The Value of Value Factors

(1)

Juni 2020

The Value of Value Factors

Time-Dependent Development of Value Factors

on the Swedish Electricity Market

(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

The Value of Value Factors

Agnes Bråve, Nora Ekström, Sara Särnblad and Katarina Vanky

This bachelor thesis investigates the development of value factors on the Swedish electricity

market and how the development can be explained. Value factor is a parameter that indicates

how well an energy source’s market price corresponds to the average spot price for the

electricity mix. Value factors for nuclear-, thermal-, wind-, solar- and hydropower are calculated

for the years 2014-2019. Electricity production- and spot price data has been sourced from

Svenska Kraftnät, Nord Pool and Uppsala University. The influence of weather conditions, spot

price, production and consumption on the development of the value factors is discussed. The

Pearson correlation coefficient is used for analytical purposes, showing the correlation between

two specific variables. The conclusion is that the value factor for each power source is the result

of the conditions present during the specific time period. The value factors for solar-and

thermal power are discontinuous since they are temperature-dependent. For nuclear-, wind-

and hydropower, the value factors are more continuous during the time span. This is due to, for

instance, their important roles in the Swedish energy system and their ability to match demands.

(3)

Preface

This study is a bachelor thesis from the MSc program in Sociotechnical Systems Engineering profiling on Energy systems at Uppsala University. It was done in a collaboration with Johan Lindahl and Energiforsk. The produced value factors will be used by Energiforsk in a future study.

(4)

1

Important concepts

Day-ahead market Energy exchange trading for the next day. Intraday market Energy trading within the 24 hours of the day.

Spot price The electricity price that is set every day on Nord Pool. Market value The relationship between the average spot price of the

electricity produced by a power source and its production share on the market.

Production profile Indicates how the production for an energy source varies with time.

Prosumer An electricity user that acts as a producer and a consumer.

IEA-PVPS International Energy Agency Photovoltaic Power System Program. Since its inception in 1993, it has carried out a series of joint R&D projects related to solar cells.

Regulation power Power sources that can be regulated in order to match the electricity demands of consumers.

Baseload The minimum level of consumer demand

on the electricity grid.

Intermittent power Power that is produced when the weather conditions allow.

Thermal power Electrical power produced by a steam boiler fueled by for instance biogas or coal.

(7)

4

1. Introduction

Today’s society is faced with a multitude of challenges. One of the most urgent is the rise of the average temperature in the atmosphere which inevitably will lead to changes in the

environment. Global warming and its potential consequences have led to the development and implementation of fossil free electricity production such as solar and wind power. From 2014 to 2019 installed wind power in Sweden increased from 5088 MW to 8681 MW, an increase of 71 % (Energimyndigheten, 2020). Solar power similarly grew from 79,4 MW installed power in 2014 to an estimated 700-800 MW in 2019, an increase of 781 % (Villas, 2019) (Lublin, Zofia and Friberg, 2016). Since these sources of electricity production are intermittent and dependent on ambient conditions, such as the weather and geographical location, they affect and change the electricity market. This affects traditional electricity production sources where the output of electricity to the market can be regulated and controlled for.

The parameter to be investigated in this thesis is the value factor. The value factor indicates whether the electricity market’s demand is in line with the power source’s profile of

production and is therefore useful in a multitude of ways. For instance, to further understand

the complex connections at play between baseload, regulation power and intermittent power

sources in the electricity production system. The value factor of a power source is calculated

as the market value for a certain production relative to the average price on the market. A

value factor > 1 indicates that the value of electricity production exceeds the average spot price during a certain time period (usually one year), and conversely if the value factor is < 1. Hence a, value factor > 1 for a power source indicates that the electricity market demands on electricity production conforms to the production profile of that power source (Hirth, 2015). The organization Energiforsk is currently conducting the study Electricity from new power

plants. The purpose of their study is to provide a current compilation of costs involved in

producing electricity and present factors that influence electricity production costs. Their project will also highlight different system costs for integration of generated electricity in the power system from the different power sources. Energiforsk are interested in the value factors for solar-, wind-, thermal-, nuclear- and hydropower to be calculated for the time period 2014-2019.

1.1 Aim of the report

The aim of the study is to calculate and analyze value factors for solar-, wind-, thermal-, nuclear- and hydropower during the time span 2014-2019 in Sweden. The development of the value factors will then be analyzed and discussed. The further aim is to look deeper into

factors that might have an impact on the outcome of the value factor. These results are

(8)

5

1.2 Thesis questions

In order to achieve the aforementioned aim, the two following questions will be investigated: 1) How has the value factor for the five different electricity sources changed during the

period 2014-2019?

2) How can the change in value factors be explained during the period?

1.3 Delimitations

The study was delimited to study the Swedish electricity market during 2014-2019. It was also delimited to five different types of electricity production.

To analyze the change in value factors the focus was delimited to data concerning weather conditions, production, consumption and spot price. To retrieve weather data, conforming to reasonable accuracy, a mean value was calculated from five different weather stations in every specific price area. This can be misleading, since the geographical areas differ in size and population density. Spot price data are based on available hourly spot prices sourced from the day-ahead trading market. Earlier studies show that the day-ahead prices are strongly associated with market values (Lingfors et al., 2019).

Export and import of electricity to and from neighboring countries was not considered, neither international factors such as fuel- and oil prices.

Thermal power does not include electricity produced by nuclear power in this report. The model for simulating solar power data includes assumptions and is described further in 3.2.2 Solar power data. Solar power is limited exclusively to photovoltaic (PV) solar cells.

1.4 Outline

(9)

6

2. Background

In this section all relevant information pertaining to the report and its purpose are presented. Information concerning the specifics of the Swedish energy system are outlined, as well as descriptions of the calculation model used.

2.1 Swedish power sources

The Swedish electricity production system consists of several different energy sources. Electricity sold to consumers is a combination between base load, regulation power and intermittent sources. Currently in Sweden, energy sources producing electricity to the market consists of nuclear-, thermal-, wind-, hydro- and solar power. Nuclear- and hydropower provide the main share of produced electricity and thereof have the largest production shares. In 2018, nuclear- and hydropower had a combined market production share of around 80 %. Wind- and solar power are intermittent sources which means that their production shares are determined by uncontrollable events such as weather conditions. Wind- and thermal power produces about 10 % each of the total amount of produced electricity (Svenska Kraftnät, 2020).

Figure 1. Electricity production 2018 with data from Statistiska centralbyrån. The solar power is only the power on the grid, i.e., the self-consumed power produced by prosumers is

not included (Statistiska Centralbyrån, 2019).

(10)

7

production ahead of time. This means that regulating power sources, such as hydropower, are becoming increasingly important in order to be able to regulate electricity production as the weather varies (Uniper, n.d.). Nuclear power produces the baseload whilst hydropower offers both baseload and controllability in regard to meeting peak demands of consumers (Svenska Kraftnät, 2016). Figure 2 shows production for all five power sources depicted from 2014-2019.

Figure 2. Hourly average of production during 2014 -2019. The generated solar power includes the power from prosumers.

Solar power has a unique role in the energy system since installed solar power can produce electricity directly to the consumer. For example, living in a house on which the solar panels are installed. Privately installed solar power can also feed electricity back to the grid. These actors are both consumers and producers, often referred to as prosumer (IVA, 2016).

2.2 Developments of Swedish power sources

The electricity sources used in today’s society are constantly evolving. New technologies are being developed and society is striving to use more and more renewable energy. Green power sources such as wind- and solar power are growing quickly. Both have an important role today and will have an even more important role in the future Swedish electricity grid. This is based on the decision to decommission nuclear power. Following a referendum in 1980 the Swedish government decided that nuclear power should be disassembled by 2010 at the latest. However, political objectives have fluctuated, and settlement plans have changed. Several nuclear reactors have been shut down, but after 2020 there will still be six reactors in use and those are prepared to operate until 2040 (Lindholm, 2020). Nuclear power is a major part of

0 2000 4000 6000 8000 10000 12000 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 2014 2015 2016 2017 2018 2019 Pr od uc tio n (M W )

(11)

8

Swedish electricity production and when it is decommissioned it must be replaced (Energiforsk, n.d.).

There will be changes in the Swedish energy system as wind-, solar- and thermal power increase and take up larger production shares. This creates demands on the national grid, but also on the local and regional power system. Different types of production are connected at different voltage levels; therefore, all parts of the network are affected, and when these growing types of energy sources reaches certain levels, the grid must be adapted. There are also major challenges in the expansion of transmission lines, that is, the ability to transport electricity from one part of Sweden to another. This makes it possible to cover electricity demand where there is a shortage (IVA, 2016).

The expansion of intermittent power sources means a more distributed electricity generation, but also a greater uncertainty in production.It is not only the electricity grid that is changing structure and being rebuilt, more players are getting involved, including companies, industries and private individuals as well as other investors (IVA, 2016).

2.3 Electricity price areas

Since November 2011 Sweden is divided into four different electricity price areas. These are numbered from 1 to 4, from north to south of Sweden. Luleå (SE1), Sundsvall (SE2),

Stockholm (SE3) and Malmö (SE4), see Figure 3.

Figure 3. Electricity price areas in Sweden.

(12)

9

The supply is greater in northern Sweden, since most of the electricity generation takes place there on an annual basis, while demand is greater in southern Sweden because of greater population density. Many industries are also located there. In the north of Sweden,

hydropower dominates and in the central-southern part of Sweden nuclear power is a major source of electricity production. The current electricity distribution and the country's different needs in different areas lead to a certain imbalance in the electricity grid. Electricity is

transported by transmission lines from northern to southern Sweden to accommodate demands from consumers. Because of this the electricity price is generally higher in the southern part of Sweden than in the northern parts. For SE1 and SE2 the spot prices are almost invariably identical (see Figure 4). The purpose of price areas is to make it more profitable to produce electricity where it is used and reduce the need to transport electricity therefore reducing transmission losses. In the long run the prices will probably be leveled (Energimarknadsbyrå, 2020).

Figure 4. Average price per hour and electricity price area year 2014.

2.4 Electricity market

Nord Pool is the Nordic electricity market exchange. They offer two different marketplaces: day-ahead and intraday market trading. The day-ahead market is the main exchange for trading electricity where producers can buy and sell electricity for the next day. It creates an opportunity to preview the supply and demand in advance, which in turn makes it possible to plan the next 24 hours and use power sources in the best possible way (Nord Pool, n.d.-a). The intraday market is trade within the 24 hours of the day and helps secure the balance

0 100 200 300 00 - 01 02 - 03 04 - 05 06 - 07 08 - 09 10 - 11 12 - 13 14 - 15 16 - 17 18 - 19 20 - 21 22 - 23 Sp ot pri ce (SEK /M W h) Time (h)

(13)

10

between supply and demand. Therefore, the intraday market is a supplement to the day-ahead market (Nord Pool, n.d.-b).

There are several factors affecting the pricing of electricity, the spot price, on the market. Overall, supply and demandare the two main factors that are strongly linked to each other. If the total production decreases and the demand increases, the spot prices rise

(Energimarknadsbyrå, 2020).

2.5 Weather impact on electricity production

This section will highlight how the weather can impact electricity production. There are many uncontrollable events that affect electricity production of different power sources. Even dispatchable power sources are to some extent affected by changes in weather and season to match changes in consumer demands. This results in power sources’ production shares changing throughout the year.

The variation of wind

Wind power is an important and growing part of the Swedish energy system. The deployment has been successful because Sweden’s coastlines and mountainous areas provide beneficial wind conditions. A modern windmill utilizes about 50 % of the kinetic energy in the wind. It starts producing electricity at wind speeds of 4 m/s and can produce effectively until the wind reaches a speed of around 25 m/s. The maximum output power can be obtained at wind speeds of 12-14 m/s (Jämtkraft, n.d.). The wind speed varies constantly but in the long term it is possible to see a pattern in the variation. The energy content of the wind also varies day and night, as there is generally higher wind speeds during the night. The production also varies annually and between seasons. There is a lot more wind during the winter months, about two-thirds of the electricity extraction from the wind occurs during this time (Centrum för

Vindbruk, 2013).

Solar radiation and temperature

Through solar radiation electricity can be produced using solar cells. Electricity production from solar power plants depends on factors such as available solar radiation, surface area and temperature. Temperature affects the efficiency of the cell by reducing its performance at increased temperature, meaning, electricity generation is improved when it is sunny at the same time as it is windy in comparison with a sunny and windless day, since the wind cools the panel. In Sweden, production from solar cells varies greatly over the year due the geographical location of Sweden at high latitudes. The main production takes place during spring, summer and autumn. During the winter sunshine hours are few, which means that the production is lower (Söder, 2014).

(14)

11

as long as it does not cover the panel. This means that Sweden have good conditions for electricity generation from solar cells despite sub-optimal solar conditions. Sweden has reasonably low average temperatures outdoors, which means a reduced temperature inside the solar cell and therefore the solar cell works better in operation (Utellus, n.d.).

Precipitation

Annual precipitation can affect power sources’ ability to produce electricity, especially a lack thereof. Hydropower, for instance, is dependent on a steady flow of water from rivers and dams (Naturskyddsföreningen, 2019). Although hydropower is less affected by general weather conditions, annual precipitation affects its performance in electricity production. A prolonged period of drought can result in dams not refilling properly and therefore limiting its ability to produce electricity (SMHI, 2018). For example, the dry summer in 2018 affected hydropower plants in Arvika, Sweden, where a 40 % decrease in production was noted due to the lack of flowing water from connected reservoirs (SVT, 2018).

2.6 Value factor

The value factor is a central parameter for this thesis. It indicates how well an energy source’s market price corresponds to the average spot price for the electricity mix (Hirth, 2013). The market value depends on the production share for the energy source in the electricity mix and the spot price is an active electricity price on the day-ahead market (Nord Pool Spot, 2011). If the value factor is > 1 it means that the price of electricity production for an energy source exceeds the average spot price. This means that the electricity produced is worth more than the spot price during the time period. For a value factor < 1 the opposite applies. The value factor is affected by market demand together with the energy source’s production profile (Hirth, 2013). The market value represents the relationship between the average spot price of the electricity produced by a power source and its’ production share on the market. The market value is an integral part of the value factor and points to whether the power generation from a specific energy resource is matched to the market’s demands during the time period.

Model description of the value factor The equation used for calculating the value factor is:

𝑉𝐹 =!"""""_!̅! , (1)

where 𝑝%%%% [SEK], the market value, and 𝑝̅ [SEK], the average price value, are defined as $ 𝑝$

%%%% =∑#"$%&"!"

(15)

12

𝑝̅ ='₍∑()*'𝑝), (3)

where 𝑎₎ and 𝑝₎ are the production share for the specific energy source and the corresponding spot price at time step t and T is the number of time steps.

2.7 Previous work

Previous work has been conducted regarding value factors. Value factors are commonly used in order to evaluate different energy sources market value in comparison with the actual price and can be an important tool in further analyzes, both socioeconomical and environmental. To evaluate different price scenarios on the Swedish electricity market when increasing the solar power share in the electricity mix, Uppsala university investigated four cases, when 1 %, 5 %, 10 % and 20 % of the electricity demand was covered by solar power (Lingfors et al., 2019)

(16)

13

3. Methodology

In this section the methodology used for calculations regarding value factors and correlations will be described. The model parameters used will be presented as well as data and tools utilized.

3.1 Model description

Value factor

To reach the aim of the study and to calculate the value factors for each of the years 2014-2019 equation (1), (2) and (3) were used where T is the time in hours for one year, more specifically 8760 hours per year except for leap years (i,e., 8784 hours). All the values used are measured hourly during one specific year.

Correlations

To find connections concerningthe value factor and outstanding parameters such as

precipitation, the Pearson correlation coefficient, r, a statistic that calculates linear correlation between two continuous variables, was used. The output is values between –1 and 1, i.e., corresponding to completely negative- and positive correlation, respectively. Negative

correlation means that when one of the parameters increases the other one decreases. Positive correlation is when both parameters develops in the same direction, thus when one increase or decrease the other one will do the same. A value of 0 indicates that no correlation can be identified between two factors (Kent State University, 2020).

The model for the Pearson correlation coefficient can be describes as 𝑟 = +,-(/,1)

3-&4(/)∗3-&4(1) (4)

Where cov(x,y) is the covariance of x and y. var(x) and var(y) are the variance of x and y (Kent State University, 2020).

Sensitivity analysis

To determine the relevance of using self-consumed power produced by solar power

(17)

14

3.2 Data

In this section the data for calculating the value factors, the sensitivity analysis and the correlations are described.

Data value factor

Two main kinds of data have been collected: hourly spot price- and corresponding production data. The spot price data were collected from Nord Pool and reported in SEK/MWh (Nord Pool, 2020). The production data were retrieved from Svenska Kraftnät and reported in MWh/h (Svenska kraftnät, 2019).

Solar power data

The solar power production data used in the value factors calculations were obtained from Uppsala University. The data were generated through simulations using the method from Lingfors & Widén (2016), since a large share of the total solar power production is self-consumed by prosumers and therefore most often not included in the statistics from Svenska Kraftnät. The simulation result was generated in proportion to the geographical location of the population and the available solar radiation, this to ensure that the solar production is

distributed realistically. Some assumptions were made in the simulation to provide the optimal and most realistic data. For instance, the simulation was made with a fixed tilt of 45° towards cardinal south even though some solar panels have another degree of tilt. For each year the production was calculated from the average installed power at the beginning and the end of the year and was weighted against values on the installed capacity from IEA-PVPS (Honghua et al., 2012). 2019 data were obtained from Statistics Sweden since IEA-PVPS values are missing. The distribution of installations in the different price areas from Statistics Sweden is used for 2016-2019. For 2014-2015, where Statistics Sweden data do not exist, the 2016 distribution was used and weighted against the IEA-PVPS as mentioned above.

Weather data

To provide a basis for analysis for the development of value factors climate data were retrieved from the Swedish Meteorological and Hydrological Institute (SMHI). Data

(18)

15

4. Results and discussion

In this section the results will be presented and discussed simultaneously. In section 4.1 an in-depth analysis for 2014 is presented. The purpose of this is to show the impact of weather conditions, geographical location, production, consumption, and spot prices are related and affect the value factor. The parameters that affect the value factor are production,

consumption and spot price. These parameters in turn are affected by seasonal changes in consumption behavior due to fluctuations in, for example, temperature. A deeper analysis for one specific year generates a greater understanding of the outcome and change of value factors over time. In section 4.2, value factors and Pearson correlation coefficients for 2015-2019 are presented, followed by a comparative discussion over the time period. In section 4.3, results from the sensitivity analysis are presented and discussed. Lastly, in section 4.4, a discussion on the future implications of the results is presented.

4.1 Analysis of value factors and correlations for 2014

(19)

16

Table 1. Value factors for each price area and average value factor for every energy source 2014.

Hydropower Nuclear power Thermal power Wind power Solar power

SE1 1.023 0.997 0.983 0.959 1.072

SE2 1.023 0.987 0.983 0.959 1.072

SE3 1.024 0.985 0.984 0.957 1.075

SE4 1.026 0.984 0.987 0.954 1.069

Average 1.024 0.986 0.984 0.957 1.072

Table 2. Pearson correlation coefficient table for 2014.

Weather conditions 2014

(20)

17

Figure 5. Correlation between daily share of daily max for production of solar and thermal power and temperature for each electricity price area during 2014.

During the summer months when the temperature is rising electricity consumption decreases. Therefore, consumption and production have a negative correlation with temperature. During this season some nuclear power plants close for maintenance and therefore have a negative correlation with temperature as well. It follows that nuclear power’s production share must be replaced by other production sources. Therefore, it works as a price-controlling signal on the Swedish electricity market. This is shown in Figure 6 below. This explains why annual temperature has a low positive correlation with the spot prices and nuclear power has a negative correlation to spot price.

Figure 6. Correlation between nuclear power production and spot price during 2014. -30,0 -20,0 -10,0 0,0 10,0 20,0 30,0 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 14- 01-01 14- 02-01 14- 03-01 14- 04-01 14- 05-01 14- 06-01 14- 07-01 14- 08-01 14- 09-01 14- 10-01 14- 11-01 14- 12-01 Te m pe rat ure (° C) Pr od uc tio n sh ar e (% ) Date

Thermal power Solar power (generated) Temperature SE1 Temperature SE2 Temperature SE3 Temperature SE4

0,0 50,0 100,0 150,0 200,0 250,0 300,0 350,0 400,0 450,0 500,0 0,0 50000,0 100000,0 150000,0 200000,0 250000,0 2014-01-01 2014-02-01 2014-03-01 2014-04-01 2014-05-01 2014-06-01 2014-07-01 2014-08-01 2014-09-01 2014-10-01 2014-11-01 2014-12-01 Sp ot pri ce (SEK /M W h) Pr od uc tio n (M W ) Date

(21)

18

Wind speed has the strongest correlation with wind power for price areas SE2 and SE3. The reason is that there were more wind power plants in SE2 and SE3 compared to SE1 and SE4 during 2014. Also, the wind speed in SE4 has a strong positive correlation with solar power. The explanation is that SE4 is the price area with most developed solar power infrastructure. Regarding precipitation, the correlations are too weak to draw any conclusions. Hydropower can conceivably be linked to precipitation since hydropower stores precipitation in reservoirs. To see this correlation more clearly a time period surpassing one year must be used because of the considerable capacity for storage in reservoirs between seasons.

Production, consumption and spot price 2014

Production and consumption have a strong positive correlation. Therefore, it is reasonable to investigate correlations with each type of power source. It can be noted that both hydropower and thermal power have a strong positive correlation with consumption and production. One explanation for this is that hydropower is the regulation power source in the energy system and therefore matches the demands of production and consumption well. This is reflected in the value factor, which has a stable value > 1 for 2014 (Table 1). Thermal power production varies seasonally to follow the needs of consumers and have, thus, strong correlations to both production and consumption. Although, not as strong as hydropower, which may explain why the value factor for thermal power is < 1 in 2014 (ibid).

Nuclear- and wind power also have a positive correlation linked to consumption and

production but not as strong as thermal- and hydropower. This is because these power sources fulfill another purpose in the electricity market, they cannot be regulated the same way. As mentioned in the previous paragraph, nuclear power has a certain continuous production share most part of the year, except for a short time in the summer. The correlation for wind power can be explained by the simple reason that it is an intermittent source. It is not always producing when power is needed.

Solar power is the sole power source in 2014 that has a negative correlation with consumption and production, even though it is weak. This may be because solar power contributes most to the electricity grid during the summer which coincides with the time of year when

(22)

19

Figure 7. Correlation between consumption, production and spot price during 2014. Spot prices for SE1-3 are hidden under the curve for SE4.

The correlation between both solar- and hydropower relative each spot price area are positive, while it is negative for the rest of the power sources. This can be linked to the value factor and explains why it for solar- and hydropower is > 1 and for the remaining < 1, see Table 1. Spot prices and daily consumption increases in the morning when the population wakes up, solar systems start to produce, and hydropower can quickly be regulated to conform to the demand. Therefore, according to the value factors for thermal-, wind- and nuclear power it can be determined that these power sources do not precisely match the consumption continuously. For hydro- and solar power, the opposite applies.During 2014, solar- and hydropower had the largest market value whilst thermal-, wind- and nuclear power had the lowest value. This corresponds to the presented value factors for the power sources.

In Table 1, the value factors for all power sources in price areas SE1 and SE2 are the same. However, the value factors for SE3 and SE4 differ. This is linked to that electricity prices in SE1 and SE2 are in principal identical and are lower compared to SE3 and SE4. SE4 usually has the highest price of the four price areas. This is partly because Swedish electricity generation and consumption is not evenly distributed in the country currently, nor was it 2014. This means that much of the electricity produced is transported from north to south via transmission lines. This is costly, i.e., electricity prices are higher, which result in higher value factors in SE3 and SE4.

4.2 Analysis of value factors and correlations for 2014-2019

In this subsection notable value factors, distinguishable connections between power sources and outstanding parameters over the time period 2014-2019 will be discussed in depth. The section is divided into discontinuous- and continuous value factors. Discontinuous value

0,0 50,0 100,0 150,0 200,0 250,0 300,0 350,0 400,0 450,0 500,0 0,0 100000,0 200000,0 300000,0 400000,0 500000,0 600000,0 14- 01-01 14- 02-01 14- 03-01 14- 04-01 14- 05-01 14- 06-01 14- 07-01 14- 08-01 14- 09-01 14- 10-01 14- 11-01 14- 12-01 Sp ot pri ce (SEK /M W h) Pr od uc tio n (M W ) Date

(23)

20

factors imply value factors that vary over the time span whilst continuous value factors refer to value factors that have remained stable. Firstly, the value factors- and Pearson correlation coefficient tables for 2015-2019 are presented followed by a general discussion of the results over the time period (often in comparison with the discussion regarding the results of 2014). For a general overview of average consumption, production, spot price, temperature, wind speed and precipitation for 2014-2019, see Appendix A.

SE1 1.001 0.995 1.116 0.962 0.919

SE2 1.002 0.995 1.116 0.962 0.919

SE3 1.008 0.990 1.120 0.948 0.929

SE4 1.014 0.984 1.108 0.940 0.956

Average 1.006 0.991 1.115 0.953 0.931

(24)

21

SE1 1.024 0.982 1.023 0.969 1.053

SE2 1.024 0.982 1.023 0.969 1.053

SE3 1.025 0.980 1.028 0.967 1.044

SE4 1.027 0.979 1.029 0.965 1.039

Average 1.025 0.981 1.026 0.968 1.047

(25)

22

SE1 1.037 0.979 0.997 0.939 1.051

SE2 1.037 0.979 0.997 0.939 1.051

SE3 1.041 0.975 0.996 0.936 1.055

SE4 1.044 0.973 0.998 0.936 1.042

Average 1.040 0.976 0.997 0.937 1.050

(26)

23

SE1 1.016 0.997 0.981 0.962 1.072

SE2 1.016 0.997 0.981 0.962 1.072

SE3 1.018 0.996 0.982 0.960 1.075

SE4 1.023 0.993 0.977 0.956 1.069

Average 1.018 0.995 0.980 0.957 1.072

(27)

24

SE1 1.027 0.984 1.052 0.947 0.974

SE2 1.027 0.984 1.052 0.947 0.974

SE3 1.029 0.983 1.052 0.945 0.969

SE4 1.034 0.982 1.090 0.940 0.972

Average 1.029 0.983 1.061 0.945 0.972

(28)

25 Discontinuous value factors

This section discusses the variations of value factors for thermal- and solar power. As shown in Figure 7, these value factors are discontinuous over the time span. The development of the value factor for these power sources is opposite to each other. An explanation for this is that they both depend on temperature, but in opposite ways.

Figure 7. Value factor development 2014-2019 for each power source.

Temperature and thermal power have continuous strong negative correlations throughout the time period which is reasonable and not unexpected since thermal power needs cold

temperatures to fulfill its purpose and this occur every year (Table 2, 4, 6, 8, 10, 12).

However, the average value factors for thermal power is > 1 in 2015, 2016 and 2019, in other words the years with characteristic temperature difference between seasons in Sweden and normal average precipitation (Table 3, 5, 11). While the other years resulted in value factors < 1 because of the high temperatures, extended summers and less average precipitation (Table 7, 9 and Appendix A), which means that there is not as much demand for the heat from thermal power.

Solar power has strong positive correlations with temperature and wind speed throughout the time period (Table 2, 4, 6, 8, 10, 12). The two correlations that seems to differ compared with the other years is production and consumption. In 2014 the correlation between

production/consumption and solar power was –0.29 and close to zero the other years (see Table 2). Therefore, the varying value factor can not only be explained by the solar power’s

0,000 0,200 0,400 0,600 0,800 1,000 1,200 2014 2015 2016 2017 2018 2019 Val ue fac tor Year

(29)

26

strong correlation with temperature but also the increased production shares across the period. This is the result of the 781 % expansion of solar power that happened from 2014 to 2019 (Figure 8). For instance, the high value factors of 2014 and 2018 are similar, yet the

production share is higher 2018. The spot prices were higher in 2018 which also contributed to a high value factor but shows no correlation with solar power (Figure 9 and Table 10). The wind speed’s correlation in price area SE4 is strongly positive as described in 4.1.1.

Figure 8. Solar power average hourly generated production 2014-2019.

Figure 9. Hourly average spot price per year during 2014-2019. 0 50 100 150 200 250 00: 00 05: 00 10: 00 15: 00 20: 00 01: 00 06: 00 11: 00 16: 00 21: 00 02: 00 07: 00 12: 00 17: 00 22: 00 03: 00 08: 00 13: 00 18: 00 23: 00 04: 00 09: 00 14: 00 19: 00 00: 00 05: 00 10: 00 15: 00 20: 00 2014 2015 2016 2017 2018 2019 Pr od uc tio n (M W ) 0 100 200 300 400 500 600 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 00: 00 04: 00 08: 00 12: 00 16: 00 20: 00 2014 2015 2016 2017 2018 2019 Sp ot pri ce (SEK /M W h)

(30)

27 Continuous value factors

Nuclear-, wind- and hydropower exhibit stable value factors for the time span in question (Table 1, 3, 5, 7, 9, 11). As shown in Figure 7, similarities in value factors can be seen between the years for wind-, nuclear- and hydropower.

Wind power has unsurprisingly the strongest positive correlation with wind speed. The correlation has increased in SE1 and SE2 over the years, therefore also the wind power production share has also increased. This is because there has been an expansion of wind power plants in these price areas. During time period, the correlation between wind power and wind speed in SE2 and SE3 has been relatively constant (Table 2, 4, 6, 8, 10 and 12).

There is a variation in correlation between wind power and the spot price over the years. The correlations are consistently negative. This is because wind power produces the most during winter when the spot prices are generally lower compared with the prices in the summer. The correlation is strongest 2014 and 2017, around –0.33, and almost non-existent during 2015 and 2016 (Table 2, 3, 4, 5, 6, 7, 8, 9). The reason can be the cold temperatures and that hydropower can regulate production which affects spot prices.

The development of value factors for wind power is stable, around 0.95. Several factors can explain this trend. One of them is that the average wind speed in Sweden is similar over the years (Appendix A). The reason why the value factor is < 1, is because it is windiest during nighttime hours when a smaller amount of electricity is consumed and since it cannot be regulated, i.e., it does not match the demand. Due to the characteristics of hydropower its value factor has remained stable over the time span (Table 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12). The development of the average value factor for nuclear power has been relatively stable over the last six years, only differing slightly between the highest recorded value in 2018 (Table 10) and the lowest value in 2017 (Table 8). Nuclear power exhibits strong negative

correlations to temperature and solar power further concluding the effect of nuclear power maintenance closure each year during the warmer summer months. Therefore, the impact on the value factor is more closely connected to the maintenance shutdown than the temperature per se. Strong positive correlations can be traced to thermal power since both power sources meet their share of maximum production during winter (Table 2, 4, 6, 8, 10, 12).

Since nuclear power provides the baseload in the Swedish energy system its value factor will consistently be < 1, since if nuclear power decrease, spot prices will increase. This is

(31)

28

hydropower had prerequisites for continual production in 2015 and therefore acted as a greater influence on the average spot price than nuclear power. In 2018 on the other hand average precipitation was lower and nuclear power had the upper hand in spot price regulation, resulting in a negative correlation (Table 10).

Table 13. Sum of production, consumption and precipitation with average spot price, temperature and wind speed for 2015 and 2018.

Prod (TWh) Cons (TWh) Spot price (SEK/MWh) Temp (°C) Precipitation (mm) Value factor (Nuclear power)

Spot price and nuclear power correlation

2015 154 131 204 5.7 2726 0.991 0.32

2018 153 135 461 5.5 1935 0.995 –0.28

4.3 Sensitivity analysis

The purpose of this sensitivity analysis was to compare the resulting value factors when using generated solar power production data versus reported to Svenska Kraftnät. In order to do this the value factors are calculated exclusively using solar power production data and total

production data from Svenska Kraftnät.

(32)

29

Table 14. Difference in solar power value factors for collected data from Svenska Kraftnät versus generated data.

2014 2015 2016 2017 2018 2019 SE1 0.006 0.001 0.015 -0.007 0.013 -0.017 SE2 0.006 0.002 0.015 -0.007 0.013 -0.017 SE3 0.005 0.021 0.014 -0.011 0.014 -0.018 SE4 0.005 0.032 0.013 -0.011 0.011 -0.017 Average 0.006 0.014 0.014 -0.009 0.013 -0.018

This comparison is of interest, since the solar power data reported to Svenska Kraftnät is not representative of the total power contribution from solar power, i.e., in the reported data, only the solar power fed back to the grid, is included. This is noticeable, for example when

comparing daily hourly averages weighted against maximum production. As seen in Figure 10, the Svenska Kraftnät solar power data differ slightly from the generated, since it only shows the percentage of solar power sent to the grid whilst the generated solar power data also include the self-consumed power.

Figure 10. Weighted daily averages of generated and collected solar power data in 2014.

In Figure 11 generated solar power data for 2019 are compared against collected solar power data from Svenska Kraftnät. In comparison with Figure 10, an increase in prosumer usage of solar power can be distinguished whilst collected data from Svenska Kraftnät has remained similar. This shows the increase of installed solar power used by prosumers and more correctly reflects the current energy system. This could also indicate that a larger part of production is reported to Svenska Kraftnät in 2019, compared to 2014, as a consequence of more installed solar power.

0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0% 00: 00: 00 01: 00: 00 02: 00: 00 03: 00: 00 04: 00: 00 05: 00: 00 06: 00: 00 07: 00: 00 08: 00: 00 09: 00: 00 10: 00: 00 11: 00: 00 12: 00: 00 13: 00: 00 14: 00: 00 15: 00: 00 16: 00: 00 17: 00: 00 18: 00: 00 19: 00: 00 20: 00: 00 21: 00: 00 22: 00: 00 23: 00: 00 Pr od uc tio n sh ar e (% ) Time (h)

(33)

30

Figure 11. Weighted daily averages of generated and collected solar power data in 2019.

In summary, the choice of solar power data does have an influence on the resulting value factors as shown in Table 14. Although, it is clear that the generated solar power data

provided an opportunity to investigate the development of solar power production in Sweden, whilst Svenska Kraftnät solar power data remained essentially unchanged from 2014-2019 providing an inaccurate depiction of the current energy system.

4.4 Future developments

The future development of the Swedish energy system is uncertain. Nuclear power is on its way to be decommissioned and intermittent power sources are about to increase their production shares significantly. Since nuclear power has a significant role in today’s energy system the future value factors for other power sources could come to change.

Due to the expansion of wind power in recent years, an assumption is that the wind power production share will continue to increase. The development of wind power is yet dependent on how the future energy- and climate politics in Sweden will evolve. It can be argued that the value factors for wind power could change since the values could be weaker for deviations with wind power expansion, yet the production share would increase which also would affect the future value factors.

Solar power is another power source expanding in Sweden yet the production share is lower comparing to other similar countries. Germany had a solar power production share of 7% 2015 (Energimyndigheten, 2016) and 9.1% in 2019 (Fraunhofer ISE et al., 2020). Sweden had in comparison 0.1% in 2015 and 0.4% in 2019. The main reasons why Sweden do not have such high solar penetration is that Germany have had an early commitment to solar power with a stable financial support system and interests in inventors (Energimyndigheten, 2016). The value factors for solar power could come to change since the production share is

increasing, due to removal of other power sources, however, to what extent depends on political aspects, i.e., financial support and interests.

0,0% 10,0% 20,0% 30,0% 40,0% 50,0% 00: 00: 00 01: 00: 00 02: 00: 00 03: 00: 00 04: 00: 00 05: 00: 00 06: 00: 00 07: 00: 00 08: 00: 00 09: 00: 00 10: 00: 00 11: 00: 00 12: 00: 00 13: 00: 00 14: 00: 00 15: 00: 00 16: 00: 00 17: 00: 00 18: 00: 00 19: 00: 00 20: 00: 00 21: 00: 00 22: 00: 00 23: 00: 00 Pr od uc tio n sh ar e (% ) Time (h)

(34)

31

Both Denmark and Germany have support systems for both solar- and wind power that are adjusted continuously and have therefore affected the development of the specific power sources (Riksrevisionen, 2017).

A greater market for gas turbines as regulating power is expected to complement the increasing production shares for wind- and solar power. This will lead to requirements for faster thermal power start-ups (Nohlgren et al., 2014). Due to this, an assumption is that the value factor for thermal power could come to change.

(35)

32

5. Conclusions

In conclusion, the study revealed the complex relationships between power sources, production, consumption and spot prices in the Swedish power system. There are no major changes in value factors for nuclear-, wind- and hydropower over the time period, only small discernable differences. The value factors for thermal- and solar power have varied between < 1 and > 1.

It is possible to conclude that certain factors have greater impact than others on the outcome of the value factor. A number of general patterns have been found. For instance, temperature has a greater impact on the value factors compared to wind speed and precipitation. There is also a clear link between the functionality of the power source and how it matches the needs of the consumers. A better match between production and demand gives a favorable value factor. This is why, for example, hydropower has a stable value factor of around 1 during the time period and wind power has a less stable value factor of < 1. Nuclear- and hydropower have fundamental roles in the Swedish energy system, which affects all the other power sources’ value factors.

(36)

33

References

Centrum för Vindbruk, H. på G. (2013). Var och hur blåser det ?

Energiforsk. (n.d.). Ny forskning visar på stora utsläpp och höga kostnader om svensk

kärnkraft ersätts. Energiforsk.

https://energiforsk.se/program/karnkraft-omvarld-och-teknik/nyheter/karnkraftens-roll-i-energisystemet/alternativ-till-karnkraft/ Energimarknadsbyrå, K. (2020). Elområden. Konsumenternas Energimarknadsbyrå.

https://www.energimarknadsbyran.se/el/elmarknaden/elomraden/ Energimyndigheten. (2016). Solceller i omvärlden.

https://www.energimyndigheten.se/globalassets/fornybart/solenergi/solen-i-samhallet/solceller-i-omvarlden.pdf

Energimyndigheten. (2020). Statistikdatabas. Statistikdatabas.

http://pxexternal.energimyndigheten.se/pxweb/sv/Vindkraftsstatistik/Vindkraftsstatistik/ EN0105_5.px/

Fraunhofer ISE, Burger, B., & Schneider, K. (2020). Public Net Electricity Generation in Germany 2019. Public Net Electricity Generation in Germany 2019: Share from

Renewables Exceeds Fossil Fuels, 1–4. www.ise.fraunhofer.de

Hirth, L. (2013). The market value of variable renewables. The effect of solar wind power variability on their relative price. Energy Economics, 38(2013), 218–236.

https://doi.org/10.1016/j.eneco.2013.02.004

Hirth, L. (2015). Market value of solar power: Is photovoltaics costcompetitive? IET

Renewable Power Generation, 9(1), 37–45. https://doi.org/10.1049/iet-rpg.2014.0101

Jämtkraft. (n.d.). Hur fungerar vindkraft? Jämtkraft. https://www.jamtkraft.se/om-jamtkraft/var-fornybara-produktion/vindkraft/hur-fungerar-vindkraft/

Lindholm, K. (2020). Kärnkraft. Energiföretagen.

https://www.energiforetagen.se/energifakta/elsystemet/produktion/karnkraft/

Lingfors, D., Åberg, M., & Widén, J. (2019). Effekt- och elprisscenarier vid hög andel solel i

det svenska elsystemet. 1–10.

Lublin, Zofia and Friberg, D. (2016). Uppföljning av utvecklingen för investeringar i

solenergi.

https://www.energimyndigheten.se/contentassets/998e44aa242f4ed0b60f3d60da251269/ uppfoljning-av-utvecklingen-for-investeringar-i-solenergi-2016-er2016_31.pdf

Naturskyddsföreningen. (2019). Ny lagstiftning för vattenkraften är här!

Nohlgren, I., Svard, S. H., Jansson, M., & Rodin, J. (2014). Electricity from new and future

facilities. (in Swedish) El från nya och framtida anläggningar. ELFORSK. 91.

http://www.elforsk.se/Programomraden/El--Varme/Rapporter/?rid=11_26_ Nord Pool. (n.d.-a). Day-ahead market.

https://www.nordpoolgroup.com/the-power-market/Day-ahead-market/

Nord Pool. (n.d.-b). Intra day market. https://www.nordpoolgroup.com/the-power-market/Intraday-market/

(37)

34

Nordling, A., & IVA. (2016). Sveriges framtida elnät - En delrapport. www.iva.se NordPool. (2020). Historical Market Data.

https://www.nordpoolgroup.com/historical-market-data/

Riksrevisionen. (2017). Det samlade stödet till solel. In Riksrevisionen.

https://www.riksrevisionen.se/download/18.78ae827d1605526e94b2db34/15184354963 56/RiR_2017_29_SOLEL_ANPASSAD.pdf

SMHI. (2018). Torka.

SPSS TUTORIALS: PEARSON CORRELATION. (2020). Kent State University.

Statistiska Centralbyrån. (2019). Elektricitet i Sverige. https://www.scb.se/hitta-

statistik/sverige-i-siffror/miljo/elektricitet-i-sverige/?fbclid=IwAR0WBu6WrJVFMnaV95ErPeAZTHBIvQb_dFuZok1Tm0BQ7FlC 4XlI6NgVA4I

Svenska kraftnät. (2019). Elstatistik.

Svenska Kraftnät. (2016). Reliable electricity supply.

Svenska Kraftnät. (2020). The control room. https://www.svk.se/en/national-grid/the-control-room/

SVT. (2018). Torkan minskade vattenkraften.

Söder, L. (2014). På väg mot en elförsörjning baserad på enbart förnybar el i Sverige. https://www.diva-portal.org/smash/get/diva2:727697/FULLTEXT01.pdf

Uniper. (n.d.). Reglerkraft - El i rätt tid och mängd.

https://www.uniper.energy/sverige/vattenkraft/reglerkraften Utellus. (n.d.). Det Nordiska klimatet utmärkt för solceller.

https://www.utellus.se/2019/09/det-nordiska-klimatet-utmarkt-for-solceller/

(38)

35

Appendix A

Average value for spot price, temperature and wind speed. Average value for production, consumption and precipitation. Year 2014-2019

The Value of Value Factors

Juni 2020

The Value of Value Factors

Time-Dependent Development of Value Factors

on the Swedish Electricity Market

The Value of Value Factors

Preface

Table of contents

Important concepts

1. Introduction

1.1 Aim of the report

1.2 Thesis questions

1.3 Delimitations

1.4 Outline

2. Background

2.1 Swedish power sources

2.2 Developments of Swedish power sources

2.3 Electricity price areas

2.4 Electricity market

2.5 Weather impact on electricity production

2.6 Value factor

2.7 Previous work

3. Methodology

3.1 Model description

3.2 Data

4. Results and discussion

4.1 Analysis of value factors and correlations for 2014

4.2 Analysis of value factors and correlations for 2014-2019

4.3 Sensitivity analysis

4.4 Future developments

5. Conclusions

References

Appendix A