A package deal for the future: Vehicle-to-Grid combined with Mobility as a Service

(1)

TVE-STS; 19004

Examensarbete 15 hp Juni 2019

A package deal for the future:

Vehicle-to-Grid combined with Mobility as a Service

Amanda Bränström

Jonna Söderberg

(2)

Teknisk- naturvetenskaplig fakultet UTH-enheten

Besöksadress:

Ångströmlaboratoriet Lägerhyddsvägen 1 Hus 4, Plan 0

Postadress:

Box 536 751 21 Uppsala

Telefon:

018 – 471 30 03

Telefax:

018 – 471 30 00

Hemsida:

http://www.teknat.uu.se/student

Abstract

A package deal for the future: Vehicle-to-Grid combined with Mobility as a Service

Amanda Bränström, Jonna Söderberg

The aim of this report is to evaluate how a future commercially owned fleet of self-driving electric vehicles (EV:s) would be able to provide power in order to avoid power exceedances in the power grid. Exceedances occur when network agreements between grid operating companies are exceeded. Exceedances are problematic, since they infer penalty fees for the paying company and make dimensioning the grid capacity more difficult for the supplying company. Capacity deficiency regarding the infrastructure of the grid is expected to increase, likely resulting in higher penalty fees.

Integrating transport and power systems by using self-driving EV:s as Mobility as a Service combined with Vehicle-to-Grid (V2G) technology is a potential solution for this problem. By modeling the EV-fleet as the New York City taxi fleet, a usage pattern deemed to resemble Mobility as a Service is created. An economic value for the V2G service is estimated by comparing the availability of the EV-fleet with local exceedances from Uppsala as well as regional occurring exceedances. The highest income during the first quarter of 2019 is 96 000 SEK for the whole fleet, or 1100 SEK per EV and hour-long exceedance. The time of exceedance and the power magnitude have to interplay with the availability of the EV-fleet in order to enable the system. The EV battery capacity highly impacts the system, but is concluded to not be a limiting factor due to market logic. Lastly, key features such as market formation as well as geographical and technical aspects are presented and discussed.

ISSN: 1650-8319, TVE-STS; 19004 Examinator: Joakim Widén

Ämnesgranskare: Umar Hanif Ramadhani Handledare: Oskar Fängström

(3)

1

Acknowledgements

This report has been put together with guidance from Sweco Uppsala. The authors wishes to express their sincere gratitude to supervisors Oskar Fängström and Anna Lundgren, as well as the Energy group at Sweco Uppsala. A lot of helpful assistance has also been provided by Håkan Österlund at Upplands Energi and Inga-Lill Åkerström at Svenska kraftnät.

Thank you!

(6)

4 Syllabus

Capacity deficiency Challenges in the power grid in two different ways: The power lines are “full” and can not transport any higher amount of power at the given time, or there is not enough electricity produced at a given time. The former challenge will be of interest in this report

EV Vehicle, or car, running on electricity supplied by an internal battery

(Power) Exceedance Overstep of network agreement regarding power consumption, resulting in additional penalty fees

Mobility as a Service A shared transport service which enable people to access transport on an as-needed basis

Network Agreement Agreement between power grid operating companies regulating power and energy consumption as well as the associated fee

SvK Svenska kraftnät, the Swedish transmission grid operating company

V2G Vehicle-to-Grid, technology for creating a bidirectional communication and power flow between the EV and the power grid, enabling an EV to charge from and discharge to the power grid

Peak shaving Leveling out peaks in electricity consumption

Power peak When power output from the power system is high in relation to the consumption of the rest of the day Power shaving Reducing power output from the power grid, not

necessarily related to a power peak

(7)

5

1. Introduction

You are on your way to work. In your left hand, you have a nice cup of coffee. With your right hand, you are texting a colleague. Outside the car window, you see the city pass by and people riding around in cars, just like yourself. Once outside your

workplace, the self-driving car stops smoothly by the curb and an automated voice wishes you a nice day. You grab your bag and get out the car. Once the door shuts behind you, the car drives off to one of the power hubs nearby and docks with it, ready to provide power to the grid through its internal battery. Without anyone noticing, the omnipresent self-driving cars will make sure you and your fellow citizens’ lives are balanced and powered.

This scenario is futuristic, but not impossible. Our society today is completely dependent on electricity, and this does not seem to change in the foreseeable future (Energimyndigheten, 2016). The Swedish power grid is faced with handling power consumption and production that it was not dimensioned for, resulting in capacity deficiency in several areas in Sweden. Challenges regarding the grid capacity is also expected to grow. For the society, it limits how much cities can grow and develop, both economically and geographically. In order to avoid overloads, the usage of the grid is regulated in network agreements between the transmission grid operating company Svenska kraftnät and underlying grid operators, as well as between underlying grid operators, where maximum power output and input is determined (Svenska kraftnät, 2018). The challenges presented pose important questions on how the energy system will look in the future, and give reasons to think beyond the solutions available today.

The electric vehicles (EV:s), mentioned above, could be one of these solutions.

The system of combining bidirectionally charging EV:s with services to the power grid is called Vehicle-to-Grid (V2G). By assuming these EV:s to be self-driving, the system studied in this report can be considered flexible and a model for future implementation of new ways of transport – Mobility as a Service. In order to evaluate how the power capacity available in a fleet of EV:s can be combined with daily power needs, the EV:s power capacity can be juxtaposed with power peaks and valleys. The service provided by the EV-fleet will be power shaving, in the form of power compensation to avoid power peaks. In this report, power peaks exceeding the agreed power consumption, with penalty fees as a consequence, will be examined. By doing this, an economic value of the potential grid services provided by the EV:s will be presented, both per EV and for the whole fleet. An evaluation of how viable the interplay between the EV:s and the exceedances is will also be made, as well as evaluations of both the impact of battery capacity and key features when implementing the system.

(8)

6

1.1 Project aim

The aim of this project is to evaluate how a future commercially owned fleet of self- driving electric vehicles (EV:s) would be able to provide power in order to avoid power exceedances in the power grid.

1.2 Research questions

In order to achieve the project aim, the report will answer the following questions:

1) What is the potential economic value of a fleet of EV:s providing service to the grid?

2) What is the interplay between the EV-fleet and the grid, regarding availability for grid services?

3) If the battery capacity of the EV is changed, what is the impact on the system?

4) Based on today’s discussion and results from previous research questions, which key features are important to address when implementing the system¹?

1.3 Delimitations and limitations

Delimitations

▪ Regulations linked to the presented energy system are not discussed.

▪ How a future market for the presented energy system would look like will only be discussed briefly.

▪ Prerequisites for the implementation of the system, such as charging stations, smart grids and communication software etc., are regarded as fulfilled and therefore not discussed. The EV:s are also assumed to be charged in a way not adding to already existing power peaks. The charging pattern is therefore not discussed.

Limitations

▪ Since the data is sourced, there is no way to ensure the correctness of the data.

▪ Due to legal constraints, only some data regarding electricity usage is available for analysis.

▪ The pricing structure of the future fees follows today’s pricing.

1.4 Overview of the report

An overview of the electricity grid infrastructure and actors, as well as challenges connected to capacity deficiency and role of the EV-fleet are presented in the

1 The system consisting of an EV-fleet and the power grid.

(9)

7

Background chapter. In the chapter Theory, data and methodology, more technical specifications and theory regarding the system will be presented. In short, the method behind the results consists of calculating the number of EV:s involved in power shaving services and comparing this to exceedance fees. Two cases are examined, with

exceedances originating from different levels of the electricity grid. A sensitivity analysis is accomplished by changing the battery capacity of the EV:s and evaluating the system again. Finally, results will be given in the Result chapter, followed by a Discussion chapter where results will be discussed. The report ends with a Conclusion, answering the research questions.

2. Background

This section gives an introduction to the power system structure and network

agreements in Sweden as well as information about the growing concern of capacity deficiency. Future outlooks as well as a solution involving V2G and Mobility as a Service will also be presented.

2.1 Managing capacity deficiency today

Capacity deficiency is a term often used in the Swedish debate of power supply.

(Energimarknadsinspektionen, 2018). A large number of grid operators form and maintain the structure of the power supply, and the linkages defining power flow in the system between these actors are network agreements. How do the contracts relate to capacity deficiency? In this chapter, an overview of the grid structure and its operators as well as capacity deficiency and network agreements will be presented, with focus on the region of Uppsala.

2.1.1 Electricity grid operators

Svenska kraftnät (SvK) is a government authority who owns the transmission grid. SvK also has the responsibility of operating the system, i.e. to make sure that the

transmission capacity and reliability is sufficient (Svenska kraftnät, 2017a). Customers of SvK are almost solely grid operating companies who own the regional grids. There are around 170 grid owning companies in Sweden today, with E.ON Elnät Sverige, Vattenfall Eldistribution and Fortum Distribution being the largest. The regional grid operating company has a monopoly, but also responsibility, to provide electricity to its geographical region. Regional grid operating companies contract power consumption to local grid operating companies (Södra Hallands Kraft, n.d.). The main featured regional grid operator discussed in this report is Vattenfall Eldistribution and the main featured local grid operator is Upplands Energi.

(10)

8 2.1.2 Capacity deficiency

Capacity deficiency means that the power lines lack the capacity needed to deliver the desired amount of electricity to the user, i.e. transmission capacity deficiency. One way to express it is that the lines are “full” and can not transport any higher amount at the given time. Another cause of capacity deficiency is that there is not enough electricity produced at the given time (Energimarknadsinspektionen, 2018). The former type of capacity deficiency will be of most interest when evaluating the system presented in this report. Today, urban regions like Uppsala, Västerås, Stockholm, and Malmö are

affected by transmission capacity deficiency (Kellner, 2019). It affects the cities in terms of having to deny establishment of new factories and server halls, as well as new neighbourhoods (Energimarknadsinspektionen, 2018). Today in Uppsala, capacity deficiency occurs during approximately 200 hours a year. The region of Uppsala has a capacity need of around 300 MW, and there is a limit of how much power the grid operator Vattenfall can provide (Lindblom, 2018).

Persson (2018), the Chief Financial Officer at Energimarknadsinspektionen, stresses that the capacity deficiency is a smaller problem at a national level, and that Sweden in total has a grid capacity that is sufficient. Instead, problems derived from capacity deficiency mostly occur on local and regional levels (Energimarknadsinspektionen, 2018). One reason could be that the capacity of the transmission grid limits the allowed power consumption stated in network agreements with regional grid operating

companies. Upgrading the infrastructure takes time and big economic investments.

Persson means that the way of making a greater capacity available is a combination of smart technologies, demand flexibility and a well-functioning cooperation between grid operators and municipalities (Energimarknadsinspektionen, 2018).

2.1.3 Network agreements

Network agreements exist between both transmission and regional grid operating companies, as well as between regional and local grid operating companies. The structure of the agreements differ depending on which actors are involved (Svenska kraftnät, 2019). On transmission level, the agreements serve as a tool for SvK to plan which capacity is needed for delivering power to a geographical area, as well as a way for SvK to cover costs of maintenance and operations of the grid (Svenska kraftnät, 2018a).

The exceeding of a network agreement might affect the grid in negative ways. Today, exceedances at some power grid stations might get so high above planned capacity that the grid components become overloaded. With an increasing capacity deficiency in the transmission grid, especially in connection with bigger, expanding cities, this problem has potential to grow. SvK sees a trend indicating that the amount of exceedings will increase before actions to develop and upgrade the grid will be taken. In fact, there is an acute need of decreasing the exceedances due to capacity deficiency. According to SvK, one way to counteract the increasing amount of exceedances and to make an even

(11)

9

clearer stand that exceeding a network agreement is not acceptable, is to raise the penalty fee. The intentions are to make the grid operators take actions to avoid the higher costs, and to make the fee reflect the seriousness of the capacity deficiency. SvK points out that it has nothing to do with an economical gain for SvK itself (Svenska kraftnät, 2018a). From January 1st 2019, SvK has changed the structure of the penalty fee (Svenska kraftnät, 2018b). SvK mentions in a referral from 2018 that there is no guarantee that the structure of the fee will look and function the same way in the future, but it is necessary to address the capacity deficiency by making exceedances more expensive (Svenska kraftnät, 2018a).

One consequence of raising the fee of exceeding network agreement is that the cost for the regional grid operator will increase. In turn, end-consumers might be affected too if the regional grid operator decides to raise the cost of the contract with the local grid operator in order to compensate the greater fee charged by SvK (Svenska kraftnät, 2018a).

Just as regional grid operating companies pay fees for using electricity from the

transmission grid, local grid operating companies pay fees for using electricity from the regional grid (Österlund, 2019a). One example is the network agreement Upplands Energi has with the overlaying regional grid operated by Vattenfall Eldistribution. At the time of April 2019, Upplands Energi has already exceeded the agreement with Vattenfall Eldistribution multiple times. For the company, finding solutions for

regulating power consumption is economically motivated. Therefore, Upplands Energi in cooperation with the software company Ngenic AI, has started to control whether heat pumps in houses are on or off in order to adjust the power consumption during cold winter hours. This method has contributed to making it possible for Upplands Energi to shave power peaks by 2 MW (Österlund, 2019a).

2.2 Today’s outlook for the future

In order to investigate the role of a V2G EV-fleet in the power system, a future outlook is conducted. What will the energy system look like and which challenges will be faced? In this part of the report, predictions from SvK about the energy system and plans for the future of Uppsala will be presented.

2.2.1 System services

Today, SvK stresses that the power system is facing major changes. New production methods of electricity and the way electricity is used and stored are some of these changes. This opens up for new ways of managing the grid through new types of system services. According to SvK, it is not obvious what these system services will look like.

With the technical development taking place, it is difficult to say whether new system services will be performed by production facilities or by network components.

According to SvK, it is unclear whether the system services will be implemented with

(12)

10

the help of, for example, regulations or market solutions. Another question is how to divide the responsibility and costs regarding the system services between SvK, electricity producers and grid operating companies. The system services may be provided by commercial operators on market terms (Svenska kraftnät, 2017a).

2.2.2 The future power grid

SvK has formulated a scenario of the Swedish power grid in the year 2040, based on current national and international politics, driving forces and decisions made today. In the scenario, no revolutionary technology breakthroughs, big market changes or big extension of the national power grid is assumed (Svenska kraftnät, 2017a). The most central challenges in this scenario are stated in the Table 1 below.

Table 1. Some challenging aspects of SvK:s scenario of a possible outcome for the Swedish power grid year 2040. The aspects are selected by relevance to this report.

Scenario Outcome

Decommissioning of nuclear power. Decreasing the power and frequency stability in the grid.

Increasing share of intermittent electricity production in terms of wind

power and, to some extent, solar power.

Increasing demand of flexibility and balancing in the power grid.

Increasing power consumption and reducing production capacity.

Degrading of the power supply capacity in the south of Sweden, with possible power

deficiency as a result.

Increasing production and consumption flexibility, as well as

energy storage in the system.

Improving the power adequacy.

(Svenska kraftnät, 2017a)

In the scenario for the power grid 2040, wind power will more or less replace the loss of nuclear power. In total, the energy production is large enough to cover the nuclear power loss, but the weather dependence makes the production unpredictable. Without the nuclear power, the south of Sweden risks a power deficiency of 400 hours a year.

The power deficiency will demand an electricity market with flexibility (Svenska kraftnät, 2017a).

2.2.3 Batteries in the power system

The Royal Swedish Academy of Engineering Sciences (IVA) stresses that a

combination of different energy sources can be used both on transmission grid level to improve the quality of the electricity, and in the distribution grids to improve the local stability of power supply. By using batteries, power is obtained from the batteries

(13)

11

instead of the grid. This can be used for frequency regulation and local peak shaving.

Other ways batteries can integrate with the grid today is by balancing fluctuations in electricity production, to avoid bottlenecks, and to ensure an uninterrupted power supply (Nordling, 2016). In this report, peak shaving will be the central feature investigated, although this does not exclude the possibility of any mentioned feature.

However, services mentioned above are energy-intensive and require characteristics of the batteries they do not possess today. According to Vattenfall, new markets for making battery storage economically viable will develop. But to reach a future of these battery services, the batteries have to be optimized for the services they are meant to perform. Put in the words of Vattenfall’s Batteries Director; “the constant cycling of the batteries are very energy-intense and affect the lifespan” (Nasner, 2019). Also, market incentives and cheaper production need to fall into place to make the development of these new energy services a viable solution (Nasner, 2019). According to IVA, price drops are occurring regarding lithium-ion batteries in the vehicle industry. With the use of batteries with reasonable price, expensive upgradings of the grid can be avoided (Nordling, 2016).

2.2.4 Development of cities

To understand how the V2G EV fleet may operate and how it can be implemented, the way cities are planning for the future is of great interest.

Local changes

One of the cities experiencing capacity deficiency today is Uppsala (Lindblom, 2018).

On the 28th of May 2018, the Uppsala municipal board adopted “Energiprogram 2050”.

Energiprogram 2050 is the plan and vision of the municipality regarding the

development of the energy system, as a part of making Uppsala fossil free in 2030 and climate positive in 2050. One of the aims is to develop the energy system and to integrate it with other systems in society, such as the transportation system (Uppsala kommun, 2018).

The municipality is aware of the fact that with a higher amount of local and renewable energy, the importance of managing and decreasing power peaks will grow. An aim to use renewable energy sources in combination with smart usage and energy storage integrated with the grid has therefore been formulated. By storing energy, it can be used when the power demand exceeds the production (Uppsala kommun, 2018). The

discussion regarding capacity deficiency in section 2.2.1 is in other words present in Uppsala as well.

In the Energiprogram it is also stated that an important part of future energy storage will be integrated in the infrastructure of the transport sector, and forecasts suggest that the transport system will be completely electrified. The municipality predicts that with technological development regarding energy storage and usage, commercial solutions

(14)

12

may develop for both the power grid and power consumers (Uppsala kommun, 2018).

One of these technological developments could be the development of new transport services, such as V2G integrated with Mobility as a Service.

Mobility as a Service

New technology and development of solutions for shared mobility, such as self-driving cars, is likely to affect how the public transports itself. The public’s travel pattern influence how the town or city itself develops, in terms of attractiveness as a place to live. However, there are big changes needed in order to enable more people to

participate in the public transportation system. Mobility, or transport, as a service is one of these potential developments, which would enable people to access transport on an as-needed basis. Shared mobility can take many forms, but the trends now point away from peer-to-peer platforms, such as car-pooling, towards a future with integrated services from several mobility providers into one single service. This development is aided by the development and use of digital solutions (Polle et al., 2018).

One of the possible developments mentioned above, include self-driving vehicles.

Studies have predicted that fully autonomous vehicles will start being phased into transport systems around year 2020-2025. Autonomous solutions such as these, may be a way to make public transportation more efficient as a system, providing transportation at a low cost for more people (Polle et al., 2018). A growing research and policy

consensus that transport systems based on privately owned internal combustion engine vehicles have a finite lifespan (Cooper et al., 2019), indicate that a future transportation system could be based on electric, autonomous vehicles. Further on, the system in the form of a commercially owned, self-driving EV-fleet with potential to interact with the electric grid has interesting potential for the future (Nelder et al., 2017). The fleet would essentially be enabling a new way of transportation.

2.3 The role of an EV-fleet in the power system

One way of managing a growing capacity deficiency and a way to even out power peaks in the grid might be to use energy storage in the form of V2G technology. In this

section, a presentation of how an EV-fleet could integrate the transportation and power systems by providing V2G services will be made. Challenges connected to using energy storages in the form of EV batteries will also be presented briefly.

2.3.1 Vehicle-to-grid (V2G)

In a future with smart electrical grids as system standard, EV batteries, which have a quick response rate, could be an asset to the grid in the form of providing charge to meet power demands at peak times. If a large number of EV:s could be centrally coordinated, the vehicles would be able to provide grid services as well as transport services, skipping manual intervention as would happen with vehicles owned by private

(15)

13

persons (Cooper et al., 2019). Consumer acceptance of V2G as well as attitudes are social challenges linked to V2G (Noel et al., 2019).

An ordinary EV is charged by connecting to the electricity grid, but unable to supply power back to the grid. With V2G technology, it would be possible to to create a “/.../

bidirectional communication and power flow between the EV and the power grid.”

(Noel et al., 2019). To make the V2G system work, there must be a way of connecting the EV to the grid bidirectionally, i.e. a specialized charger, as well as a way of

communicating to the EV when to charge from and discharge to the grid (Noel et al., 2019).

Some interesting aspects in a future scenario are the ways the V2G system may

integrate energy and transportation systems and what kind of services the system could offer the grid. With the ability to get information on the state of the power supply EV:s can “/.../ offer stability and flexibility as a market participant /.../” (Noel et al., 2019) and with a well-functioning synchronization, the EV:s can offer services that balances energy flows (Noel et al., 2019). In order to compensate instead of contributing to power peaks, a well-managed implementation is needed, especially on regional and local grid levels (Nordling, 2016).

In order to be part of the energy market, the EV-fleet has to meet the demands from the grid operators. The grid operators need to be ensured that offered power capacity will be charged and discharged at the right time. Highlighted advantages of V2G are that the EV:s together have a high capacity at a relatively low price, can react quickly when needed and have a high availability. At the same time, the EV:s have a limited energy supply capacity and the cost per unit of energy is higher in comparison with competitive solutions (Noel et al., 2019).

When the energy market has been charted regarding V2G, the considered highest valued service for today is stabilizing the imbalance of momentary power production and consumption, i.e. frequency regulation. Serving as a baseload power is considered unsuitable, with reasons such as not being able to provide continuous energy long-term.

With the prediction of a greater amount of intermittent energy sources, the mismatch between the electricity demand and generation might also be a problem that can be solved by the EV:s backing up the system. Another potential service is providing power compensation to the grid (Noel et al., 2019). In this report, power compensation due to capacity deficiency caused by transmission capacity deficiency will be focused on. The general idea of V2G examined in this report is illustrated in Figure 1 below.

(16)

14

Figure 1. The V2G system examined in this report.

The EV-fleet can offer its service to grids on different scales, even as small as micro grids. On a local level, the EV-fleet can improve the power supply and help avoid critical situations that otherwise might occur. However, it is important to remember that this kind of market allowing these kind of services does not exist today, and there is no way of knowing exactly how the future will develop regarding how the EV-fleet might integrate with the market. Thus, some potential services might still be unknown (Noel et al., 2019).

V2G solutions today are in a stage of development and there are only a few examples of the technology being used to provide services to an electricity market. The first example took place at the University of Delaware in the US, where the EV:s successfully

participated in frequency regulation (Noel et al., 2019). Another successful V2G-project took place in Denmark. Today, the company Nuvve runs the first commercial V2G project providing frequency regulation to the grid (Noel et al., 2019). Some conclusions from the now ended “Parker project” are that V2G technology is scalable and the market is ready, but also that the supply chain for both vehicles and charging infrastructure, as well as a clear business case are not yet in place (Andersen et al., 2019). Pilot projects involving V2G are also running in Sweden. Nissan, together with the municipality of Kungsbacka and E.ON, have initiated a project with the aim to install ten V2G units in Kungsbacka (Nissan News, 2018).

2.3.2 EV battery challenges

For an EV, there are limitations in the ways it can be used. The battery has a given energy storage that can not be exceeded. With services to the grid taken into account, the whole energy capacity can not be used (Mohammed et al., 2017). Another limitation is the fact that due to reducing stress on the battery, it is recommended not to charge or discharge the battery to its highest and lowest energy capacity. It is recommended to charge the battery 80 % of its maximum capacity (Nissan USA, n.d). The charging and discharging cycle pattern the V2G system would demand affects the battery in a negative way; it reduces its lifetime and performance (Mohammed et al., 2017).

(17)

15

3. Theory, Data and Methodology

In order to answer the research questions, theory, data and methodology for different components of the modeled system have been compiled. The chapter is divided into sections presenting each component of the modeled system based on its corresponding theory, data and methodology. The chapter is summarized in the last section.

Overview

An overview of this chapter is provided in Figure 2 below. Sections 3.1-3.4 aim to answer research question 1, 2 and 3. How the sections come together to answer all four research questions is presented in a Methodology summary, see section 3.5.

Figure 2. Overview of methodology sections.

Notes on methodology

Estimating future developments is always a tricky business. In this report, conditions that apply today together with assumptions that are continuously presented throughout the chapter have been used as an estimation of a future scenario. Locating the system to Uppsala and its estimated taxi fleet is motivated by the fact that capacity deficiency is a present and addressed problem in the area. Data obtained from Upplands Energi,

operating near Uppsala, will therefore be the local scenario examined in this report. The system will also be examined by using the taxi fleet of Uppsala for V2G services to the transmission grid. Where the EV:s will connect to the power grid is delimited to only being briefly discussed in chapter 5.

3.1 Exceedances

• Local exceedances

• Regional exceedances

• Quarter one 2019

3.2 Capacity of EV:s

• Capacity of Nissan Leaf e+

• Compare capacity with exceedances, powerwise and energywise

3.3 Taxi fleet

• New York City taxi fleet scaled to Uppsala

• Compare availability of the fleet with time of exceedances

3.4 Sensitivity analysis

• How will battery capacity impact the system?

• Capacity of Tesla Model 3

• Capacity need during

exceedances

(18)

16

The softwares used for data handling and calculations in this project are Microsoft Excel and Matlab R2018b. In addition to this, literature on the topics of V2G, Mobility as a Service and power exceedances has been used.

3.1 Exceeding network agreements

As mentioned in section 2.1.3, penalty fees apply when network agreements are exceeded. The structure of the fees differs depending on which actors are involved;

transmission and regional grid operators, or regional and local grid operators.

In this report, two cases will be examined; a Local case and a Regional case. The EV:s will connect to the electricity grid on a local level in order to perform power shaving services on higher levels of the grid, such as transmission level (Noel et al., 2019). The exceedances examined in the two cases are real exceedances of network agreements from different levels of the grid. The reason behind looking at a local and a regional case is to get a picture of the potential income and availability for the EV-fleet

providing power shaving services to grid operating companies at different levels of the power system. Since the magnitude of exceedances in the Regional case is considerably higher than exceedances in the Local case, not to mention the capacity of an EV battery, it is reasonable to not compare the stations of the two cases directly. Instead, three stations from the Local case will be compared to each transmission station. Details regarding magnitude of exceedances can be found in Table 2 and Appendix C.

3.1.1 Local case Theory

As mentioned in section 2.1.3, Upplands Energi has to pay fees when exceeding network agreements with the overlaying regional grid operator. Agreements are signed one year at a time (Österlund 2019c). Upplands Energi has agreements for individual stations, as well as a joint agreement for the total power consumption. Penalty fees apply when exceeding agreements at individual stations, as well as the joint agreement.

The former fee depends on the amount of exceeding power at a given time. The latter fee depends on the mean power consumption of two consecutive months. In this report, only exceedances at individual stations will be considered. The fee for exceeding the joint network agreement will not be considered since it does not represent a power peak in time (Österlund, 2019a). Exceedances occur almost exclusively during winter and are highly weather dependent (Österlund, 2019c).

Data

Data for exceedances during the first quarter of 2019 was sourced from Upplands Energi and is presented in Table 2 below. Three out of a total number of four

exceedances happened during the same day, with simultaneous exceedances at stations 1 and 2.

(19)

17

Table 2. Exceedances of network agreement during the first quarter of 2019, Local case.

Station Date and Time Exceedance [MW] Exceedance Fee [SEK]

1 190121, 18-19 1.163 52 335

2 190121, 18-19 1.680 75 600

2 190204, 07-08 0.070 3150

3 190121, 17-18 0.386 17 370

Total exceedance fee:

148 455 (Österlund, 2019a)

The data presented in Table 2 above is a good estimation of how quarter one usually looks like, according to Håkan Österlund, maintenance engineer at Upplands Energi.

Depending on weather conditions, exceedances may vary +/- 10 %, which affects the cost (Österlund, 2019b).

Methodology

Since the cost and time of exceedance at each individual station are known and

considered a good general estimation of exceedances during the first quarter of a year, results from using this data is presented in Table 10 and Figure 8, combined with the needed number and usage pattern of the EV:s. How the data is handled in order the answer research question 1 and 2 is described in more detail in section 3.5.

3.1.2 Regional case Theory

When a regional grid operating company is connected to the transmission network, the operator needs to pay a fee to SvK. SvK has a transmission grid tariff, which determines the fee an operator has to pay for transporting electricity on the transmission grid. It depends on both the amount of energy and power used. The fee related to power

depends on how many kW the network agreement covers (Svenska kraftnät, 2019). The regional grid operator is not allowed to exceed its network agreement, the only accepted way of using a higher amount of power than allowed is to get a temporary network agreement (Svenska kraftnät, 2018a).

Data

The regional grid operating company pays hourly fees when exceeding the network agreement (Svenska kraftnät, 2019). The data was sourced from SvK and the details regarding fees 2019 are shown in Table 3 below.

(20)

18

Table 3. Penalty fees for exceeding a network agreement, depending on the hour of exceedance, for 2019.

Hour of exceedance (during 24 hours) Exceedance fee [kr/MWh]

First hour 560

Second hour 1400

Third hour and following hours 2800

(Svenska kraftnät, n.d)

Data for exceedances during the first quarter of 2019 at three different transmission stations was sourced from SvK. The three stations were chosen to show varying patterns of exceedance magnitude and occurrence. The geographic locations of the three stations are unknown, due to the confidentiality of the data. The station data is presented in Tables 1A, 2A and 3A in Appendix A.

Methodology

As stated above in section 3.1, the EV:s will connect to the grid on a local level in order to perform power shaving on higher levels of the grid. When examining the

exceedances of the network agreement between regional and transmission grid

operating companies, the EV:s are assumed to be able to power shave seemingly at the point of exceedance and without losses. This is done in order to enable comparison of income with the Local case.

To determine an income for the EV-fleet providing power shaving at the transmission stations, data on exceedances at each station is used. The results from using this data are presented in Tables 1C, 2C and 3C in Appendix C as well as Table 11. How the data is handled in order to answer research question 1 is described in more detail in section 3.5.

To compare the exceedances at the transmission stations with the usage pattern of the EV:s, for each station, a day consisting of compiled mean values of all exceedances during the first quarter of 2019 was created, see Tables 1B, 2B and 3B in Appendix B.

All exceedances were added, and the mean value of each minute of occurring

exceedances was calculated by dividing the accumulated exceedance with the number of days with exceedance at that particular minute. By doing this, the data could be represented in a more comprehensible way and in addition to this, give an indication on the time and mean value of exceedances. Thus, note that the day of compiled mean exceedances does not show exceedances happening during the same day, but rather when in time they happen. By taking a mean value of simultaneous exceedances,

extreme exceedances might be smoothed out, eliminating the most extreme cases. These compiled days of exceedances combined with the usage pattern of the EV:s is presented

(21)

19

in Figure 9, 10 and 11. How the data is handled in order to answer research question 2 is described in more detail in section 3.5.

3.2 Capacity of EV:s

To determine the economic value the grid compensation services might bring to the company operating the EV-fleet, as well as the interplay of the EV-fleet and

exceedances, it is essential to know how many EV:s are involved in power shaving. To determine the number of EV:s, technical specifications regarding power and energy storage capacity have to be defined. In this section, the theory, methodology and data behind the results will be presented.

3.2.1 Theory

To determine the number of EV:s needed to match a particular exceedance, in both the Local and Regional case, calculations regarding the EV capacity are needed. Two characteristics of the EV are essential when comparing its capacity to an exceedance;

how the EV can match the exceedance powerwise, and how the EV can match the exceedance energywise, i.e. power over time. The specification demanding the higher number of EV:s will be the deciding factor. This entails investigating the specifications of the EV battery.

To calculate the amount of EV:s needed to match an exceedance powerwise, the following quantities, shown in Table 4, are needed:

Table 4. Quantities needed to calculate the amount of EV:s, powerwise.

Quantity Unit Description

Power exceedance MW Occurring exceedance in

the transmission or regional grid

Maximum charge rate MW Maximum charge rate

With these quantities mentioned above known, the number of EV:s can be calculated as follows: The Maximum discharge rate, i.e. the maximal power that an EV is capable of supplying to the grid, is assumed to be the same as the Maximum charge rate, see equation (1):

𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑑𝑖𝑠𝑐ℎ𝑎𝑟𝑔𝑒 𝑟𝑎𝑡𝑒 = 𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑐ℎ𝑎𝑟𝑔𝑒 𝑟𝑎𝑡𝑒 (1) The Number of EV:s, powerwise, can now be calculated using equation (2):

𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐸𝑉: 𝑠, 𝑝𝑜𝑤𝑒𝑟𝑤𝑖𝑠𝑒 = 𝑃𝑜𝑤𝑒𝑟 𝑒𝑥𝑐𝑒𝑒𝑑𝑎𝑛𝑐𝑒

𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑑𝑖𝑠𝑐ℎ𝑎𝑟𝑔𝑒 𝑟𝑎𝑡𝑒 (2)

(22)

20

To calculate the amount of EV:s needed to match an exceedance energywise, the following quantities, shown in Table 5, are needed:

Table 5. Quantities needed to calculate the amount of EV:s, energywise.

Quantity Unit Description

Power exceedance MW Occurring exceedance in the transmission or regional grid, during one hour

Energy storage MWh Maximal energy storage in an EV battery Percentage energy

storage

- The percentage of the storage of the battery that is ideal to charge

Percentage to grid

- The percentage of how much of the storage of the battery that can be discharged to the grid; this is dependent on

how much energy that should be left in the battery to perform transport services

When calculating the amount of EV:s involved energywise, Power exceedance needs to be translated into required energy, i.e. power over time. Today, penalty fees are charged one hour at a time. This means that the measured power at every new hour is assumed to have been constant over the past hour. As a response to this simplification, the EV:s featured in this report need to supply a constant power during one hour. An assumption is thus made; to cover an exceedance, the amount of EV:s needed to cover the whole exceedance is deemed to be the “sufficient” amount of EV:s. The required energy is calculated by multiplying the value of Power exceedance with one hour, as equation (3) shows:

𝐸𝑛𝑒𝑟𝑔𝑦 𝑒𝑥𝑐𝑒𝑒𝑑𝑎𝑛𝑐𝑒 = 𝑃𝑜𝑤𝑒𝑟 𝑒𝑥𝑐𝑒𝑒𝑑𝑎𝑛𝑐𝑒 ∙ 𝑂𝑛𝑒 ℎ𝑜𝑢𝑟 (3) The energy storage capacity is not used to its maximum, in order to extend the battery life. Further on, the battery does not discharge completely to the grid when power shaving, due to the fact that the EV must have the capability to perform transportation services. Thus, the Available energy storage to the grid is given in equation (4):

𝐴𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 𝑒𝑛𝑒𝑟𝑔𝑦 𝑠𝑡𝑜𝑟𝑎𝑔𝑒 = 𝐸𝑛𝑒𝑟𝑔𝑦 𝑠𝑡𝑜𝑟𝑎𝑔𝑒 ∙ 𝑃𝑒𝑟𝑐𝑒𝑛𝑡𝑎𝑔𝑒 𝑒𝑛𝑒𝑟𝑔𝑦 𝑠𝑡𝑜𝑟𝑎𝑔𝑒 ∙ 𝑃𝑒𝑟𝑐𝑒𝑛𝑡𝑎𝑔𝑒 𝑡𝑜 𝑔𝑟𝑖𝑑 (4) The Number of EV:s, energywise, can now be calculated using equation (5):

𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐸𝑉: 𝑠, 𝑒𝑛𝑒𝑟𝑔𝑦𝑤𝑖𝑠𝑒 = 𝐸𝑛𝑒𝑟𝑔𝑦 𝑒𝑥𝑐𝑒𝑒𝑑𝑎𝑛𝑐𝑒

𝐴𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 𝑒𝑛𝑒𝑟𝑔𝑦 𝑠𝑡𝑜𝑟𝑎𝑔𝑒 (5) To determine the Numbers of EV:s needed to match exceedances both powerwise and energywise, the largest output of equations (2) and (5) will be chosen as the resulting number of EV:s.

(23)

21 3.2.2 Data

Data regarding technical specifications of the EV battery has been based on the current EV model 2019 Nissan LEAF e+. The model has been selected due to the price

plausibility for a company investing in a large number of EV:s, as well as being a well known EV model of today. Nissan is also a company involved in projects regarding V2G today, see section 2.3.1. The data is presented in Table 6 below.

Table 6. Data regarding the EV model 2019 Nissan LEAF e+ and assumed Percentage to grid unique to this report.

Quantity Value

Maximum charge rate 0.050 MW*

Energy storage 0.062 MWh*

Percentage energy storage 0.8*

Percentage to grid 0.5

(*Nissan USA, 2019)

The value of Percentage to grid is determined based on assumptions made in section 3.3.1, i.e. the primary operations of the EV-fleet is transportation, which limits the energy available for grid services. With the given data, the following quantities in Table 7 have been calculated with equations (1)-(5) mentioned in section 3.2.1. The value of Available energy storage is rounded off in Table 7.

Table 7. Calculated quantities based on previous equations and known data regarding the EV model 2019 Nissan LEAF e+ in Table 6.

Quantity Value

Maximum discharge rate 0.050 MW

Available energy storage 0.025 MWh

Together with specifications concerning the exceedances presented later in the report, the number of EV:s needed to match each exceedance will be calculated.

3.2.3 Methodology

By using the calculated quantities in Table 7 combined with known hourly exceedances, the number of EV:s needed to match an exceedance can be calculated as shown in Figure 3 below.

(24)

22

Figure 3. Method for calculating number of EV:s needed to match an exceedance.

3.3 Taxi fleet

In order to determine the availability of the EV-fleet, the usage pattern of the taxi fleet of New York City has been examined. The reason for examining this usage pattern is based on the assumption that this might resemble the usage pattern of the EV-fleet in the future, with relatively short and frequent trips in a limited area.

3.3.1 Theory

No matter the economic value of the grid service, an assumption has been made that the EV-fleet will remain a taxi fleet, in other words; the EV:s will not stop driving people around. This assumption is made to ensure that the EV-fleet can still be considered offering Mobility as a Service. Therefore, it is reasonable to assume that the EV:s available for grid services will be the ones not performing transport services at that moment.

3.3.2 Data

To examine the usage pattern of the New York City taxi fleet, a dataset provided by the Illinois Data Bank has been used. The dataset consists of data from almost 700 million taxi trips during the years 2010-2013, in the form of pickup and drop-off dates, times, and coordinates, the metered distance reported by the taximeter, taxi identification number (medallion number), fare amount, and tip amount (Donovan and Work, 2016).

The authors of this report believe that in a future with self-driving EV:s that are commercially owned and operated, the usage of the service will result in more, but

Available power

• 𝑃𝑜𝑤𝑒𝑟 𝑒𝑥𝑐𝑒𝑒𝑑𝑎𝑛𝑐𝑒

𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑑𝑖𝑠𝑐ℎ𝑎𝑟𝑔𝑒 𝑟𝑎𝑡𝑒= 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐸𝑉: 𝑠, 𝑝𝑜𝑤𝑒𝑟𝑤𝑖𝑠𝑒

Available energy

• 𝐸𝑛𝑒𝑟𝑔𝑦 𝑒𝑥𝑐𝑒𝑒𝑑𝑎𝑛𝑐𝑒

𝐴𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 𝑒𝑛𝑒𝑟𝑔𝑦 𝑠𝑡𝑜𝑟𝑎𝑔𝑒= 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐸𝑉: 𝑠, 𝑒𝑛𝑒𝑟𝑔𝑦𝑤𝑖𝑠𝑒

Determine number of EV:s

• Determine Number of EV:s as the largest output from previous steps

(25)

23

shorter trips. This corresponds well to the usage pattern in the dataset, and motivates using it for modelling a future taxi fleet. In order to scale the fleet to Uppsala, the number of taxis operated by the two largest taxi companies in Uppsala was used to estimate the number of taxis in Uppsala. The total came down to an approximation of 150 (Uppsala taxi, 2019) plus 180 (Taxi kurir, 2019) taxis. In total 330 taxis were estimated to operate in Uppsala, with an estimated hourly income of 370 SEK per hour (Axelsson, 2019). The method used was telephone call to the companies.

3.3.3 Methodology

In order to calculate the availability of the taxi fleet, the following data extracts were chosen: pickup and drop-off dates and times, as well as the medallion number. From this, the activity of each active taxi during each minute of a certain day could be determined. The activity is here defined as the number of taxis performing transport service during a particular minute. By dividing the activity each minute with the total amount of active taxis during a day, a percentage of active taxis during each minute was calculated.

Due to the magnitude of the dataset and the fact that access to big data-software was unavailable, delimitations had to be made. This means that only a few days were sampled, and some of the days do not have 24 hours of data, since some parts of the data was not read. Winter days were chosen due to the fact that power peaks are more frequent during cold winter months (Svenska kraftnät, 2017b). Data from the first two days of the months November, December and February were chosen. Only one day from January was chosen, since the first day of January involves taxi usage during New Year’s, which can be considered exceptional usage. As mentioned in 3.3.2, not all days had 24 hours of sampled data. These are data points lacking information, and thus not included in calculating a mean day, as Figure 4 below shows.

Figure 4. Method for calculating the activity of the taxi fleet during a mean winter day.

Since the activity is now a percentage, the availability during each minute of the mean winter day is defined as 1-percentage of active cars. This is a simplification due to the fact that data only exists for when the taxis are occupied. Activity occurring when taxis are off-duty is therefore unknown.

The population in New York City in 2010 was 8 175 133 (NYC Government, n.d.) and the total number of taxis in the dataset is 13 164 (Donovan and Work, 2016). The quota

Calculate the activity each minute of each

day

Sum up the activity of all minutes of all days, omit data points that lack information due to

delimitations

Calculate a mean day by dividing the summed

up activity for each minute by the sampled

amount of days

(26)

24

taxis/inhabitant in New York City is approximately 0.0016. The current population in Uppsala is 376 354 (SCB, 2019) and using the estimated number of taxis concluded in section 3.3.2, the corresponding quota for Uppsala is 0.00088. The higher quota in New York City motivates using this usage pattern to represent a more frequent use of taxi services in Uppsala, i.e. an EV-fleet providing Mobility as a Service.

Finally, the number of taxis available was scaled to Uppsala by multiplying the percentage of available taxis with the estimated number of taxis in Uppsala. When plotting the result, the first and last 20 minutes of each day were cut, since these represent a non-realistic increase and decrease indicating that all taxis stand still at midnight, which is not the case. Trimming the data like this does not affect the V2G system, since no exceedances investigated occur during this time.

3.4 Sensitivity analysis

In order to evaluate one aspect of the system’s sensitivity, the impact of the EV batteries, i.e. charge rate and energy storage, will be evaluated based on the following questions:

▪ If the capacity of the EV battery were to change in accordance to probable development of batteries, what would the impact on the system² be?

▪ In order to fully cover the exceedance with an unchanged number of EV:s, what would the needed EV battery capacity have to be?

By doing this, how sensitive the system is to battery capacity can be further

investigated. The exceedances in the most extreme scenario, illustrated in Figure 11, will be examined. When answering the second question, exceedances that can not be covered will be examined, since these represent situations needed to be resolved for the system to improve.

The method for answering the first question involves using data for an EV battery considered to represent the general future development of EV batteries. A sound estimation is to use the battery of Tesla Model 3, since the model is in the same price range and market as the Nissan Leaf e+, but with better range (Shepero, 2019). In this sensitivity analysis, a Tesla supercharger with a charging rate of 120 kW is assumed to be used (Shepero, 2019). The specifics regarding the EV model is presented in Table 8 below. Note that the last two parameters are unchanged. The values can be compared to specifics regarding the Nissan Leaf e+ model in Table 6.

(27)

25

Table 8. Data regarding the EV model Tesla Model 3.

Quantity Value

Maximum charge rate 0.120 MW*

Energy storage 0.075 MWh*

Percentage energy storage 0.8**

Percentage to grid 0.5

(**Lambert, 2017; *Nissan USA, 2019)

The variable Percentage to grid is based on assumptions stated in section 3.3.1. By using equations (1)-(5) in section 3.2.1, the following quantities were calculated, see Table 9.

Table 9. Calculated quantities based on previous equations and known data regarding the EV model Tesla Model 3 in Table 8.

Quantity Value

Maximum discharge rate 0.120 MW

Available energy storage 0.030 MWh

The number of EV:s are determined by using the method presented in Figure 3. The result is compared to the availability of the EV-fleet and presented in section 4.3.

The method for answering the second question includes using specifics for an ideal EV according to Table 5. The number of EV:s available for power shaving is determined as the mean number of available EV:s during the hour of exceedance.

By using equation (4), Available energy storage is calculated. Energy exceedance is calculated by using equation (3). The result is presented in Table 12.

3.5 Methodology summary

To answer the first research question: “What is the potential economic value of a fleet of EV:s providing service to the grid?”, the income must be estimated. By knowing the actual cost of exceeding a network agreement at a given time, the income of each exceeding for the EV-fleet operating company can be determined as in equation (6).

𝐼𝑛𝑐𝑜𝑚𝑒 𝑝𝑒𝑟 𝑒𝑥𝑐𝑒𝑒𝑑𝑖𝑛𝑔 = 𝐴𝑣𝑜𝑖𝑑𝑒𝑑 𝑓𝑒𝑒

𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐸𝑉: 𝑠 (6)

(28)

26

Since the Avoided fee is an hourly fee, the Number of EV:s must be sufficient to cover the exceedance the whole hour to earn that hourly income. Therefore, Number of EV:s will be compared to the mean number of EV:s available during the hour of exceedance to determine if Number of EV:s is sufficient at the time. The income will be presented as both the total income during the first quarter of 2019 and the average income per EV and exceedance. The income will be determined for both the Local case and Regional case. For an overview, see Figure 5 below.

Figure 5. Flowchart of how to answer research question 1. The method is applied in both the Local and Regional case.

To answer the second research question: “What is the interplay between the EV-fleet and the grid, regarding availability for grid services?” the usage pattern can be compared with the number of EV:s needed for power shaving, distributed over 24 hours. The interplay between the transport and grid service can be illustrated by plotting the usage pattern and the needed number of EV:s to power shave in the same graph.

This will be determined for both the Local case and Regional case. In the former case, every exceedance during the first quarter of 2019 will be presented in the same graph, to illustrate which exceedances could potentially be covered. In the latter case,

exceedances of a day consisting of compiled mean values of all exceedances during the first quarter of 2019 for each transmission station will be presented in separate graphs, to illustrate which of these exceedances could potentially be covered. Representing the exceedances this way is done to show occurrence and magnitude of exceedances in order to illustrate the interplay of exceedances and the EV-fleet at each individual station. The number of EV:s needed to match each of these exceedances is presented in Tables 1B, 2B and 3B in Appendix B. For an overview, see Figure 6 below.

Figure 6. Flowchart of how to answer research question 2. The method is applied in both the Local and Regional case.

To answer the third research question: “If the battery capacity of the EV is changed, what is the impact on the system³?” a sensitivity analysis will be made, see section 3.4.

Cost of exceedance

Required number of EV:s

involved

Is a sufficient number of EV:s

available?

If yes, income

Plot availability of the EV-fleet

of Uppsala

Needed number of EV:s to

match exceedances

Plot needed number of EV:s

at time of exceedances

Determine interplay

(29)

27

To answer the fourth research question: “Based on today’s discussion and results from previous research questions, which key features are important to address when

implementing the system?”, results from the three former research questions will be discussed in a wider context, with themes stemming from the Background (chapter 2).

4. Results

In this part of the report the results of research question 1, 2 and 3 is presented in different sections. In order to answer research question 1 and 2, the Local case and the Regional case is presented in sections 4.1 and 4.2. The third research question is answered in the Sensitivity analysis, section 4.3. The fourth research question is answered in the Discussion, chapter 5.

4.1 The economic value

To answer the first research question: “What is the potential economic value of a fleet of EV:s providing service to the grid?”, the income of the EV-fleet in the Local and Regional cases, respectively, has been determined. Power shaving services generating income for the fleet can only be provided if there is a sufficient amount of EV:s

available at the time of the exceedance, i.e the fleet must cover a whole exceedance. The availability of the EV-fleet is illustrated in Figure 7, section 4.2.1. The determined number of EV:s involved in covering exceedances was, in both cases, determined as the number of EV:s energywise.

4.1.1 Local case

The total income for the EV-fleet during the first quarter of 2019 in the Local case is presented in Table 10. The Table shows the total income for the EV-fleet and the average income per EV and exceedance. Note that two exceedances occurred

simultaneously at station 1 and 2. In this case, the income is related to which of these exceedances, (a) or (b), the company operating the EV-fleet prioritize; the highest income per EV or the highest total income per exceedance. In Table 10, the index (a) means that this exceedance, together with the unlabeled exceedances, are considered when calculating the “Total income” and “Average”. In the same way, the index (b) means that this exceedance, together with the unlabeled exceedances, are considered when calculating the “Total income” and “Average”.

(30)

28

Table 10. Exceedances of network agreements during the first quarter of 2019 in the Local case, shown for each station. Note that two exceedances occurred simultaneously

190121, (a) and (b). Income per EV is rounded off to whole numbers. Average and Total income are rounded off to two significant numbers.

Station Date and Time of exceedance

Exceedance [MW]

Exceedance Fee [SEK]

# EV:s needed to

match exceedance

#EV:s available

Income per EV [SEK]

1 190121, 18-19

1.163 52 335 (a) 47 149 1114 (a)

2 190121, 18-19

1.680 75 600 (b) 109 149 694 (b)

2 190204, 07-08

0.070 3150 3 255 1050

3 190121, 17-18

0.386 17 370 16 184 1086

Total income:

73 000 (a) 96 000 (b)

Average:

1100 (a) 940 (b)

4.1.2 Regional case

The total income for the EV-fleet during the first quarter of 2019 and the average income per EV and exceedance based on covering exceedances at three transmission stations, A, B and C, are presented below in Table 11. The results come from the results in Table 1C, 2C and 3C presented in Appendix C. Each of these tables contains the number of EV:s needed to match an exceedance, the number of available EV:s at the time of an exceedance and resulting total income and income per EV and exceedance.

An average of the latter result is presented in Table 11.

Table 11: Total income for the EV-fleet, average Total income and Average income per EV and exceedance. Exceedances occurred during the first quarter of 2019. Results are presented for each station. Average income per EV and exceedance and Total income are rounded off to whole numbers. Average is rounded off to two significant numbers.

Station Total income [SEK] Average income per EV and exceedance [SEK]

A 11 760 19

B 15 680 42

C 17 360 17

Average 15 000 26

(31)

29

4.2 Availability for grid services

To answer the second research question: “What is the interplay between the EV-fleet and the grid, regarding availability for grid services?”, the number of EV:s needed to match each exceedance is compared with the usage pattern of the EV fleet in the Local and Regional case, respectively.

4.2.1 Availability of the EV-fleet during an average winter day

The estimated average number of EV:s available in Uppsala during an average winter day is presented in Figure 7 below. The mean value of the availability of the taxi fleet in New York City an average winter day has been scaled to the amount of taxis in

Uppsala.

Figure 7. Estimated average number of EV:s of the commercially owned EV-fleet available in Uppsala during an average winter day.

4.2.2 Local case

The number of EV:s needed to match every exceedance during the first quarter of 2019, see Table 10, compared with the average availability of the EV:s during a winter day is shown in Figure 8 below. Note that two exceedances occurred simultaneously at station 1 (**) and 2 (*), during 18-19 190121. Figure 8 shows that the number of EV:s are sufficient to cover three out of four exceedances.

(32)

30

Figure 8. Comparison of EV-fleet availability and EV:s needed to match exceedances in the Local case. Note that during 18-19 190121, 109 or 47 EV:s are involved, depending

on exceedance covered. The red* and yellow** bars are stacked.

Figure 8 shows that the interplay between the EV-fleet and the grid regarding

availability for grid service seems to be sufficient in the Local case if not considering covering both exceedances during the time 18-19, see Figure 8. The largest exceedances occurs during the time when the EV-fleet is most occupied with transport service, 18-19. During 07-08, the margin is the greatest, with an additional 200 EV:s idle.

4.2.3 Regional case

The number of EV:s needed to match exceedances at station A during a day consisting of compiled mean values of all exceedances occuring during the first quarter of 2019, see Table 1B, compared with the average availability of the EV:s during an average winter day is shown in Figure 9 below. The figure shows that the number of EV:s are not always sufficient to cover every exceedance, given the assumption made in section 3.5.

A package deal for the future: Vehicle-to-Grid combined with Mobility as a Service

Examensarbete 15 hp Juni 2019

A package deal for the future:

Vehicle-to-Grid combined with Mobility as a Service

Amanda Bränström

Jonna Söderberg

Abstract

A package deal for the future: Vehicle-to-Grid combined with Mobility as a Service

Table of contents

Acknowledgements

1. Introduction

1.1 Project aim

1.2 Research questions

1.3 Delimitations and limitations

1.4 Overview of the report

2. Background

2.1 Managing capacity deficiency today

2.2 Today’s outlook for the future

2.3 The role of an EV-fleet in the power system

3. Theory, Data and Methodology

3.1 Exceeding network agreements

3.2 Capacity of EV:s

3.3 Taxi fleet

3.4 Sensitivity analysis

3.5 Methodology summary

4. Results

4.1 The economic value

4.2 Availability for grid services