27-415-3-ED

Introduction

From 2001 to 2011 the number of households in Norway increased from 1.96 million to 2.21 million. Compared to other Nordic countries, such as Sweden and Finland, Norwegian households have a significantly higher electricity consumption per household (The World Bank, 2014). This could be due to a number of different factors, such as differences in energy mix and uptake of electric vehicles (EV), and energy demanding appliances in the home. Although the consumption is higher, the carbon intensity of the national electricity grid is much lower than the EU average of 329 g CO2/kWh (IEA, 2017b) at 22 g CO2/kWh (Meld. St. 25 (2015–2016), pp. 19–20). This means that potential investments in residential photovoltaic (PV) panels will do little to decarbonise the domestic sector, but could potentially benefit households financially.

Furthermore, energy prices are expected to rise by 4% each year (Bartnes and Skaare Amundsen, 2016) due to expansion projects to the European market and increased fees for grid connections. Residential photovoltaic (PV) panels and storage solutions are available technologies that could provide homeowners with a buffer against rising prices and increase energy security for households (Noll, Dawes and Rai, 2014).

This study determines under what conditions an investment in a residential PV system in combination with a storage solution would be economically beneficial for a 'typical' Norwegian household.

Background

Current investments in photovoltaic and storage systems

The number of PV systems is increasing worldwide and in 2017 the global capacity increased to 75 GWp, a 50% increase from 2016 (International Energy Agency (IEA), 2017a). On a global scale, residential PV systems are conventionally installed to feed into national electricity grids, which has typically been rewarded using Feed–in Tariffs (FiTs). However, in Norway, the location of interest in this study, PV modules in conjunction with a storage system (batteries) have traditionally been installed to power off–grid buildings such as cabins and small holiday homes, with 95% of national installed PV capacity being off–grid in 2003 (IEA, 2004). FiTs were only introduced in the beginning of 2017, with electricity providers offering a "plus customer" service, paying private customers 1 NOK/kWh (0.12 USD/kWh) of self–generated power fed to the grid (Otovo, 2018a). As of the end of 2017, Norway had a total installed capacity of 45 MWp from PV systems, an 18 MWp and 59% increase from the record year of 2016 that saw a 366% increase from the previous year (Norwegian Solar Energy Group, 2018).

Energy demand and greenhouse gas emissions from households

Energy consumption and greenhouse gas (GHG) emissions are strongly linked together, and many countries are introducing energy efficiency measures to alleviate the emissions from households (Balaras et al. 2007). There is some disagreement in literature about the total global carbon footprint from the domestic sector, however Berardi (2017) concludes that residential buildings represent 30–40% of the total energy demand and 19–40% of GHG emissions each year. The energy demand estimate holds true for the Norwegian economy where 31% of energy consumption is traced back to households (Statistics Norway (SSB), 2017a). The country is among the highest electricity (kWh) per capita consuming countries in the world with a demand of 23,000 kWh/capita/annum (The World Bank, 2014; Bøeng, 2014), yet only 2% of emissions are linked to energy consumption in buildings (SSB, 2017b). This is because renewable energy resources have been used to power Norwegian industries and homes since the 1870's. Hydropower is abundant across the country and today 96.3% of the annual demand is covered using this technology (SSB, 2017a). Consequently, the carbon footprint of electricity bought from the national grid has a very low carbon intensity at 22 g CO2/kWh (Meld. St. 25 (2015–2016), pp 19–20), far below the European average of 329 g CO2/kWh (IEA, 2017b).

Households in Norway are primarily using electricity for all activities in the house, including heating (SSB, 2014a), which in light of the current electricity generation results in the current building stock having a much smaller carbon footprint than other European countries. Some biofuel is used to support the electric heating systems for the colder parts of the year, yet residential energy consumption in 2012 consisted of 80% electricity (SSB, 2014a). Due to the flexibility in the system and the abundant access to elevated water reservoirs, the price per kWh is significantly lower in Norway than the EU average, at 0.98 NOK (0.10 €/kWh) (SSB, 2018b) and 0.21 €/kWh (Eurostat, 2016) respectively. Relative to income, this provides Norwegian residents with cheap, low–carbon energy all year round.

Norwegian national law states maximum requirements for the rate of heat flow (U–values) through all external surfaces to ensure energy efficient housing across the country and to lower energy consumption on a national scale. The technical building regulations (TEK10, 2017) were updated in 2016 according to scientific research and national targets to reduce energy consumption. Among the amendments were the banning of paraffin tanks for domestic heating to reduce emissions, and the previous requirement to incorporate connections to district heating facilities was removed to increase the flexibility of systems (TEK10, 2017). This is part of the government's commitment to reduce GHG emissions in response to UNFCCC publications, the Kyoto Protocol and the 2015 Paris Agreement. The Norwegian government is committed to reduce emissions by 40% against 1990 levels by 2030 (Utenriksdepartementet, 2016). Working towards an agreement with the European Union (EU), Norway is seeking to be a part of their target to reduce emissions by 85–90% by 2050 and becoming a low carbon economy.

The current, positive progress against the long term GHG emissions reduction target is recorded by SSB and verified by Eurostat, the European counterpart. In 2016, total emissions from Norwegian territories were 3.0% higher than 1990 levels, but saw a reduction of 1.1% from the previous year. The emissions from domestic energy consumption was 0.4 Mt CO2e, a reduction of 73% against 1990 levels and 20% lower than 2015 (SSB, 2017b). Although the emissions are going down, the domestic sector still accounts for 31% of the energy demand, in contrast to 25.3% in the rest of Europe (SSB, 2017a; Eurostat, 2017), indicating a potential to reduce energy consumption. A proposition from the Energy and Climate Committee in the Norwegian Government encourages the government to invest in domestic electricity generation, as well as other domestic energy efficiency measures. Domestic generation of renewable energy is unlikely to significantly reduce emissions directly, but will make it possible to offset emissions from other sectors by making renewable energy that's already on the grid available to fossil fuel heavy industries. Generating renewable energy on residential buildings could also encourage and enable the ongoing transition from diesel/petrol vehicles to electric vehicles (EV) to be sustainable and not increase the demand on the national electricity grid.

Definitions

Residential photovoltaic (PV)

There are many definitions of the term "residential PV" in existing literature. The definition used in this study is a slight amendment to the one used by the International Renewable Energy Agency (IRENA) (2012); "residential PV systems typically do not exceed 20 kW and are usually roof–mounted". As the Norwegian government, through ENOVA, only supports systems up to 15 kWp, this will be used as the max capacity in this report. Contrary to other definitions (Goodrich, James and Woodhouse, 2012; Rickerson et al. 2014), this specifies not only the maximum installed capacity, but also the way in which it is mounted to the building.

The 'typical' Norwegian house and household

A typical residential household in Norway is a single occupancy, detached house, with a total living area (TLA) ranging from 123 m2 to 322 m2 depending on when the household was built (EPISCOPE, 2016). This study considers seven building age groups, reflecting the groupings of published statistical data on building specifications. All appliances and systems in the house are powered using electricity from the national grid, including heating of living spaces and hot water (SSB, 2014a). Supporting electric heating through space heaters, underfloor heating and air source heat pumps (ASHPs), solid fuels (biomass) are burned in the colder months to provide extra heat (16% of total heating demand).

For easy comparison of results, all house ages considered in this study will be assumed to have the same layout of rooms, scaled up or down to fit the average TLA. The living spaces distributed over two floors have standard ceiling heights (Standard Norge, 2012) and a cold roof with an insulated loft. 20% of the external surfaces are glazed, distributed over the east, south and west facing walls. The typical house has a standard gable roof, with one sloped surface oriented directly to the south. Measurements give a slope of 35 (see Section 2.3), which is slightly steeper than the "typical" slope of 27–33 (University of Bergen, nd). The U–values of all external surfaces are defined by TABULA (2016) and vary depending on the construction time period.

On average, there are 2.3 residents in the house (SSB, 2017c), of which two are adults and are either in work or retired.

Greening economy

The term "Green Economy" was first introduced in a series of reports for the UK government by a group of environmental economists, the first named "Blueprint for a green economy" (Pearce, Markandya and Barbier, 1989). The reports make a case for using "sustainable development" as a measurement for economic progress and the successfulness of policies and projects. There is no internationally agreed definition to the term, with at least 8 definitions identified by the United Nations (UN) (UN Sustainable Development, nd). United Nations Environment Programme (UNEP) has defined a green economy as "one that results in improved human well–being and social equity, while significantly reducing environmental risks and ecological scarcities. It is low carbon, resource efficient, and socially inclusive" (UNEP, 2011).

As Norway cannot yet be defined by this definition, but is committed to working towards it (Utenriksdepartementet, 2016), this study uses the term "greening economy". Making good progress in the energy and building sector (Nykamp, 2017), with renewable energy covering most of the electricity demand and the construction sector transitioning to building "green buildings" (United States Environmental Protection Agency, 2016), Norway is in this paper defined as a "greening economy" – an economy that is making progress towards becoming a Green Economy and is actively working towards sustainable development.

Methodology

The approach used in this study is similar to that of Arteconi et al. (2017), which simulates the energy demand of an industrial building and estimates the potential benefits of utilising solar irradiation and storage to heat the building using TRNSYS. This provides a clear overview of the behaviour of the system which is essential when determining profitability. The results are then used to calculate the payback periods of the two systems of interest; PV–storage and PV–only.

By defining a "typical" Norwegian house and household using published statistical data, it is possible to establish a baseline for energy demand in households constructed in seven time periods (<1954, 1955–1970, 1971–1980, 1981–1990, 1991–2000, 2001–2010, and 2011<). Four different locations are considered using weather data from Oslo, Bergen, Trondheim and Narvik (Figure 2). Using TRNSYS, a dynamic simulation software, 28 simulations are run to estimate electricity generation potential for each building in each location. Results from the simulations are then used to determine payback periods for two PV systems; a PV system with storage capacity (PV–storage) and a PV system without storage capacity (PV–only).

Technical description of the system

The systems used in this study is a single family house and has a rooftop PV system connected to a battery installed in the house (Figure 1). The PV panels consist of monocrystalline silicone cells and have a manufacturer declared efficiency of 23% (SunPower, 2017); 3% lower than the most recent PV technologies to account for the time it takes for technologies to become cost efficient on the market (Green et al. 2017). The installed PV capacity is 3 kWp, which covers a roof area of 20 m2. The study compares one storage option, a lithium–ion battery such as the Tesla Powerwall (a battery with 5 kWh capacity is used for financial calculations), to feeding excess generated electricity to the national grid. Using market prices, current cost/kWh for electricity, and system feed–in tariffs, payback times are estimated which form the foundation for further discussion and conclusions. The total electricity generated, , is

where is the total power output (kW), is time (hr) and is the efficiency of the PV system.

Figure 1. System layout of an AC coupled residential PV/battery system as used in this study. Source: Weniger, Tjaden and Quaschning (2014).

Norway is split into seven different climate zones, defined by Tokle, Tønnesen and Enlid (1999). This study considers four of the most populous, focussing on a major city within each of them (Figure 2):

  1. Oslo: located in the South–East with an inland climate (zone 1).
  2. Bergen: located in the West with a coastal climate (zone 2).
  3. Trondheim: located in the middle of the country, with a coastal climate (zone 4).
  4. Narvik: located in the North, with an arctic coastal climate (zone 6).

Input data

To simulate the behaviour of a PV system and energy demand in a house, a weather file containing hourly weather data is required. For this research four different weather files, sourced from WhiteBoxTechnologies (nd), are used to account for how differences in outdoor temperatures impact the energy demand in residential buildings. They provide data on the reflectance of the environment (albedo) from cloud cover, solar irradiation (direct and indirect) and temperature, which is subsequently incorporated in the TRNSYS model and reflected in the results.

Figure 2. Norway's seven climate zones, defined by SINTEF (adapted from Tokle, Tønnesen and Enlid, 1999), and the four cities considered in this study.

The energy efficiency of residential buildings have increased with time, in line with updated regulations and to meet national energy targets (Forskrift om tekniske krav til byggverk (Byggteknisk forskrift), 2017; Utenriksdepartementet, 2016). In order to incorporate the difference in building standards and thermal performance into the analysis, the typical house, as defined in Section 1.4, with specifications from seven different age groups is modelled using a dynamic simulation software, TRNSYS. All data about building size, the thermal performance of all surfaces (U–values) and heating systems for each building age group is gathered and published by the Norwegian University of Science and Technology (NTNU) as part of the EU funded project EPISCOPE (TABULA, 2016). All U–values used include the cold bridges from lintels, junctions and wall ties, as this is the standard for reporting thermal performance in Norway. Together, this data forms the foundation of the TRNSYS models and an annual energy demand (kWh) for the houses modelled is estimated in addition to a simulated output from the installed 3 kWp PV system.

Retrofits including microgeneration technologies are most commonly undertaken in combination with improvements to thermal performance such as installing additional insulation, a cost efficient way to reduce energy demand (Sunikka–Blank et al. 2012). It is therefore assumed that a household investing in PV panels and storage solutions has done this already. Just under 50% of the Norwegian building stock had been renovated by 2012 where thermal performance was improved or more efficient heating systems such as air source heat pumps (ASHP) were installed (ENOVA, 2012). EPISCOPE outlines three states of renovation for each building; original state, improved state and ambitious state. Based on the assumptions outlined, the "improved state" data is used.

To incorporate typical fluctuations in electricity consumption caused by switching on and off electrical loads in the household, an occupancy profile working at the same temporal scale as the weather file is required. The hourly profile described by Nord et al. (2018) (Figure 3) for a Norwegian house is therefore used as the schedule basis for load consumption and heating patterns.

2.3 Simulation data and payback periods

It was assumed that the PV panels are fitted on the south facing roof surface at an azimuth angle of 180, with 3 kWp installed capacity as estimated in the EPISCOPE project. An average slope of 35 for the roof was found measuring 400 houses; 100 in each location. To determine the in–plane irradiance on the PV cells, the irradiance measured on the horizontal plane was converted to the plane of the PV array. This

Figure 3. Occupancy patterns used in energy demand model. Source: Nord et al. 2018.

was done in the TRNSYS simulation for direct and diffuse irradiance. The albedo was incorporated as a function in the weather file.

Residential storage solutions were not explored in the simulations, but the potential for profitability of a 5 kWh battery connected to the PV system was considered against a grid connected system using a payback period method (Bhandari, 2015). Assuming electricity prices increase by 4% each year, the payback period, , is found using Equation 2:

where is the payback period and is annual saved cost for year 1 (). Full calculations can be found in Appendix 1.

Data sources

All building statistics on household size, occupants' employment status, measured energy consumption (by fuel), and age of buildings was gathered using SSB (2018b; 2013; 2014a, 2018c) and information on building fabrics and energy efficiency parameters by building year was extracted from the EU funded EPISCOPE project through the TABULA Webtool (TABULA, 2016).

Results

Performance of the simulated houses and installed PV panels

Table 1 shows the results for simulated energy demand in the typical households modelled across the four locations. From Table 1 it can be observed that the energy demand varies from 96 to 171 kWh/m2, depending on the volume of the house, thermal performance (as a function of age of building) and location. The energy demand is highest during the winter months and less energy is used for heating from April to the second week in October. Buildings built before 1970 typically have a higher overall consumption, although more recent buildings are more efficient per m2 (with an exception from those built in the 1990s).

The simulated output from the installed PV panels is described in Table 2 as total generation (kWh) and percentage of demand for the respective households. When comparing the simulated electricity generated in the four locations, Oslo and Trondheim sees a higher annual output than Bergen and Narvik, with outputs ranging between 5 and 21% of annual energy demand. The slope (=35), TLA and type of installed PV technology remains the same for all locations and building periods and the output is therefore only a function of climate data. All local limiting factors, such as shading from vegetation or built structures, are assumed to be negligible for the sake of producing comparable data.

All the buildings modelled are located within a latitude range of 58.8N <  < 68.7N, with Narvik, the northernmost location, being above the Arctic circle. Consequently, it experiences midnight sun in the middle of summer (June–August) and the opposite phenomenon, polar nights, in the middle of winter (November–January). This is clearly illustrated by the outputs from the simulated PV output (Figure 4.d).

Table 1. Simulated energy demand from a typical Norwegian household in four major Norwegian cities, constructed to building standards from seven different time periods. Energy demand is given as total (kWh) and as demand per living area (kWh/m2). Blank cells are due to errors in the simulation.

Oslo (58.9N)

Bergen (60.4N)

Trondheim (63.4N)

Narvik (68.4N)

kWh

kWh/m2

kWh

kWh/m2

kWh

kWh/m2

kWh

kWh/m2

Pre 1954

37508

148

39892

157

45892

181

1955–1970

36088

158

35405

155

37770

166

43457

191

1971–1980

18894

124

18708

123

19767

130

23325

153

1981–1990

15306

124

15034

122

15941

130

18857

153

1991–2000

25790

162

25747

162

27113

171

31710

199

2001–2010

34422

107

33940

105

36197

112

41360

128

Post 2011

19831

108

19712

107

20898

114

24352

132

Table 2. Simulated output from a 3 kWp PV system in four major Norwegian cities expressed as real performance per kWp, total generation (kWh), and total generation as a percentage of total energy demand from households build in seven different time periods.

Oslo (58.9N)

Bergen (60.4N)

Trondheim (63.4N)

Narvik (68.4N)

kWh/kWp/annum

1070

736

914

760

Total (kWh)

3211

2207

2742

2281

Share of demand

Pre 1954

6%

7%

5%

1955–1970

9%

6%

7%

5%

1971–1980

17%

12%

14%

10%

1981–1990

21%

15%

17%

12%

1991–2000

12%

9%

10%

7%

2001–2010

9%

7%

8%

6%

Post 2011

16%

11%

13%

9%

Annual irradiation patterns

The four locations included in the study have different irradiation levels throughout the year (Figure 4.a–d). Oslo (Figure 4.a), being the southernmost location, experiences the highest annual generation at 3,211 kWh, and exceeds the generation in Bergen (Figure 4.b), Trondheim (Figure 4.c) and Narvik (Figure 4.d) by 145%, 117% and 141% respectively. As a direct consequence of location, Oslo experiences higher levels of irradiation and total generated electricity throughout the year than the other cities considered in this project.

Figure 4.a–d. Total irradiation throughout a year in four Norwegian cities modelled using TRNSYS; Oslo (a), Bergen (b), Trondheim (c) and Narvik (d).

Payback periods

Assuming no maintenance costs and a life expectancy of 30 years (Otovo, 2018a) the initial cost will be the only expense related to the PV system. Figure 5 shows the payback period of different shares of export to grid versus self–consumption in Trondheim, the location with a mid–range generation output. Four scenarios, two with and two without an installed battery storage solution, are explored where y=0 illustrates total system payback time in years.

Figure 5. Payback periods of four scenarios of residential PV systems in Trondheim, Norway. With (B.) and without (A.) a battery energy storage system for different export to grid/self–consumption (%export / %consumption).

The four scenarios analysed in Figure 5 consider the best case and worst case scenarios for the two systems included in this study; PV only and PV with storage. The two storage scenarios assume that the battery has the capacity to store excess daily generated electricity, which the 5 kWh battery assumed used is capable of. FiTs are assumed to remain at 1 NOK for the lifetime of the PV panels, and not to be affected by the rising energy prices. Scenario A.50/50, with 50% export to grid and 50% self–consumption, has the shortest payback period of 14 years, and assumes that 50% of all electricity generated is consumed in the households without the use of a storage solution. Even though this is the best case scenario without storage, the worst case scenario, A.100/0, also has a shorter payback period than both the alternatives with storage.

Due to diurnal variation in resource accessibility, a more realistic period of self–consumption without the use of storage would be closer to 30% (Herrando, Markides and Hellgardt, 2014). From Table 3 it can be observed that PV–only systems in all locations have shorter payback periods than PV–storage for all four reference locations considered.

Table 3. Payback periods (years) for PV–only systems and PV and storage combined systems in four major Norwegian cities with current incentive schemes and estimated increase in electricity prices.

 (years)

Oslo

Bergen

Trondheim

Narvik

PV–only

12

17

14

17

PV–storage

19

25

22

25

Discussion

When looking at the same reference buildings across the different locations, the results for energy demand clearly show a higher consumption of electricity in areas of high latitude (Narvik). This is caused by the colder climate and a greater need for additional heating throughout the year. Locations in the south that experience warmer temperatures are naturally less dependent on additional heating, which is reflected by the lower estimated demand (Table 1). However, the estimated total PV outputs (Figure 4) shows that latitude is not the only significant factor in determining outputs; although Bergen is located further south than both Trondheim and Narvik, it experiences a lower output. Bergen is a coastal city in a very rain–prone area due to its micro–geographic placement. It consequently experiences significantly more rain than the other locations considered (Dannevig and Harstveit, 2013). Increased rain and cloud cover increases the albedo and reduces the irradiance level. Compared to Narvik, which is located above the arctic circle and experiences no beam radiation for two months (Figure 4), the installed PV system in Bergen is estimated to have a lower total generation despite not being affected by polar solar conditions. From Equation 1, the factor that incorporates this is , the system efficiency, which includes all weather factors and technological limitations. Although there are many contributing elements that can impact the efficiency, temperature is likely to be the most significant (Meral and Dinçer, 2011); PV panels are more efficient in colder conditions and the relatively high generation in Narvik, despite limited irradiation during winter, confirms this.

Overall, the results correspond roughly with the reported annual consumption average of

20,230 kWh (SSB, 2014a) in Norwegian households. The simulated total energy demands are overall slightly higher, but when comparing estimated demand per area (Figure 6) the simulated results are slightly lower than those from published statistics (SSB, 2014b). The discrepancies between the simulated and statistically reported energy demand could indicate that the households used for the model have a larger TLA than the dwellings contributing to the statistics. This will be discussed further in Section 4.1.

Energy balance in residential buildings in Norway

As a first assessment, it is common to use building typologies to establish an estimated energy balance for buildings (Ballarini, Corgnati and Corrado, 2014), which is the approach this study has taken. However, in order to make the typology representative for a larger number of buildings it is necessary to compare the results to detailed building stock data. SSB has published data on the number and age of residential buildings, detailing TLAs (SSB, 2018b) and geographic region. The statistics do not go into detail about building fabrics, which are needed to produce a representative TRNSYS model.

The buildings included in the EPISCOPE project are real examples of buildings considered to be representative for a whole age group. When producing the TRNSYS models all specific features in construction materials, insulation and systems as described by TABULA (2016) were assumed to be typical for all buildings in this age group.

Following the evolution of Norwegian building standards, the estimated energy demands per area (Figure 6) show a steady decrease in more recent years. This is reflecting the increasing targets for energy efficiencies in Norwegian building regulations (Forskrift om tekniske krav til byggverk (Byggteknisk forskrift), 2017, para 14–2) and observed results from a similar study in Greece (Dascalaki et al. 2011). The slight increase in demand in the 1991–2000 building is likely due to a reduction in energy efficiency targets in Norwegian legislation for those years.

Figure 6. Comparison of energy demand in households located in four Norwegian cities simulated in this study against observation based statistics. Source of grey column: Statistics Norway (2014b).

Cost efficiency assessment of an installed PV system

The study estimates the cost savings made possible by installing a PV–storage system or PV only system. Comparing these to the reference systems (typical dwellings in their current state) who use electricity from the national grid makes it possible to estimate the payback period of the investments. The total investment cost for a PV system, with or without storage, does not only include the price for PV panels, but also the cost of other components such as an inverter and wires. Efficiency is the cause of varied PV panel prices, and even though the simulated system uses PV panels of 23% efficiency, the most cost efficient monocrystalline PV panels on the market are currently 16–18% efficient (Solcellespesialisten, 2018a). This does mean that until the market efficiency reaches the simulated efficiency, the result outputs (Table 2) are more optimistic than what it is realistic to achieve. The breakdown of costs for a PV system (with and without storage) is shown in Table 4, using prices on the Norwegian market at the time of conducting this research (April, 2018). In the overview, the available financial support scheme, ENOVA (2018), is included, which supports the installation of PV panels, but not storage systems. Other incentive schemes will be discussed further in Section 4.3. As there are no available incentive schemes for storage solutions, the cost of a PV–only system is reduced the most. This could, however, change with time, as discussed in Section 4.5.

Table 4. Breakdown of investment cost for a PV–storage and PV–only system based on market prices in April, 2018.

Component

Price

Unit

Comments

Single PV module

(P=1; 295 Wp)

2,890

NOK1/module

Solcellespesialisten (2018a)

Total PV module (x 10)

28,900

NOK

2,890 NOK/module x 10 modules

Inverter

11,635

NOK

Solcellespesialisten (2018b)

Metal structure and installation costs

16,365

NOK

Estimated from price breakdown (Solcellespesialisten, 2018c)

Complete battery storage system (5kWh)

67,423

NOK

Solcellespesialisten (2018d)

Total investment cost

(PV–storage)

94,323

NOK

Total investment cost

(PV only)

56,900

NOK

PV incentives (capital subsidy)

–13,750

NOK

ENOVA (2018): 10,000 NOK for installation and 1,250 NOK/kWp installed

Final costs (PV storage)

80,573

NOK

Final costs (PV only)

43,150

NOK

1 1 NOK = 0.1225 USD (18 July 2018)

All the costs other than the PV modules and battery storage system are based on an estimate for a 3.1 kWp electrical residential PV system (Solcellespesialisten, 2018c). As the households in this study already have electric heating systems, no other costs would be associated with the installation of the PV system.

Profitability of PV systems, with and without storage

The four locations selected for this research project are located in different climate zones (Figure 2) to determine whether some locations are unfit for investments in PV systems, based on resource availability. As Table 2 describes, the two locations where the installed system has the lowest performance are the two that experience the least direct irradiation (Figure 4.b, 4.d); Bergen and Narvik. The differences in outputs will consequently impact the rate of which the system can pay for itself through consumption of self–generated electricity and export to the grid, rewarded with FiTs.

The Norwegian government launched FiTs alongside the "plus–customer" concept in January 2017 (Norwegian Water Resources and Energy Directorate (NVE), 2015) which allowed electricity providers to award customers feeding self–generated electricity to the grid. A number of energy companies have incorporated this since the start of the scheme, and the current FiT is around 1 NOK (0.123 USD) (Otovo, 2018b), depending on which provider the contract is with. For comparison, the current cost of electricity from the national grid is 0.98 (0.120 USD) NOK (SSB, 2018a) making it more profitable to export all self–generated electricity to the grid. This is likely going to change in the coming years as the cost/kWh is estimated to rise by 4% every year, due to grid expansion projects to other European countries (Section 1.1). A maturing electricity market is also likely going to affect this, making it more profitable to consume self–generated electricity within the household. Johann and Madlener (2014) discuss the profitability of PV–storage systems in German households where they estimate that up to 50% of energy could be consumed within the household without a storage system.

Increasing self–consumption is increasingly discussed in literature as more consumers are investing in PV systems, and as FiTs are one of the most common incentive schemes. The general consensus is that adding a battery storage system will increase self–consumption by 17–23% (Luthander et al. 2015). All systems will be positively impacted by FiTs, although, the initial investment support targeting PV systems, make the PV–only system more profitable than PV–storage systems in the shorter term (Figure 5). This initial investment capital subsidy is uncommon in the European market, and makes a big difference to the affordability of the technology. The storage system, not having any available capital subsidies, will only be able pay itself off at the end of the PV lifetime (year 19–25), making it unprofitable for the current market. Assuming a self–consumption rate of 30% the payback period for PV–only systems in the four different locations will range between 12–17 years (Table 3). Based on the results highlighted in Figure 5, PV–storage systems are likely to be less profitable than PV–only systems, which is largely due to a lack of available financial incentives. However, if incentive schemes for investments in storage solutions would be available in Norway, this would be likely to change. A reduction in FiTs or a greater increase in cost/kWh than what is expected could also make battery storage solutions more profitable for the consumer.

Alternative technologies and methods to lower energy consumption

Although this work is interested in the profitability of PV–storage systems, it also recognises that there are more than one way to reduce energy consumption. Many approaches to improve energy efficiency are based around the Kyoto pyramid as defined by Dokka and Hermstad (2005) (Figure 7). Based on the

Trias Energetica method described by Lysen (1996), the Kyoto pyramid is a strategic approach on how to design low energy buildings in Norway; first reduce energy demand by improving the building fabric, then tackle occupants' consumption behaviour before investing in renewable energy technologies. The building envelopes (facades, roof, glazing) in Nordic houses are already relatively energy efficient due to cold weather, and will see some, but not major improvements from reducing U–values (Nippala and Vainio, 2017). However, there is a "knowledge–action gap" between people's self–reported environmental attitudes and values and their observed behaviour (Frederiks, Stenner and Hobman, 2015). This increases the potential for reducing energy consumption by altering occupants' behaviour, which will be explored further in section 4.4.1. Step 3, 4 and 5 on the Kyoto pyramid are, however, where the greatest reduction potential lies (Nippala and Vainio, 2017).

Figure 5. Kyoto pyramid – a hierarchy for improving energy efficiency in buildings (Dokka and Hermstad (SINTEF), 2005).

Ensuring efficient electricity use

Nord et al. (2018) describes how occupant behaviour (OB) affects the total energy demand of Norwegian households as well as the stress on the grid. Using a zero emissions building as the reference household, the study limits the impact of other variables and isolates OB as the only variable to affect energy use, assuming the weather is the same both sample years. OB entails all scheduled use of electric equipment and alterations to the air flow rate, such as turning lights on and off, opening windows and other openings, the use of hot water, and the pattern of heating. Figure 8 shows the results from the study published by Nord et al. (2018) and illustrates the impact conscious use of energy can have on the total energy consumption and stress on the grid. This confirms the results from the present study in terms of simulated energy consumption throughout the year, but also illustrates the impact occupants have on the annual demand pattern (load). However, the study estimates that the OB model only reduces the stress on the grid by a maximum of 5%, and has the potential to increase the stress as well. By comparing this to the results of the generated electricity in Table 2, the benefits from utilising solar energy are greater than altering consumer behaviour. Bahaj and James (2007) conclude that installing PV systems in social housing can increase consumer awareness and potentially reduce overall consumption, so the benefits of altering OB and solar PV might both be achieved by investing in a PV system.

Figure 6. Loads on the grid from domestic heating, cooling and hot water consumption (DHW) in two different models, the reference model (a) and conscious occupant behaviour (b). Source: Nord et al. (2018).

Intermittency – a case for residential energy storage systems

At the end of 2016, Norway had a total installed capacity of 27MW from PV panels across the country. This accounted for 0.02% of total energy generation, and did not affect the national grid to any considerable extent (NVE, 2017). However, looking to other economies who have had a rapid increase in PV installations (e.g. Japan, China, Germany), intermittent power fed into the grid can have negative consequences (Yu, Popiolek and Geoffron, 2014). After the windfall in profits on the PV market between 2004–2008, Germany saw a significant increase in power intermittency and a reduced stability on the electricity grid as a result of many years of policies favouring residential PV panels with high FiTs (Hoppmann, Huenteler and Girod, 2014). With a strong continuation of PV deployment and decreasing prices of PV systems, the government was faced with two challenges; 1. integrating an increasing capacity of intermittent power to the electricity grid without reducing stability, and 2. altering incentive schemes to match up with the reduced cost of investment.

In order to increase self–consumption and support investments in storage systems, the German government introduced a new requirement; for all new PV plants, of any size, to have a remote control for the government to be able to temporarily disconnect them from the grid (Ausschuss für Umwelt, Naturschutz und Reaktorsicherheit – Deutsche Bundestag, 2011). This meant that electricity generating dwellings could miss out on FiTs and the benefits of generating electricity if they didn't consume it on site. Simultaneously, the government introduced a self–consumption bonus to substitute FiTs when not exporting to the grid to make it more profitable to invest in a storage system. As Norway already have established reward systems and incentive schemes for PV investments, the biggest issue facing the implementation of PV systems is the support of the market to build the PV industry and create jobs (Hoppmann, Huenteler and Girod, 2014). Once the market has matured and the initial investment costs have decreased, it might be more profitable for self–generators to consume the power generated rather than exporting it. Learning from the German experience, storage options could be more profitable on the Norwegian market in a few years when the initial incentive schemes have been reduced and the market has stabilised.

Representativeness of results

This study appreciates that the definition of the typical household is not representative for all households in Norway and consequently, that the results are only representative for a share of the Norwegian market. As single occupancy, detached dwellings are the main target market for private PV systems, the results are likely to represent 92% of Norwegian households (SSB, 2017d). However, not all of these are economically independent and able to afford the initial cost of investment, even with governmental incentives. Using an occupancy profile similar to Aragon et al. (2018) it is possible to determine what percentage of suitable households are likely to be able to afford to invest in a PV system. Based on statistics published by SSB, households consisting of a married or cohabitating couple with or without kids (of any age(s)) are the most likely to have their main source of income from employment or saved up retirement money (SSB, 2017e). This group of people represents 52% of the total population and have a median salary of 556,400–963,300 NOK (SSB, 2017e), which is significantly higher than the remaining 48%. In light of this, the results from this study are likely to be relevant for 52% of the Norwegian population, based on annual income and ownership status of houses. It is to be expected that a small percentage of households that do not fit this description are interested in investing in PV systems, but in the context of this study they will not be accounted for.

Future research and recommendations

To further this research, a greater analysis of actual compared to simulated performance is required. Using statistical methods, any statistical relationships between building age and energy consumption (Aksoezen et al. 2015) should be determined in order to provide better guidance for household owners looking to improve the energy rating of their house and to invest in low–carbon technologies. This study could also be beneficial for local and national governing bodies to guide the development of planning policies. By adapting an evidence based approach there would be a greater opportunity to decarbonise the domestic sector through focussing their efforts on parts of the building stock with the highest energy saving potentials.

To better understand the evolution of the PV–storage market it would be beneficial to gain better insight into how self–consumption can be maximised. Luthander et al. (2015) reviews studies on how self–consumption rates are impacted by the use of battery storage, demand side management and combinations of technologies. With an electric vehicle (EV) market share of 29% (IEA, 2017c), Norway has a high potential to utilise the storage capacity of the EVs and create a vehicle–to–grid system where EVs are connected to the grid when stationary and used as buffers to intermittency on the grid (Kempton and Tomić, 2005). Utilising the intermittent energy from installed PV panels to power the increasing number of EVs could both reduce the intermittency issue, and reduce the increased stress they put on the grid in "peak charging hours" (Drude, Pereira Junior and Rüther, 2014). A more detailed study on the potential for connecting the private transport sector to the domestic energy generation in Norway, would complement the findings from this research and widen the perspective on how residential PV could be integrated better into to the national electricity grid or in micro–grids.

Conclusions

This study conducted a TRNSYS simulation to simulate and, consequently, better understand the energy demand from different households of different ages and locations in Norway. Using the Kyoto pyramid it was found that households that have seen a moderate retrofit to building fabrics are likely to benefit more from using solar energy to reduce energy imported from the grid, than to invest in further upgrades to external surfaces. This is likely due to the colder Norwegian climate, which encourages energy efficient buildings to keep houses warm during the colder months. The results from the simulations are similar to reported observed demand and assumptions made about building fabric and retrofit state for the "typical" house are therefore likely to reflect the state of the current building stock.

The two 3 kWp PV systems explored in this study, with and without storage, have distinctly different payback periods due to differences in availability of financial incentives and governmental support schemes. A PV–only system will, depending on latitude and local climate, have a payback period of 12–17 years. In contrast, a PV–storage system has a payback period of 19–25 years, which consequently has the possibility of never fully recovering the initial cost. PV–only systems are therefore much more likely to be economically beneficial for the household as the last 8–13 years of the system lifetime, on average 25–30 years, will bring economic benefits even if only 30% of the total electricity generated is consumed within the household.

This study does not consider scenarios where storage solutions are incentivised, but looking to other, more mature, PV markets like Germany, it is likely that storage systems will be favoured more in the future. This would alter the payback period of the two systems and make a higher rate of self–consumption more beneficial. With current incentive schemes a PV–storage system will not be more beneficial than a PV–only system within the lifetime of a PV panel being the limiting factor. However, in a political–financial environment with lower FiTs and reduced initial investment support, this would be likely to change.

Even though investments in residential PV system in a greening economy like Norway are unlikely to decarbonise the domestic sector, a reduction in demand from the grid could benefit the transport sector through EVs. Alternative storage solutions, such as utilising EVs in vehicle–to–grid systems, should therefore be researched further specifically for the Norwegian market. As the largest per capita market in the world for EVs, utilising their storage capacity could bridge the gap between cost and benefit for the consumer, while simultaneously decarbonising private transport.

Acknowledgements

I would like to thank the University of Southampton for providing me with the resources needed and the opportunity to write this dissertation in the final year of my bachelor degree. Despite countless hours of frustration I have learnt so much from undertaking this research project, both in the scientific arena, but also on a more personal level. Writing this dissertation has taught me that patience is key and to realise and use the expertise from people I surround myself with.

I would especially like to say thank you to my supervisor, Prof. Patrick James, who have supported my every step along this process. You provided me with the tools I needed for this project and inspired me greatly with your knowledge and expertise in the field of micro renewables.

In addition, I would like to thank my university tutor, Prof. Simon Kemp, for encouraging me throughout my undergraduate studies. Your passion for your students and the work you do to promote sustainable development has inspired me immensely.

Finally, I would like to thank Matthew Ferguson, Johanne Skrefsrud and Anna Pennington for listening to my frustrations along the way and for helping me with all the small things I didn't realise I needed help with.

Thank you!

References

Aksoezen, M., Daniel, M., Hassler, U. and Kohler, N. (2015). Building age as an indicator for energy consumption. Energy and Buildings, 87, pp.74–86.

Aragon, V., Gauthier, S., Warren, P., James, P. and Anderson, B. (2017). Developing English domestic occupancy profiles. Building Research & Information, pp.1–19.

Arteconi, A., Ciarrocchi, E., Pan, Q., Carducci, F., Comodi, G., Polonara, F. and Wang, R. (2017). Thermal energy storage coupled with PV panels for demand side management of industrial building cooling loads. Applied Energy, 185, pp.1984–1993.

Ausschuss für Umwelt, Naturschutz und Reaktorsicherheit (Deutsche Bundestag) (2011). Öffentliche Anhörung zum Thema „Ausbau der erneuerbaren Energien/Energiewende". [ebook] Berlin. Available at: http://lvbw–wka.de/media/linnen/EEG_Prot_17_47_DB_Umweltausschuss.pdf [Accessed 25 Apr. 2018].

Bahaj, A. and James, P. (2007). Urban energy generation: The added value of photovoltaics in social housing. Renewable and Sustainable Energy Reviews, 11(9), pp.2121–2136.

Balaras, C., Gaglia, A., Georgopoulou, E., Mirasgedis, S., Sarafidis, Y. and Lalas, D. (2007). European residential buildings and empirical assessment of the Hellenic building stock, energy consumption, emissions and potential energy savings. Building and Environment, 42(3), pp.1298–1314.

Ballarini, I., Corgnati, S. and Corrado, V. (2014). Use of reference buildings to assess the energy saving potentials of the residential building stock: The experience of TABULA project. Energy Policy, 68, pp.273–284.

Berardi, U. (2017). A cross–country comparison of the building energy consumptions and their trends. Resources, Conservation and Recycling, 123, pp.230–241.

Bhandari, K., Collier, J., Ellingson, R. and Apul, D. (2015). Energy payback time (EPBT) and energy return on energy invested (EROI) of solar photovoltaic systems: A systematic review and meta–analysis. Renewable and Sustainable Energy Reviews, 47, pp.133–141.

Bjartnes, G. and Skaare Amundsen, J. (2016). Kraftmarkedsanalyse mot 2030. [online] Oslo: NVE. Available at: https://www.nve.no/Media/5160/2016_12_nve_kraftmarkedsanalyse_mot_2030.pdf [Accessed 14 Apr. 2018].

Bøeng, A. (2014). På verdenstoppen i bruk av strøm. Samfunnsspeilet, 4, pp.9–15.

Byggteknisk forskrift (TEK17). Forskrift 19. juni 2017 nr. 840 om tekniske krav til byggverk. Forskrift om tekniske krav til byggverk 2017 (Norway) Available from: https://lovdata.no/dokument/SF/forskrift/2017–06–19–840#KAPITTEL_14 [Accessed 20 Apr. 2018].

Dannevig, P. and Harstveit, K. (2013). Klima i Norge. In: Store Norske Leksikon. [online] Available at: https://snl.no/Klima_i_Norge#–Nedb%C3%B8r_og_skyer [Accessed 18. Apr. 2018].

Dascalaki, E., Droutsa, K., Balaras, C. and Kontoyiannidis, S. (2011). Building typologies as a tool for assessing the energy performance of residential buildings – A case study for the Hellenic building stock. Energy and Buildings, 43(12), pp.3400–3409.

Dokka, T. and Hermstad, K. (2005). Fremtidens energieffektive boliger – en håndbok for planlegging av passivhus og lavenergiboliger. [online] Trondheim: SINTEF. Available at: http://www.arkitektur.no/?nid=155887&pid0=154999&cat=74568&pid2=157909 [Accessed 23 Apr. 2018].

Drude, L., Pereira Junior, L. and Rüther, R. (2014). Photovoltaics (PV) and electric vehicle–to–grid (V2G) strategies for peak demand reduction in urban regions in Brazil in a smart grid environment. Renewable Energy, 68, pp.443–451.

WhiteBoxTechnologies (nd). Weather Data. [online] Available at: http://weather.whiteboxtechnologies.com/ [Accessed 17 July 2018].

ENOVA (2012). Energieffektivisering av norske boliger. Potensial– og barrierestudien. [online] Available at: https://www.enova.no/download/?objectPath=upload_images/7260B28F559045159FD9213FA7AB989E.pdf&filename=Underlagsrapport%20Potensial–%20og%20barrierestudien%20Bolig%20Prognosesenteret.pdf [Accessed 23 Apr. 2018].

ENOVA (2018). El–produksjon. [online] Available at: https://www.enova.no/privat/alle–energitiltak/solenergi/el–produksjon–/ [Accessed 24 Apr. 2018].

EPISCOPE (2016). NO Norway – Residential Building Typology. [online] Available at: http://episcope.eu/building–typology/country/no/ [Accessed 25 Apr. 2018].

Eurostat (2016). Electricity prices by type of user. [online] Available at: http://ec.europa.eu/eurostat/tgm/table.do?tab=table&init=1&language=en&pcode=ten00117&plugin=1 [Accessed 13 Mar. 2018].

Eurostat (2017). Energy consumption in households. [online] Available at: http://ec.europa.eu/eurostat/statistics–explained/index.php/Energy_consumption_in_households [Accessed 3 Mar. 2018].

Frederiks, E., Stenner, K. and Hobman, E. (2015). Household energy use: Applying behavioural economics to understand consumer decision–making and behaviour. Renewable and Sustainable Energy Reviews, 41, pp.1385–1394.

Goodrich, A., James, T. and Woodhouse, M. (2012). Residential, commercial, and utility–scale photovoltaic (PV) system prices in the United States: current drivers and cost–reduction opportunities (No. NREL/TP–6A20–53347). National Renewable Energy Laboratory (NREL), Golden, CO..

Green, M., Hishikawa, Y., Warta, W., Dunlop, E., Levi, D., Hohl–Ebinger, J. and Ho–Baillie, A. (2017). Solar cell efficiency tables (version 50). Progress in Photovoltaics: Research and Applications, 25(7), pp.668–676.

Herrando, M., Markides, C. and Hellgardt, K. (2014). A UK–based assessment of hybrid PV and solar–thermal systems for domestic heating and power: System performance. Applied Energy, 122, pp.288–309.

Hoppmann, J., Huenteler, J. and Girod, B. (2014). Compulsive policy–making—The evolution of the German feed–in tariff system for solar photovoltaic power. Research Policy, 43(8), pp.1422–1441.

IEA (2004). National Survey Report of PV Power Applications in Norway 2003. [online] Available at: http://www.iea–pvps.org/index.php?id=93&eID=dam_frontend_push&docID=101 [Accessed 12 Apr. 2018].

IEA (2017b). CO2 Emissions from Fuel Combustion 2017 – Highlights. [online] Available at: http://www.iea.org/publications/freepublications/publication/CO2EmissionsfromFuelCombustionHighlights2017.pdf [Accessed 13 Mar. 2018].

IEA (2017c). Global EV Outlook 2017. [online] Available at: https://www.iea.org/publications/freepublications/publication/GlobalEVOutlook2017.pdf [Accessed 25 Apr. 2018].

International Energy Association (IEA) (2017a). Norway 2017 Review. Energy Policies of IEA Countries. [online] Available at: https://www.iea.org/publications/freepublications/publication/EnergyPoliciesofIEACountriesNorway2017.pdf [Accessed 13 Mar. 2018].

IRENA (2012). Solar Photovoltaics, Volume 1: Power Sector. RENEWABLE ENERGY TECHNOLOGIES: COST ANALYSIS SERIES. [online] Available at: https://www.irena.org/documentdownloads/publications/re_technologies_cost_analysis–solar_pv.pdf [Accessed 12 Apr. 2018].

Johann, A. and Madlener, R. (2014). Profitability of Energy Storage for Raising Self–consumption of Solar Power: Analysis of Different Household Types in Germany. Energy Procedia, 61, pp.2206–2210.

Kempton, W. and Tomić, J. (2005). Vehicle–to–grid power fundamentals: Calculating capacity and net revenue. Journal of Power Sources, 144(1), pp.268–279.

Luthander, R., Widén, J., Nilsson, D. and Palm, J. (2015). Photovoltaic self–consumption in buildings: A review. Applied Energy, 142, pp.80–94.

Lysen, E.H. (1996). The Trias Energica: solar energy strategies for developing countries. Proceedings of the EUROSUN Conference.

Meld. St. 25 (2015–2016) Kraft til endring — Energipolitikken mot 2030. [online] Available at: https://www.regjeringen.no/no/dokumenter/meld.–st.–25–20152016/id2482952/sec1 [Accessed 20 Apr. 2018].

Meral, M. and Dinçer, F. (2011). A review of the factors affecting operation and efficiency of photovoltaic based electricity generation systems. Renewable and Sustainable Energy Reviews, 15(5), pp.2176–2184.

Nippala, E. and Vainio, T. (2017). Boosting deep renovation in the Nordic Countries. In: 9th Nordic Conference on Construction Economics and Organization. [online] Lyngby, Denmark: Polyteknisk Forlag, pp.388–395. Available at: http://vbn.aau.dk/files/259711112/9th_Proceedings.pdf#page=398 [Accessed 23 Apr. 2018].

Noll, D., Dawes, C. and Rai, V. (2014). Solar Community Organizations and active peer effects in the adoption of residential PV. Energy Policy, 67, pp.330–343.

Nord, N., Tereshchenko, T., Qvistgaard, L. and Tryggestad, I. (2018). Influence of occupant behavior and operation on performance of a residential Zero Emission Building in Norway. Energy and Buildings, 159, pp.75–88.

Norwegian Solar Energy Group (2018). Solcellesystemer og sol i systemet. [online] Available at: https://static1.squarespace.com/static/597512eb579fb3d3de0207aa/t/5aaef19d562fa7d5c7274b84/1521414583385/Rapport_solkraft_markedsutvikling_2017.pdf [Accessed 12 Apr. 2018].

Norwegian Water Resources and Energy Directorate (NVE) (2015). Plusskunder. [online] Available at: https://www.nve.no/reguleringsmyndigheten–for–energi–rme–marked–og–monopol/nettjenester/nettleie/tariffer–for–produksjon/plusskunder/ [Accessed 24 Apr. 2018].

Nykamp, H. (2017). A transition to green buildings in Norway. Environmental Innovation and Societal Transitions, 24, pp.83–93.

Otovo (2018a) Otovo. [online] Available at: https://www.otovo.no/ [Accessed 24 Apr. 2018].

Otovo (2018b). Nabostrøm [online] Available at: https://www.otovo.no/grid [Accessed 24 Apr. 2018].

Pearce, D., Markandya, A. and Barbier, E. (1989). Blueprint for a Green Economy. 1st ed. New York: Earthscan.

Rickerson, W., Couture, T., Barbose, G., Jacobs, D., Parkinson, G., Chessin, E., Belden, A., Wilson, H. and Barrett, H. (2014). Residential prosumers: Drivers and policy options (re–prosumers). International Energy Agency (IEA)–Renewable Energy Technology Deployment.

Solcellespesialisten (2018a). Solcellepanel monokrystallinsk 295W 164x99.2x4cm IBC MonoSol 295 ZX4. [online] Available at: https://www.solcellespesialisten.no/solcellepanel/ibc–monosol–280–zx.html [Accessed 24 Apr. 2018].

Solcellespesialisten (2018b). Fronius Galvo 3.1. [online] Available at: https://www.solcellespesialisten.no/inverter–vekselretter/1–fase/fronius–galvo–3–1.html [Accessed 24 Apr. 2018].

Solcellespesialisten (2018c). Nøkkelferdige solcelleanlegg. [online] Available at: https://www.solcellespesialisten.no/solcelleanlegg/solcellepanel–til–hus–med–solcellepanel–pa–taket/nokkelferdige–solcelleanlegg–med–montering.html [Accessed 24 Apr. 2018].

Solcellespesialisten (2018d). Komplette systemer. [online] Available at: https://www.solcellespesialisten.no/stromlagring/komplette–systemer.html?dir=asc&order=position [Accessed 24 Apr. 2018].

Statistics Norway (SSB) (2013). Population and housing cencus, employment and education, 2011. [online] Available at: https://www.ssb.no/en/befolkning/statistikker/fobsysut/hvert–10–aar/2013–04–18 [Accessed 25 Apr. 2018].

SSB (2014a). Energibruk i husholdningene. [online] Available at: https://www.ssb.no/energi–og–industri/statistikker/husenergi/hvert–3–aar [Accessed 26 Jan. 2018].

SSB (2014b). Energy usage in households, by year of construction 2001 – 2012. [online] Available at: https://www.ssb.no/en/statbank/table/10579?rxid=f4e2f9b2–6206–48e1–bb10–fa677d7ebefa [Accessed 25 Apr. 2018].

SSB (2017a). Elektrisitet. [online] Available at: https://www.ssb.no/energi–og–industri/statistikker/elektrisitet/aar [Accessed 26 Jan. 2018].

SSB (2017b). Emissions of greenhouse gases. [online] Available at: https://www.ssb.no/klimagassn/ [Accessed 3 Mar. 2018].

SSB (2017c) Families and households. [online] Available at: https://www.ssb.no/en/befolkning/statistikker/familie/aar [Accessed 25 April 2018].

SSB (2017d). Boliger. [online] Available at: https://www.ssb.no/statbank/table/06266/tableViewLayout1/?rxid=c537ed48–10e5–4ebe–b782–bdfdcb23ae78 [Accessed 26 Jan. 2018].

SSB. (2017e). Income and wealth statistics for households. [online] Available at: https://www.ssb.no/en/inntekt–og–forbruk/statistikker/ifhus [Accessed 25 Apr. 2018].

SSB (2018a). Elektrisitetspriser. [online] Available at: https://www.ssb.no/energi–og–industri/statistikker/elkraftpris [Accessed 13 Mar. 2018].

SSB (2018b). Dwellings: Dwellings, by year of construction and utility floor space (M) 2007 – 2018. [online] Available at: https://www.ssb.no/en/statbank/table/06517?rxid=e0420aba–1427–4c99–8353–74ef6e4ac66c [Accessed 25 Apr. 2018].

SSB (2018c). Dwellings: Dwellings, by type of building and year of construction (M) 2006 – 2018. [online] Available at: https://www.ssb.no/en/statbank/table/06266?rxid=5593e6f9–8508–46be–a09c–cfc9033d0d30 [Accessed 25 Apr. 2018].

Standard Norge (2012). Calculation of areas and volumes of buildings. NS 3940:2012.

Sunikka–Blank, M., Chen, J., Britnell, J. and Dantsiou, D. (2012). Improving Energy Efficiency of Social Housing Areas: A Case Study of a Retrofit Achieving an "A" Energy Performance Rating in the UK. European Planning Studies, 20(1), pp.131–145.

SunPower (2017). SunPower® X–Series Commercial Solar Panels | X22–360–COM. [ebook] Available at: https://us.sunpower.com/sites/sunpower/files/uploads/resources/sp–x22–360–com–ds–en–ltr–mc4comp–514617–0.pdf [Accessed 25 Apr. 2018].

TABULA (2016). Norway, building data. [online] Available at: http://webtool.building–typology.eu/#bd [Accessed 1 Feb. 2018].

The World Bank (2014). Electric power consumption (kWh per capita). [online] Available at: https://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC [Accessed 25 Apr. 2018].

Tokle, T., Tønnesen, J. and Enlid, E. (1999). Status for energibruk, energibærere og utslipp for den norske bygningsmassen. Trondheim: SINTEF.

UN Sustainable Development (nd). Green economy. [online] Available at: https://sustainabledevelopment.un.org/index.php?menu=1446 [Accessed 25 Apr. 2018].

United Nations Environment Programme (UNEP) (2011). Towards a Green Economy: Pathways to Sustainable Development and Poverty Eradication. Towards a Green Economy: Pathways to Sustainable Development and Poverty Eradication. [online] Available at: https://www.unenvironment.org/search/node?keys=Towards+a+Green+Economy%3A+Pathways+to+Sustainable+Development+and+Poverty+Eradication [Accessed 25 Apr. 2018].

United States Environmental Protection Agency. (2016). Green Building – Basic Information. [online] Available at: https://archive.epa.gov/greenbuilding/web/html/about.html [Accessed 25 Apr. 2018].

University of Bergen (nd). Byggeskikk. [online] Available at: http://home.hib.no/mediesenter/haandverk/byggeskikk/takkonstruksjoner.htm [Accessed 9 Apr. 2018].

Utenriksdepartementet (2016). Norway's follow–up of Agenda 2030 and the Sustainable Development Goals. New York: UN. [online] Available at: https://www.regjeringen.no/en/dokumenter/follow–up–sdg2/id2507259/ [Accessed 24 Apr. 2018].

Weniger, J., Tjaden, T. and Quaschning, V. (2014). Sizing of Residential PV Battery Systems. Energy Procedia, 46, pp.78–87.

Yu, H., Popiolek, N. and Geoffron, P. (2014). Solar photovoltaic energy policy and globalization: a multiperspective approach with case studies of Germany, Japan, and China. Progress in Photovoltaics: Research and Applications, 24(4), pp.458–476.

Appendix 1

The payback periods are calculated based on assumption on how much electricity generated from the PV system is consumed within the household, either with or without the use of energy storage. All assumptions are described in Section 2.3.

Table A.1. – Oslo

Table A.2. – Bergen

Table A.3. – Trondheim

Table A.4. – Narvik

The formulae used in Tables A.1–4 in this appendix are as follows:

A

B

C

D

E

1

Year

100% export to grid annual savings

100% self–consumption annual savings

70% export to grid / 30% self–consumption (PV–only)

30% export to grid / 70% self–consumption (PV–storage)

2

N=1,2,3,…

=PVoutput for location * 1 NOK (FiT)

=PVoutput for location*1.04^N–1

=initial investment cost (PV–only)–B2*0.7–C2*0.3

=initial investment cost (PV–storage–B2*0.3–C2*0.7

Table A.1. The system payback process for two PV systems in Oslo (PV–only and PV–storage).

Year

100% export to grid annual savings

100% self–consumption annual savings

70% export to grid / 30% self–consumption (PV–only)

30% export to grid / 70% self–consumption (PV–storage)

0

–40463

–77362

1

–3211

–3211

–37252

–74151

2

–3211

–3339

–34002

–70850

3

–3211

–3473

–30713

–67456

4

–3211

–3612

–27381

–63964

5

–3211

–3756

–24007

–60371

6

–3211

–3907

–20587

–56673

7

–3211

–4063

–17120

–52866

8

–3211

–4225

–13605

–48945

9

–3211

–4394

–10039

–44905

10

–3211

–4570

–6420

–40743

11

–3211

–4753

–2747

–36452

12

–3211

–4943

984

–32029

13

–3211

–5141

4774

–27467

14

–3211

–5347

8626

–22761

15

–3211

–5560

12541

–17905

16

–3211

–5783

16524

–12894

17

–3211

–6014

20576

–7721

18

–3211

–6255

24700

–2379

19

–3211

–6505

28899

3137

20

–3211

–6765

33176

8836

21

–3211

–7036

37535

14724

22

–3211

–7317

41978

20810

23

–3211

–7610

46508

27100

24

–3211

–7914

51130

33603

25

–3211

–8231

55847

40328

26

–3211

–8560

60663

47283

27

–3211

–8902

65581

54478

28

–3211

–9258

70607

61923

29

–3211

–9629

75743

69626

30

–3211

–10014

80995

77599

Table A.2. The system payback process for two PV systems in Bergen (PV–only and PV–storage).

Year

100% export to grid annual savings

100% self–consumption annual savings

70% export to grid / 30% self–consumption (PV–only)

30% export to grid / 70% self–consumption (PV–storage)

0

–40463

–78410

1

–2207

–2163

–38269

–76234

2

–2207

–2249

–36049

–73997

3

–2207

–2339

–33803

–71698

4

–2207

–2433

–31528

–69333

5

–2207

–2530

–29224

–66899

6

–2207

–2631

–26890

–64395

7

–2207

–2737

–24524

–61817

8

–2207

–2846

–22125

–59163

9

–2207

–2960

–19692

–56429

10

–2207

–3078

–17224

–53612

11

–2207

–3202

–14718

–50709

12

–2207

–3330

–12174

–47716

13

–2207

–3463

–9591

–44630

14

–2207

–3601

–6965

–41447

15

–2207

–3745

–4297

–38163

16

–2207

–3895

–1583

–34774

17

–2207

–4051

1177

–31276

18

–2207

–4213

3986

–27665

19

–2207

–4382

6845

–23936

20

–2207

–4557

9757

–20084

21

–2207

–4739

12724

–16105

22

–2207

–4929

15747

–11992

23

–2207

–5126

18830

–7742

24

–2207

–5331

21974

–3349

25

–2207

–5544

25182

1194

26

–2207

–5766

28457

5893

27

–2207

–5996

31800

10752

28

–2207

–6236

35216

15780

29

–2207

–6486

38707

20982

30

–2207

–6745

42275

26366

Table A.3. The system payback process for two PV systems in Trondheim (PV–only and PV–storage).

Year

100% export to grid annual savings

100% self–consumption annual savings

70% export to grid / 30% self–consumption (PV–only)

30% export to grid / 70% self–consumption (PV–storage)

0

–40463

–77886

1

–2742

–2687

–37737

–75182

2

–2742

–2795

–34979

–72403

3

–2742

–2906

–32188

–69546

4

–2742

–3023

–29362

–66608

5

–2742

–3144

–26499

–63585

6

–2742

–3269

–23599

–60474

7

–2742

–3400

–20660

–57271

8

–2742

–3536

–17680

–53973

9

–2742

–3678

–14657

–50576

10

–2742

–3825

–11590

–47076

11

–2742

–3978

–8477

–43469

12

–2742

–4137

–5317

–39751

13

–2742

–4302

–2107

–35917

14

–2742

–4474

1155

–31962

15

–2742

–4653

4470

–27882

16

–2742

–4839

7841

–23672

17

–2742

–5033

11271

–19326

18

–2742

–5234

14760

–14840

19

–2742

–5444

18313

–10207

20

–2742

–5661

21931

–5421

21

–2742

–5888

25617

–477

22

–2742

–6123

29373

4632

23

–2742

–6368

33203

9913

24

–2742

–6623

37109

15371

25

–2742

–6888

41095

21016

26

–2742

–7164

45163

26853

27

–2742

–7450

49318

32890

28

–2742

–7748

53562

39137

29

–2742

–8058

57898

45600

30

–2742

–8380

62332

52289

Table A.4. The system payback process for two PV systems in Narvik (PV–only and PV–storage).

Year

100% export to grid annual savings

100% self–consumption annual savings

70% export to grid / 30% self–consumption (PV–only)

30% export to grid / 70% self–consumption (PV–storage)

0

–40463

–78338

1

–2281

–2235

–38195

–76089

2

–2281

–2325

–35901

–73777

3

–2281

–2418

–33579

–71400

4

–2281

–2514

–31228

–68956

5

–2281

–2615

–28847

–66441

6

–2281

–2720

–26434

–63853

7

–2281

–2828

–23989

–61189

8

–2281

–2942

–21510

–58445

9

–2281

–3059

–18996

–55619

10

–2281

–3182

–16444

–52708

11

–2281

–3309

–13855

–49707

12

–2281

–3441

–11226

–46614

13

–2281

–3579

–8556

–43425

14

–2281

–3722

–5842

–40135

15

–2281

–3871

–3084

–36741

16

–2281

–4026

–280

–33239

17

–2281

–4187

2573

–29623

18

–2281

–4354

5476

–25891

19

–2281

–4528

8431

–22037

20

–2281

–4710

11441

–18056

21

–2281

–4898

14507

–13943

22

–2281

–5094

17632

–9693

23

–2281

–5298

20818

–5300

24

–2281

–5510

24067

–759

25

–2281

–5730

27383

3936

26

–2281

–5959

30767

8792

27

–2281

–6198

34223

13814

28

–2281

–6445

37754

19010

29

–2281

–6703

41361

24387

30

–2281

–6971

45050

29951