability (kWh): limited due to capacity of compressors

Psychological Trust in gas storage - Energy provision capability (MWh), available time, acquisition of assets: limited as gas is unpopular in region

Trust Reliability of operations

- Energy provision capability (MWh), available time: limited to machine requirements in order to mitigate risks of machine failure

4.5. Case E – Aggregator of an industrial flexibility supplier

Case E includes aggregator 2, a

company that provides software systems to industrial firms enabling DSM as a source of flexibility, similar to case C. This case also includes a specific project concerning SSM creation and the delivery of primary reserve (FCR) to the

TSO. In this situation, energy market data is interpreted by aggregator 2, an energy provision strategy is mathematically determined and the energy provision load of wind turbines in the North Sea is adjusted. Here, FCR is provided. The energy provision capability is adjusted with a maximum of 4% of a group of wind turbines capacity when an abundance of energy on the grid is measured. Figure 4.5 visualises the FCR provision. When DSM is created at the IFS, this results in the same process as case C (Figure 4.3).

(28)

28

4.5.1. Pre-deployment phase

In this first stage of DSM, the aggregator visits the flexibility supplier and links their operations system to the systems of the aggregator. Data that is used for flexibility deployment concerns historical energy market data, production volumes, weather data, and process characteristics data. These types of data are then used to calculate and forecast an optimal point of energy usage to reduce consumption costs. In this phase, the customer decides the maximum flexibility band width that can be used for flexibility creation. When providing the system that enables DSM, a contract is set between the IFS and aggregator 2 that includes a fixed fee for the use of the system or a variable fee that includes a percentage of the costs savings the IFS makes. In the case of supply-side flexibility, daily contracts are set that determine the amount of primary reserve aggregator 2 has to adjust for the wind turbines. As said by the aggregator: “[for daily contracts] you are much more flexible because when there is a certain amount of wind blowing, you can be quite certain that it holds for the whole day. For weekly contracts over a large group of wind turbines, this is really hard [Aggregator 2].”

4.5.2. Flexibility deployment phase

Furthermore, concerning DSM creation, psychological barriers mainly play a role: “You have to convince people and give them examples that [flexibility] works. If you work with people who understand the issue and can convince other people, your band width increases [Aggregator 2].” Concerning adjustments of wind turbines according to energy demand, a continuous cycle occurs of measuring the deviation of the balanced energy grid in Hertz. When a deviation is measured, the system is able to decrease its energy provision within 15 seconds by 4% of the turbine’s capacity. The main factor that influences flexibility deployment for SSM, is centred around regulatory factors: “The TSO simply cannot simply apply different rules than other European TSOs […] as, due to harmonising different energy networks in countries, decisions are primarily being made on a European level [Aggregator 2].”

Table 4.4: Factors and constraints of case E Factors Constraints Influence

Regulation European FCR rules

- Energy provision capability (MWh), available time: cannot provide flexibility yet due to European regulation

Psychological Trust in software

(29)

29

4.5.3. After-deployment phase

When DSM is delivered to the market, the IFS party gets remunerated in the form of decreased energy costs. The IFS is not remunerated for providing flexibility to the market, since this action is undertaken by aggregator 2. The delivery of flexibility is transferred from the IFS to aggregator 2 to the TSO. The IFS pays the aggregator according to the contract that was set in the pre-deployment phase. In the case of SSM creation, aggregator 2 gets remunerated according to the amount of FCR that was supplied to the TSO.

4.6. Cross-case analysis

This section provides a cross-case analysis in which important topics related to the research questions will be compared between cases. Table 4.5 provides an overview of results of the within-case analysis, presenting a starting point for the cross-case analysis. The cross-case analysis will highlight the four biggest differences: the involvement of an aggregating party, response time, factors influencing flexibility deployment, and contractual differences related to relational governance.

Table 4.5: Cross-case analysis

Case A Case B Case C Case D Case E Contractual agreement No No Yes Yes Yes

Aggregator involved No No Yes No Yes

Level of automation High High High High High

Main flexibility source DSM DSM DSM Storage SSM

Response time* High High High Low High

Power provision capability 50MW 60MW 0.5MW 6TW 0.32MW

Constraining dimension Internal Internal Internal External External

Type of process primed for energy flexibility

Primary Primary Primary Primary Primary

*high = 0 sec – 30 sec, moderate = 30 sec – 15 minutes, low = 30 minutes or higher

(30)

30 certain departments are appointed for forming energy consumption strategies. These strategies are determined according to product demand, product characteristics, storage availability, and energy prices of the APX market. As mentioned by IFS 1: “We have so-called energy dispatchers, who form an electricity consumption strategy according to the day ahead market [IFS 1].” For cases C and E, the energy consumption and offered flexibility is low compared to cases A, B, and D which makes investments in these energy dispatchers difficult, hence offering opportunities for aggregators to provide flexibility assimilating services.

Secondly, when comparing different response times across the cases, one can notice that case D has a relatively high response time (approximately 2 hours) of deploying flexibility compared to almost instantaneous response times of the other cases. However, it is technically possible for case D to quickly react to the market as their processes are electrically driven. Due to regulatory constraints in the gas market, case D can only respond to market deviances in 2 hours, as said by IFS 4: “Everything has to be contracted in the gas market so it takes 2 hours to respond. […] it is incredibly slow compared to the electricity market [IFS 4].” Regulatory constraints therefore limit flexibility creation significantly compared to cases A, B, C, and E.

Thirdly, when comparing all cases, the flexibility creation is impacted by different factors. However, the main difference is centred around the difference in source of flexibility and the factors that in turn influence the flexibility deployment. For cases A, B, and C, who supply DSM, the noticeable factor influencing flexibility creation is related to internal reasons. In other words, the cases that supply DSM are mostly focusing on product quality or the certitude of supplying downstream processes with sufficient product in order to meet customer demand. On the other hand, flexibility deployment for cases D and E are mainly influenced by external factors, such as market structure and regulation. All these factors mainly influence the energy provision capability (MWh) and the available time. Where the energy provision capability (MWh) concerns the amount of flexible energy being deployed for flexible use and the available time depicts the time an asset is supplying flexibility. The main reason this difference occurs is due to the physical product involvement for DSM creation in cases A, B, and C.

(31)

31 flexibility which is readily available (automated balancing, or aFRR). When parties are automatically providing flexibility, contracts revolve around the amount of flexibility that is being delivered (MWh) and the price at which it can be delivered (€ per MWh). When an aggregator is involved, contracts that enable flexibility creation do exist and mainly revolve around specific agreements between the IFS and the aggregator concerning costs for usage of a software system of the aggregator, a fixed fee for the realised energy savings, or a percentage of the realised cost savings (variable fee). When an aggregator is involved, however, the IFS is still able to provide mFRR and gets remunerated by the aggregator.

4.7. Summary

To summarise, the following conceptualisation of the flexibility creation process can be made. This process is depicted in Figure 4.6.

Firstly, in the pre-deployment phase, the general IFS ensures that systems and/or appointed people are in place that acquire market data in order to form a sound energy consumption strategy. This strategy is determined according to product demand, storage availability, and energy prices of the APX market and defines the flexibility band width. This band width illustrates the amount of power in MWh that can be deployed for flexibility creation without harming surrounding processes or products. During this phase, the flexible assets are appointed. These devices that are generally being selected for flexibility deployment are the devices that consume the most electricity. Since monetary incentives are a key factor in the decision whether to supply flexibility or not, the devices for which the IFS can save most of its electricity costs are primed for flexibility creation. Optionally, contracts are set with aggregating parties, involving software usage fees and realised energy reduction fees. Contracts can also be set with the TSO, involving, in the case of automated balancing (aFRR), a reward for the pre-determined amount of flexibility that is made available and the delivered amount of flexibility. A license is in place between the TSO and the IFS or between the TSO and the aggregator to ensure the IFS is qualified as a flexibility supplier

(32)

32 Lastly, the amount of flexibility that is being delivered is measured and the flexibility supplying parties are remunerated for their flexibility supply through invoicing. Parties who do not use an aggregator have to pay penalties if the agreed on amount of flexibility is not being delivered. Parties that use the trading system of an aggregator have to pay fees according to contracts that have been set in the first phase and are remunerated for the delivered mFRR.

(33)

33

5. Discussion

The focus of this study is to increase knowledge as to how flexibility is created within industry and what constraining factors are present that hinder this flexibility creation. This chapter will dive deeper into the theoretical contributions of this study, the practical implications, the research limitations, and advices on further research concerning flexibility within industry. This chapter will conclude with proposed policy implications that could contribute to the enhancement of the integration of flexibility within industry.

5.1. Theoretical contributions

(34)

34 Secondly, this study provides additional insights in the market structure of IFS. Literature shows, according to Figure 2.2 (adopted from Van der Burg et al., 2018), that all IFS supply their flexibility to an aggregating party. However, this study has shown that for large IFS who operate with a dedicated energy department and a high (50 MWh) flexibility power provision, no aggregator has to be involved. Here, flexibility at the IFS is internally being created according to an energy consumption strategy and creates economic benefits for the IFS through cost savings on energy consumption and by the TSO according to the amount of flexibility (MWh) that has been supplied. This flexibility deployment results in a balanced market, which differs from creating flexibility by the IFS as an input for flexibility services that are being aggregated by an aggregator. Furthermore, the remuneration cycle as proposed in Figure 2.2 suggests that only the IFS is compensated for delivery of the flexibility as input for flexibility services by the aggregator. This study has shown that also the IFS compensates the aggregating party, if it is involved, for using energy software provided by the aggregator. In this sense, the IFS rewards the aggregator while the IFS both benefits from the delivered flexibility and the energy cost reductions it realises. Moreover, case E has shown that energy producers are also a source of flexibility provision and hence need to be adopted as an IFS. In case E, aggregator 2 controls the flexibility provision of the energy producer who owns the wind turbine. Following from the three adjustments, Figure 5.1 visualises the actors that are involved in the flexibility creation process of the IFS.

Figure 5.1: IFS market structure

(35)

35 primed for flexibility creation since creating flexibility within industry is mainly a decision concerning cost reductions. Since the marginal costs of electricity for primary factory processes have proven to be significantly higher than that of TBS, these primary processes are primed for flexibility creation. In order to cope with future flexibility demand, which is expected to be rising, factory planning and design should incorporate means to activate flexibility creation. Further research should therefore focus on how flexibility sources should be incorporated into TBS and in factory planning and design in order to realise maximum flexibility potential of TBS. When focusing on TBS as flexibility providers, aggregators should be involved that aggregate large amounts of flexibility from TBS in order to deliver flexibility services to the market (Burger et al., 2017). To summarise, TBS do have the potential of providing flexibility. However, it does not happen as of now since the economic benefits for the IFS are marginal and the creation process concerning flexibility provision from TBS is not yet explored in-depth.

Lastly, when comparing the IFS flexibility creation process (Figure 4.6) with the flexibility service management process developed by Van der Burg et al. (2018), one can clearly see distinct similarities. The main similarity is in the process of first making sure prerequisites, such as software installation and contractual agreements, have to be in place before being able to deploy flexible assets. Another similarity is that, when flexible assets are deployed, so-called actual available flexibility and actual flexibility supply are constantly being monitored. Furthermore, after the flexible assets have been deployed, remuneration takes place between various market actors. The IFS flexibility creation process adds novel insights to the model of Van der Burg et al. (2018) by specifying the flexibility creation process specifically to an industrial setting. Also, this study has found that IFS are requiring software systems of the aggregator to activate flexibility supply for which the IFS pays the aggregator a fixed fee or a variable fee which includes a percentage of the realised energy cost savings. A further addition to the model of Van der Burg et al. (2018) found in this study is that whenever an IFS agrees to supply tertiary reserve (emergency power) to the aggregator and is unable to deliver, penalties are being paid by the IFS to the aggregator. Lastly, this study has found that for IFS, the available flexibility is influenced by product demand, product characteristics, storage availability, and energy prices of the APX market, which adds insights to the first phase of the flexibility service management process of Van der Burg et al. (2018), where potential available flexibility is determined.

5.2. Practical implications

(36)

36 study shows a clear indication of a balance between two points. Namely, the relative impact an IFS can make in providing balancing measures versus the relative cost reductions a firm can realise when creating flexibility. Also, it was shown in all cases that cost reductions were realised and IFS are remunerated for supplying flexibility to the market, showing that this study can help potential IFS in supporting their decision whether to start creating flexibility or not. In other words, this study supports interested firms by showing what steps can be taken to integrate flexibility sources within their processes.

Secondly, this study shows the importance of contractual agreements between the IFS and market actors in supplying flexibility. It is found that for large IFS (cases A and B) supplying secondary reserve (mFRR), no contracts were in place between either the IFS and an aggregator or between the IFS and the TSO. Here, the IFS is being remunerated for the amount of flexibility is has provided (per MWh) and the energy cost savings it has realised. The reason no contracted flexibility is provided for cases A and B (in the form of primary, automated FRR or tertiary reserve), is because both cases operate with highly dangerous chemicals and want to be in control of their own factory. However, due to automation efforts, case A will potentially supply contracted secondary reserve after first testing its capabilities with non-contracted secondary reserve. Both cases do require a specific license that is granted by the TSO which states that the IFS is qualified for supplying mFRR. When an aggregating party is involved, contractual agreements revolve around the amount of flexible power that is agreed upon (i.e. flexible band width in MWh) and certain costs for usage of software that visualises potential flexibility provision. In this case, remuneration takes place from the IFS to the aggregator for providing the system, internally at the IFS on the basis of energy cost savings, and from the aggregator to the IFS for the amount of flexibility that has been supplied. Therefore, this study shows that for large IFS, providing flexibility to the market is not directly bounded to usage of contracts. However, for smaller parties providing flexibility through aggregators, the creation of flexibility is restricted to usage of contracts.

5.3. Study limitations and further research

(37)

37 lead to meticulously constructed advice for policymakers on what policy measures maximise the creation of flexibility within industry best.

Secondly, this study has shown from a qualitative approach how flexibility is created within industry. Knowing how this flexibility deployment process takes place for the four flexibility sources, further research should focus on a more quantitative analysis in comparing the four different flexibility sources and measure the different impacts that the flexibility sources have on balancing the market versus the relative ease of implementing the specific flexibility source. With this knowledge, one can more easily assess what type of flexibility source can contribute most to balancing the energy market.

5.4. Proposed policy measures and implications

In this study, various regulatory and market factors have been identified as possible constraints to the creation of flexibility. Therefore, it is vital to propose countermeasures that could alleviate these constraining factors.

(38)

38

6. Conclusion

Nowadays, the growing depletion of the earth its natural resources for energy provision requires everybody, from nations to municipalities, to develop and deploy new innovations that will foster a sustainable world for future generations. One of the most prominent innovations that has been integrated into the market related to energy provision are RES such as the collection of solar energy through solar panels and wind energy through wind turbines. Although such renewable energy sources result in clear environmental benefits through decarbonising the energy market, integration of RES in the energy market creates challenges due to the unpredictability of RES. In response to these challenges, this study investigated the creation of flexibility within industry, a clear solution to the balancing challenges the energy market currently faces. It was found that the creation of flexibility encompasses three subsequent phases (Figure 4.6). Firstly, in the pre-deployment phase, an energy consumption strategy is determined that derives a flexibility band width. Devices that can exploit the sources of flexibility are selected according to potential energy cost savings and contracts and licenses that enable flexibility creation are optionally being set between market actors and the IFS. Secondly, a demand call for flexibility is being received and devices are deployed according to the energy consumption strategy. The flexible device is constantly being monitored and redeployed according to this strategy. Thirdly, when flexibility in the form of primary, secondary or tertiary reserve is being delivered directly to the TSO or via an aggregator, the IFS is being remunerated according to the amount of flexible power that has been supplied (in MWh), has been made available, and the IFS saves on energy costs.

(39)

39

References

Albadi M.H. & El-Saadany, E.F. (2008) “A summary of demand response in electricity markets”, Electric Power Systems Research, Vol. 78, pp. 1989–96.

Alizadeh, M.I., Moghaddam, M., Amjady, N., Siano, P. & Sheikh-El-Eslami, M.K. (2016) “Flexibility in future power systems with high renewable penetration: a review”, Renewable and Sustainable Energy Reviews, Vol. 57, pp. 1186–1193.

Apxgroup.com. (2018). Dashboard | EPEX SPOT | Welcome. [online] Available at: https://www.apxgroup.com/market-results/apx-power-nl/dashboard/ [Accessed 25 May 2018].

Biegel, B., Hansen, L.H., Stoustrup, J., Andersen, P., & Harbo, S. (2014) “Value of flexible consumption in the electricity markets”, Energy, Vol. 66, pp. 354–362.

Burger, S., Chaves-Ávila, J., Batlle, C. & Pérez-Arriaga, I. (2017) “A review of the value of aggregators in electricity system” Renewable and Sustainable Energy Reviews, Vol. 77, pp.395-405.

Chang, M. & den Breeje, S. (2016) [online] Movares.nl. Available at: https://movares.nl/wp-content/uploads/2016/06/Movares-2016.-Introductie-Industriele-Demand-Response.pdf [Accessed 20 Jun. 2018].

Conti, S., Greco, A. & Raiti, S. (2009) “Local control of photovoltaic distributed generation for voltage regulation in LV distribution networks and simulation tools”, European Transactions on Electrical Power, Vol. 19(6), pp.798-813.

Delattre, M., Ocler, R., Moulette, P., & Rymeyko, K. (2009) “Singularity of qualitative research: from collecting information to producing results”, Tamara Journal for Critical Organization Inquiry, Vol. 7 (3/4), pp. 33-50.

Energy flexibility within industry: an industry perspective on flexibility creation