ES16029

Examensarbete 30 hp September 2016

Evaluating the Potential for Floating Offshore Wind Power in Skagerrak The Golden Triangle Nils Jonsson Forsblad

Abstract Evaluating The Potential for Floating Offshore Wind in Skagerrak Nils Jonsson Forsblad

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

Wind power is a rapidly growing industry worldwide, both on- and offshore. Most of the good locations onshore in continental Europe are in use today, which has prompted a move offshore in recent years. Europe has by far the most offshore wind turbines installed, mostly located in the North sea. The low hanging fruits are locations with relatively shallow waters (up to 45-50 meters), a high and steady wind speed and is close to grid connections onshore. Big parts of the North Sea are suitable for this, but many places with good wind conditions worldwide are too deep. The next step for the industry is to move to these deeper waters, with the help of floating wind turbines. The first prototype floating turbines have been running for a couple of years, with even larger, albeit still pretty small, wind farms in the planning stage. This thesis looks on the possibility of building large floating wind farms in the future, specifically in the eastern most part of the North Sea - Skagerrak. Several different factors and stakeholders have been mapped out and important factors such as water depth, wind speed and seabed conditions considered to create four different future scenarios. Each scenario has been evaluated technically and Levelized Cost of Energy (LCOE) has been calculated to be able to compare the different locations. Since the technology is very new and under development, the initial costs are high. This gives the lower LCOE of 149 €/MWh. Many new developments are however expected in the years to come, which would lower the investment cost considerably, by up to 40% according to some sources. This would lower the LCOE to under 100 €/MWh. It is however also found that these investments carry many other positive effects, such as developing a new carbon neutral technology in Scandinavia which could become a big export worldwide. The social acceptance of bottom fixed foundation offshore (close to shore) and onshore wind power is also falling, and this would also be a big plus for floating offshore wind as it can be built so far offshore it can't be seen from land. Both Sweden and Denmark have big power plants closing in the coming decades, nuclear power in Sweden and coal fired power plants in Denmark. These need to be replaced either by import or by new carbon neutral power production.

Handledare: Johan Sandberg Ämnesgranskare: Urban Lundin Examinator: Petra Jönsson ISSN: 1650-8300, ES16029

Nils Jonsson Forsblad

Popul¨ arvetenskaplig sammanfattning V¨arlden st˚ ar inf¨or stora utmaningar med klimatf¨or¨andringar och stigande temperaturer; detta har gjort att m˚ anga l¨ander bygger om sina energisystem f¨or att g˚ a mot mer h˚ allbar energif¨ors¨ojning med f¨ornybara energik¨allor. I Sverige kommer fyra av de ˚ aldrande k¨arnkraftsreaktorerna snart att st¨angas ner p˚ a grund av behovet av dyra nya investeringar, samtidigt som all transport beh¨over st¨alla om till nya energik¨allor, d¨ar el kan komma att vara en viktig del. Motst˚ andet fr˚ an befolkningen om var nya vindkraftverk p˚ a land kan placeras blir allt st¨orre. Detta har gjort att m˚ anga l¨ander, framf¨orallt Tyskland och Danmark, har satsat mycket p˚ a att bygga andra generationens vindkraftsverk - ute p˚ a havet Offshore. Offshore har m˚ anga f¨ordelar, inte minst ¨ar vindf¨orh˚ allande generellt b¨attre, och placeras de l˚ angt ifr˚ an land s˚ a ”st¨or” de inte heller befolkningen. Dagens offshoreteknik kr¨aver fundament som st˚ ar p˚ a botten vilket betyder att fundamenten kan bli dyra och de byggs endast p˚ a grunda djup, upp till 50 meter. F¨or att ¨oppna upp st¨orre delar av havet f¨or byggnation har ytterligare ett steg tagits i vindkraftens utveckling och generation tre av vindkraften a¨r p˚ a v¨ag - flytande vindkraftverk. De flytande vindkraftverken a¨r till stor del i demonstationsstadiet, med ett f˚ atal byggda verk som ett test av tekniken. Det finns tv˚ a st¨orre (5-8 turbiner) parker i planeringsstadiet. Detta a¨r b˚ ada viktiga steg som visar p˚ a teknikens mognadsgrad. Dessa kommer att kunna byggas p˚ a mycket stora djup, 900 meter och djupare, vilket o¨ppnar stora delar av havet f¨or elkraftproduktion. F¨or att titta p˚ a en storskalig expansion av denna teknik i Skandinavien valdes Skagerrak ut att unders¨okas n¨armare. I Skagerrak finns de b¨asta vindf¨orh˚ allandena i Svenska vatten och ligger mellan de tre nordiska l¨anderna Sverige, Danmark och Norge. Det finns m˚ anga intressenter som uppt¨acktes och kartlades, som tex. naturreservat, milit¨ara ¨ovningsomr˚ aden och fiskeomr˚ aden. N¨ar alla dessa s˚ a kallade ”show stoppers” var funna s˚ a fanns det en stor del av havet kvar som inte var upptaget av andra aktiviteter d¨ar vindkraftverk skulle kunna byggas. F¨or att minska kostnader f¨or kedjor och ankare och reducera kostnaden f¨or anslutning till eln¨aten p˚ a land valdes 6 platser ut som l˚ ag relativt n¨ara till land och hade ett djup p˚ a 100-200 meter. Dessa 6 platser visar tillg¨angliga omr˚ aden dels f¨or stora parker anslutna till de enskilda l¨anderna Sverige, Danmark och Norge, och dels parker som s˚ a kallade ”teein” anslutningar till HVDC l¨ankar mellan l¨anderna Sverige-Danmark och SverigeNorge. Samtliga parker anv¨ander sig av 6 MW turbiner, d¨ar den minsta a¨r mellan

1

Nils Jonsson Forsblad

Sverige och Danmark med 83 st turbiner (498 MW), och den st¨orsta 300 turbiner (1800 MW) i tv˚ a omr˚ aden mellan Sverige och Norge. Ur j¨amf¨orelsesynpunkt utv¨arderas kostnader f¨or dessa parker med hj¨alp av ett LCOE v¨arde, som ¨ar en livscykelkostnad f¨or energin genererad av vindkraftsparkerna. En LCOE ber¨akning g¨or att h¨ansyn inte heller beh¨ovs tas till framtida elpriser eller subventioner. Ut¨over det har en rad experter intervjuats och deras ˚ asikter och kommentarer tagits med i utv¨arderingen av parkerna. Detta visar att om de skulle byggas idag s˚ a vore priset f¨or energin 149 e/MWh, men det finns stort utrymme f¨or utveckling. Industrin och politiker har g˚ att samman och satt ett m˚ al p˚ a bland annat en minskning av investeringskostnaden p˚ a 40 % innan 2020 f¨or havsbaserad vindkraft, n˚ agot som redan a¨r p˚ a god v¨ag! Med dessa kostnadsantaganden och en byggnation n˚ agra ˚ ar senare blir priset ist¨allet under 100 e/MWh. Det finns m˚ anga f¨ordelar som inte heller m¨ats monet¨art. En f¨orbindelse mellan Norge och Sverige g¨or att vattenkraft fr˚ an Norge kan anv¨andas d˚ a vindkraftsparken inte producerar el, vilket o¨kar tillf¨orlitligheten hos parken och f¨orb¨attrar utnyttjandet av vattenkraften. Det ger a¨ven en ¨okad produktion i s¨odra Sverige, d¨ar produktion annars saknas, vilket betyder l¨agre f¨orluster f¨or eln¨atet i helhet p˚ a grund av minskade transmissionsf¨orluster. Det ger ¨aven de nordiska l¨anderna ett f¨orspr˚ ang i en industri som har m¨ojlighet att v¨axa mycket i v¨arlden de kommande ˚ aren, eftersom stora delar av v¨arldens hav ¨ar f¨or djupa f¨or traditionell offshore vindkraft. M˚ anga utvecklingsl¨ander bygger ut sin elproduktion kraftigt, tyv¨arr ofta med fossil kraftproduktion och flera av dem har d˚ aliga f¨orh˚ allanden f¨or vindkraftverk med bottenfasta fundament, h¨ar kan flytande offshore vindkraft komma att bli en stor energik¨alla i framtiden.

2

Nils Jonsson Forsblad

Executive Summary In order to enable production in larger parts of the sea a new type of wind turbine technology is being developed - floating wind turbines. This would enable construction of wind farms in depths of more than than 900 meters, and unlock some of the best locations available out of a power production standpoint. This work is assessing the feasibility of floating wind farms in Skagerrak. Skagerrak is the easternmost part of the North Sea and is shared by the three nations Sweden, Denmark and Norway. The location gives a close proximity to large volumes of energy storage from hydro power in southern Norway, a well developed supply chain in the region and a close proximity to strong network points in the Scandinavian power grid. A range of interests related to marital spacial planning are mapped out and examined, as well as the seabed, wind and water depth conditions. This shows that areas that are in favorable conditions are more than 1800 km2 in size in depths of 100-200 m. Four different scenarios with large floating wind farms (600-1800 MW) are evaluated economically by calculating the cost of energy (LCOE e/MWh). The lowest cost in the base case is 149 e/MWh. This is based on numbers from this year (2016), and they are very likely to change in the future, since the technology is in an early stage of development. The investment cost for offshore wind is expected to be greatly reduced in the years to come, with cost compression expected in excess of 40% before 2020. If CAPEX is reduced according to the cost compression presented by Carbon Trust in the future, the LCOE would lower significantly and end up at 99 e/MWh. This cost reduction is expected to continue, with industry leaders recently releasing a press release promising a cost of 80 e/MWh or less in 2025 for offshore wind.

3

CONTENTS

Nils Jonsson Forsblad

Contents 1 Abbreviations and Acronyms

6

2 Introduction 2.1 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Limitations and uncertainties of the study . . . . . . . . . . . . . . .

7 7 9 9

3 Offshore wind 3.1 Wind power potential . . . . . . . . . . . . . . . . . . 3.2 Social acceptance . . . . . . . . . . . . . . . . . . . . 3.3 Floating offshore wind technology . . . . . . . . . . . 3.3.1 Types of submerged structure (Substructure) . 3.3.2 Anchors . . . . . . . . . . . . . . . . . . . . . 3.3.3 Pilot projects . . . . . . . . . . . . . . . . . . 3.4 Offshore substation . . . . . . . . . . . . . . . . . . . 3.5 Cost Reduction . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

10 10 11 12 12 13 13 14 15

4 Transmission 16 4.1 Electric grids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 4.2 Trade between SE3 and NO2 . . . . . . . . . . . . . . . . . . . . . . 17 5 Economics of an offshore wind farm 5.1 Capital Expenditures (CAPEX) . . . . 5.1.1 Costs of substructures . . . . . 5.1.2 Cost of Mooring . . . . . . . . . 5.1.3 Cost of Electrical infrastructure 5.2 Operating Expenditures (OPEX) . . . 5.3 LCOE . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

19 19 20 20 20 21 21

6 Skagerrak 6.1 Natura 2000, national nature reserves. 6.2 Risk areas . . . . . . . . . . . . . . . . 6.2.1 Mines and Dumping Sites . . . 6.2.2 Armed forces training area . . . 6.3 Shipping lanes . . . . . . . . . . . . . . 6.4 Commercial Fishing. . . . . . . . . . . 6.5 Water depth & Seabed . . . . . . . . . 6.6 Wind conditions . . . . . . . . . . . . . 6.7 Overview of Skagerrak . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

23 23 24 24 24 25 26 27 28 29

4

CONTENTS

7 Result 7.1 Reference park . . . . . 7.2 Park A . . . . . . . . . . 7.3 Park D . . . . . . . . . . 7.4 Park E . . . . . . . . . . 7.5 Park F . . . . . . . . . . 7.6 Park BC . . . . . . . . . 7.7 Sensitivity analysis . . . 7.7.1 Discount factor . 7.7.2 Lifetime . . . . . 7.7.3 Capacity factor . 7.8 Cost Compression result

Nils Jonsson Forsblad

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

8 Discussion and Conclusion

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

30 30 32 32 33 34 35 37 37 38 39 40 41

5

Abbreviations and Acronyms

Nils Jonsson Forsblad

Acknowledgments I would like to acknowledge a few people that have made the completion of this thesis possible. First and foremost Johan Sandberg who has guided and helped me during the whole thesis, and been the link between me and all of the other experts I have talked to. Thanks to Urban Lundin who always encouraged me when it felt like I was getting nowhere, and went above and beyond his role as subject reader. Also to Kjeller Vindteknikk who provided me with some much needed data. And lastly to the experts who came with much valued and great input for the thesis, Emir Isabegovic, Martin Westerlund, Magnus Ebbesen and Henrik Stiesdal.

1

Abbreviations and Acronyms

CAPEX GIS HVDC HVAC LCOE OPEX QGIS TLP WMS

CAPital EXpenditures Geographic Infromation System High Voltage Direct Current High Voltage Alternating Current Levelized Cost of Energy (or Electricity) Operational expenditures Quantum GIS Tension Leg Platform Web Map Service

6

Introduction

2

Nils Jonsson Forsblad

Introduction

The electricity generation in many nations is under a big change. The looming threat of climate change pushes many nations to abandon old, fossil based power production, in favor of renewable energy sources. The recent agreement made at Paris climate conference (COP21) for the reduction of climate change that was negotiated in 2015 shows that this development will continue for many years. In Sweden the old nuclear power plants are closing down, 4 reactors before 2020 and the remaining 6 some years after that. Denmark faces the same problem as it needs to phase out its coal fired power plants. Their electricity generation need to be replaced by something, either new generation or through import. There has been a big move towards renewable power in recent years and the wind power industry is growing rapidly. There has been a big focus on onshore wind power in Europe, but the biggest potential can be found offshore. [1] There are many great locations to be found for offshore wind in the North Sea, since it generally has steady and high wind speeds. The North Sea has many locations well suited for bottom fixed offshore wind, but there is also big parts that are too deep for conventional offshore wind. This is also true for most of the worlds’ oceans, so to be able to utilize more of the sea new floating offshore wind turbines are currently being tested. The purpose of this thesis is to investigate the feasibility of building an offshore wind farm in Skagerrak, answering relevant questions like: What are the difficulties of constructing a wind farm in Skagerrak? What other activities are conducted in the area and how do they interact with an offshore wind farm? How much would the cost of electricity be? Skagerrak is the easternmost part of the North Sea, and is shared by the three Scandinavian nations Sweden, Denmark and Norway. A project in this ocean need to have an idea of what other activities are conducted in the area, and what challenges or opportunities they represent. The goal of this thesis is to get an overview of what these are, for example nature reserves, military zones and commercial fishing areas.

2.1

Method

Much of the map data is available from the three Scandinavian nations’ government agencies. To get access to them several government agencies and some companies were contacted by phone and email. The information was often provided in WMS

7

Introduction

Nils Jonsson Forsblad

format, which is a type of map-service that can be used in different Geographic Information System (abbreviated GIS) programs, for example QGIS or ARCGIS. WMS is an online map service which is frequently and easily updated by the responsible institution or company, which makes sure the information is as up to date as possible. When a WMS server was not available data was downloaded from the homepage of the institution, for example from http://www.naturalearthdata. com/downloads/, which was where all of the maps, borders and city locations were found. These are not as easy to keep up to date but all the files used here are relatively recent. The data and information was primarily provided by: • The Natura 2000 network • The Swedish Armed Forces (F¨orsvarsmakten) • The Swedish Coast Guard (Kustbevakningen) • The Norwegian Coastal Administation (Kystverket) • Swedish Agency for Marine and Water management (Havs och vattenmyndigheten) • HELCOM • Kjeller Vindteknikk • Geological Survey of Norway (Norges Geologiske Underøkelse) • Mareano • The Swedish County Government (L¨ansstyrelsen) • Swedish Meteorological and Hydrological Institute (SMHI) • The Norwegian County Governor (Fylkesmannen) . . . The data provided was then used to map out the different areas of interest in Skagerrak, which in turn would be the basis for creating different potential wind farms and their connections to the grid in the three different nations. The economic viability of the wind farm is evaluated using LCOE, which takes the whole life cycle of the wind farm into account. LCOE is also useful because it makes comparison to other energy solutions easy, and is not dependent on the actual price of electricity or subsidies, which would be near impossible to estimate for such a long time period.

8

Introduction

2.2

Nils Jonsson Forsblad

Purpose

The goal is to investigate the possibility of building a big wind farm using the third generation of wind power - floating offshore wind - in Skagerrak. Different stakeholders in the area is investigated and mapped out, so any showstoppers for a wind park can be found. This information will then be used to make economic calculations of future wind farms in the area. To make sure that assumptions are relevant a few different experts are interviewed, both on the assumptions and their view of the industry and the power grid.

2.3

Limitations and uncertainties of the study

Since the technology being evaluated is in the development stage, some assumptions had to be made. The uncertainty in prices is one of the biggest problems, and how these prices will change in the future. To remedy this many experts have been consulted and their opinion on the different parts factored in, but it still only gives a rough estimate. The idea from the beginning was to have some sort of simulation to see the effects of an offshore wind farm on the national grids of the different nations. After consultation with experts from Gothia power and Svenska kraftn¨at this was changed. Since the park is hypothetical, it would be near impossible to find the information necessary to make any simulations that have any valuable results, so the focus of the study was moved from a simulation to look closer on the economic and societal factors.

9

Offshore wind

3

Nils Jonsson Forsblad

Offshore wind

The future bring many new challenges for the worlds energy systems. The problems with climate change are becoming bigger and more apparent which is moving governments and companies all over the world to invest more in power generation from renewable resources [2]. A promising alternative is wind power, which is already proven on land. Wind power has been a rapidly growing industry, with the total installed capacity in Europe increasing six-fold since 2000 [3]. Most of the installed wind power capacity is on land, but there has been a big move toward offshore in the last couple of years. Offshore does however make installation and servicing more difficult and expensive.

3.1

Wind power potential

The ocean provide no obstacles for the wind such as hills and forests do on land, which makes it possible to have overall shorter towers in comparison to the turbine blades. One of the biggest limiting factor is the ocean floor like depth and seabed conditions. Offshore also provide more area for construction, which makes scaling up of parks possible. A big onshore wind farm is considered small offshore, and there are wind farms planned that consist of several hundred turbines. The fact that bigger projects reduce overall costs (including servicing costs) is also a big plus. The sea provide a more stable and high speed wind for power generation and the capacity factor for offshore wind power is 30-50% (average 42 %), compared to onshore wind power 15-35%. [4] There are, however, big variations depending on the location of these wind farms, the Danish offshore wind farms have a capacity factor between 22.7 % and 48.3 % [5] and 41.1 % average for all 13 wind farms. The highest capacity factor recorded to date was December 2015 when the 175-turbine offshore wind farm London Array had a capacity factor of 78.5 %. Wind power generally have a higher capacity factor during the winter months, something that the offshore wind industry in the North Sea has shown. [6] The best places for wind power is often far out to sea, because the wind conditions are favorable (high and stable) and the way offshore wind power is constructed today, with bottom fixed foundations, makes building in deep places very expensive, and impossible on deeper waters. Like oil platforms before it, wind power also need to move away from bottom fixed foundations and towards floating foundations. The hope is that this will become cheaper than bottom fixed foundations as the technology matures, because of streamlined production and installation that does not require highly specialized ships.

10

Offshore wind

Nils Jonsson Forsblad

Floating wind power open up deeper parts of the ocean, 900 m and deeper, and the available power at these depths are estimated to be 4000 GW in Europe alone, as shown in Table 1. Table 1: Offshore wind power potential in Europe, USA and Japan. [1]

Region

Share of offshore wind resource in deep water locations (>60 m)

Potential floating wind capacity

Europe USA Japan

80% 60% 80%

4000 GW 2450 GW 500 GW

3.2

Social acceptance

The sites for onshore wind power are getting more limited, and the social acceptance in many countries for more onshore wind is lowering. The ”Not in my back yard” mentality, in other words that you may accept and even support more electricity from wind farms, but not anywhere where you can see the turbine is becoming more common. The sites onshore are often limited in size as well and big construction projects are hard to find suitable locations for. There are also limitations because of the sound that is emitted from the turbine which also increase as turbines become larger. The grids of most developed nations are not well suited for the task of sustaining a big wind farm in many locations, and in the case of Sweden the wind power potential offshore can be found in places where power generation is currently lacking in the southern parts of the nation. The sounds and visibility of wind farms are also less of a problem out to sea. How far √ you can see to the horizon can be calculated by d ≈ 3.57 h where d is the distance in kilometers and h is the height of the object. In the case of a wind turbine with a blade length of 60 meters, this would be √ around 160 meters high at its highest. This means that turbines are visible d ≈ 3.57 160 ≈ 45km at the most with binoculars. The distance is more likely only visible to the naked eye by the height of the tower, √ which is around 100 m, d ≈ 3.57 100 = 35.7km visibility. This means that much of the ocean cannot be seen by the naked eye, and one of the hurdles for wind farm development is avoided.

11

Offshore wind

3.3

Nils Jonsson Forsblad

Floating offshore wind technology

The two main differences between regular offshore wind turbines and floating ones are that they float and that they require some sort of mooring to be set in place. There are currently a couple of companies that are building these wind turbines and they use different technologies, and it is unclear which will be the dominant one in the future and different technologies could be best suited for different locations. One of the biggest hurdles when installing bottom fixed foundation offshore wind is the requirement of big and highly specialized ships. These need to be booked years in advance and are very expensive. Floating offshore wind have the potential to be completely built in harbor, and simply transported by tugboats to be installed far out to sea. This has the potential to greatly improve build times and reduce the costs of installation for offshore wind. 3.3.1

Types of submerged structure (Substructure)

There are three main concepts for the substructure of a floating wind farm, these are spar-bouy, semi-submersible and tension leg platform (TLP). Two derived concepts from these are multi-turbine platform and hybrid wind/wave which can use any of the substructures mentioned above. The spar-bouy is designed for depths of 100+ meters, and is constructed by using a steel tube that goes down in to the water and is filled with stone, water or sand to stabilize it as ballast. This construction require a higher depth than other designs, but have improved stability and doesn’t require active ballast. The semi-submersible can be used at lower depths, but requires a large and heavy sub-structure to provide sufficient stability. The low depth requirement makes it easier to tow in to shore and repair, but it may need an active ballast systems. The tension leg platform requires low structural mass since it gets its stability from the tension in the wires set in the mooring on the bottom, this does however require a more robust bottom fixture, which is harder to build and more expensive, and also limits the depths of which this type of floating wind farm i suitable. There are also ideas of building multi-turbine platforms, which are large steel platform which can hold 2-3 turbines. These large and heavy structures give the advantage of being more stable in the water, and they hope to make savings in installation cost by making fewer trips to shore necessary for the same number of turbines, and the lesser need for anchors. [1] [7]

12

Offshore wind

3.3.2

Nils Jonsson Forsblad

Anchors

Figure 1: The three main concepts of floating wind turbines. [8]

The mooring system is very important for the safety of the wind farm. Different types of anchors can be used in different conditions, both depending on the load they will need to hold and the soil condition of the area. In the case of Skagerrak, most of the seabed is covered in mud and sandy mud as shown in Figure 1. For this type of seabed, two types of anchors can be used, a drag driven or suction pile. Which choice is better depend on which substructure the wind farm is using. For spar-bouy the driven anchor should be enough, but for other types a suction pile may be more suited. [1] Apart from the obvious difference in the way the anchors work, the main difference is the price where a suction pile is up to three times as expensive as a drag driven anchor.[9] The type of seabed in Skagerrak is discussed more closely in the section ”Skagerrak”, but the type of seabed there is considered medium clay, which makes a drag driven type of anchor possible to use in the area. (Personal Correspondence, A. Lepland, Norske Geologiske Undersokelse (NGU), Feb-Apr 2016) [9] 3.3.3

Pilot projects

There are currently 3 full-scale pilot projects constructed at different places in the world Hywind (Statoil) in Norway, Windfloat (Principle power) in Portugal and Fukushima forward in Japan. [1] These projects can be seen as a proof of concept for this budding industry, and their non-floating and land based counterparts are often used as a template for the future cost of building large scale wind power parks.

13

Offshore wind

Nils Jonsson Forsblad

Statoil Hywind was the first installed full-scale floating offshore wind turbine when it was installed south west of Norway in 2009. Statoil is a Norwegian oil company which has previous experience in the floating oil platform industry, which has been used when designing the wind turbine. The turbine is designed for >100m depths and has the spar-bouy type of mooring (as seen in Figure 1). There are currently a pilot park planned in Scotland which will consist of five 6 MW wind turbines.[10] The prototype has so far had a capacity factor of 50%, which is very high. Fukushima Forward is a Japanese project which will demonstrate and test three different turbine designs. As of today two turbines are installed, a 2 MW semisubmersible and a 7 MW spar-bouy, with a third turbine planned for 2016. These will be evaluated for stability, efficiency and environmental impact in preparation for bigger projects in the 2020s. [1] The second turbine to be installed in the world was Windfloat, which is a 2 MW semi-submersible installed outside of Portugal. A pilot park using eight 6 MW turbines is also planned in Kincardine outside of Scotland.

3.4

Offshore substation

The substation is an important part of any wind farm that is located far from the grid. It is used to transform the voltage level of the wind farm, which is usually in the medium range of 35 kV, to a high voltage, either as HVAC or HVDC. This reduces the transmission losses for the wind farm. These substations are very big and expensive. There are many fixed-foundation examples, for instance Dolwin 1, 2 and 3, Borwin 1 and 2 and Helwin 1 and 2. They have a capacity between 100 and 1000 MW, depending on the size of the wind farms they are connected to. Figure 2:

Dolwin2 being before installa-

The biggest one currently in use is Dolwin beta (or transported tion. Dolwin2), which can be seen in Figure 2. It has 916 MW capacity and its outside measurements are 100 ∗ 74 ∗ 100 (side∗side∗height), and together with its substructure it weights 23000 tons. [11] The size and weight of these substations can be considered to grow in a linear fashion with its power. This means that a substation with a capacity of 1800 MW

14

Offshore wind

Nils Jonsson Forsblad

could weigh in excess of 40 000 tons for a bottom fixed substation. [12] In the case of floating offshore wind, these substations would also need to float, and would therefore require a substructure like one used in the offshore oil industry which would reduce the weight and size needed considerably (30 meters shorter and 20-30% lighter). There are oil platforms that weigh more, which means that it is technically possible even though none have been constructed today.

3.5

Cost Reduction

There is a big potential for cost reduction in the offshore wind industry, and several organizations have looked in to cost reduction pathways. These give a general direction for the industry. Table 2: Cost reduction potential from floating offshore wind prototype to commercial farm according to Carbon Trust. [1]

Reduction

Platform

Moorings

Anchors

Installation

Turbine

Balance of system

16%

2%

2%

5%

12%

9%

Table 2 shows the potential cost reduction for floating offshore wind from prototype to commercial deployments. This gives an overall cost reduction of almost 40 %, which would greatly increase the competitiveness of the technology. [1] Since most of the parts for traditional offshore wind is still used for floating offshore wind, the cost reduction potential for that can be used as a estimate for floating offshore wind. The offshore wind industry have made an estimate for offshore wind at a total CAPEX cost reduction of 25 % by 2020 because of technology development and an additional 15 % thanks to increased volumes. This is based on cooperation from many different stakeholders in the offshore wind industry, and a continued growth of the offshore wind market. [13]

15

Transmission

4

Nils Jonsson Forsblad

Transmission The power generated in any offshore wind farm need to be transmitted to the national power grid to be of use to the country, this is usually done by either a HVAC or a HVDC cable. The output from all of the turbines are collected in a central substation, which in turn is connected to a substation onshore.

There are different pros and cons of each technology, but it usually comes down to Figure 3: Comparison of losses for HVAC price. The substations are more expensive and HVDC. [14] for HVDC-technology, but the losses in the cables are lower. This creates a break-even point for when HVAC or HVDC is more economically viable, as shown in figure 3. The technologies are constantly evolving and there is no exact point for this fact, but it is generally considered that HVDC is the better choice at cables longer than 70 km. [14] In the case of connecting two or more wind farm areas to a single export cable, there could be needed to first collect the electricity by using HVAC technology, and convert it to HVDC at a central collection point between the areas, at a HVDC substation. This has been done in a couple of offshore wind farms, for example Dolwin 1 and 2. [11] This is one of the considered scenarios, namely Scenario BC which can be found in the section ”Scenarios”. One other possibility is to build a tee-connection to an already planned HVDC connection, where a wind farm and substation would be connected to the cable. In this case the cost of the extra length of cable and one offshore substation is all that is required, and it could be beneficial if the wind farm is far from shore but close to a planned HVDC cable. There are two examples of this in Skagerrak, NorwayJylland and Jylland - Sweden are connected by HVDC VSC technology, and these connections could be expanded in the future. [15]

4.1

Electric grids

The cable (HVAC or HVDC) need to be connected to the power grid somewhere around Skagerrak. Since the cable and the accompanying substations is a significant part of the total cost of the project, the selected site should be as close to a connection point as possible. For a wind farm of more than 300 MW it should be connected

16

Transmission

Nils Jonsson Forsblad

to the 400 kV national grid, which is operated by the nations’ transmission system operators (TSO). This grid and its substations are shown in Figure 4. The cable and substation is expensive, so the cable should be kept as short as possible. If you connect the different national grids you also enable trading between the nations. Sweden - Norway and Norway - Denmark already have a big trading capacity over their borders, Sweden and Norway are connected on many points along their border which makes transmission of thousands of MW possible and Denmark and Norway are connected by four HVDC cables with a total transmission capacity of 1700 MW. Sweden and Denmark are connected by a 700 MW cable Jylland-Sweden. A connection between Sweden and Norway would however connect two previously unconnected areas (SE3 Figure 4: The Swedish, Norwegian and Danish and NO2), which could make trade national 400 kV grid and substations. [16] between the two beneficial.

4.2

Trade between SE3 and NO2

A rough estimate of the potential trade between the two areas NO2 in Norway and SE3 in Sweden has been made. Energy trading on the Nordic electricity market is handled by Nordpool [17] and historical price information can be obtained from them. The data here is based on full year hourly price data from 2013-2015, and the first three months of 2016. Trade would only happen if the difference in price between the two areas were more than 2 %, because the rest would be transmission losses, and that the cable would always be available to transmit 600 MW. A cable of this size has the potential to change the electricity price in the regions by averaging them out, but that is not considered here.

17

Transmission

Nils Jonsson Forsblad

In the years 2013- March 2016 the prices differed by more than 2% almost 44% of the time, which resulted in a total trade value of 56 million e, as shown in Table 3. Table 3: Shows the number of hours each year that the price difference in SE3 and NO2 are more than 2%, and the amount of available trade between them.

600 MW Cable

Hours [price-diff >2%]

Percentage of the year

total trade [MEUR]

2013 2014 2015 Jan-April 2016

4341 5034 2742 315

50% 57% 31% 14%

17,72 23,38 12,26 2,58

Total

12432

43,6%

55,94

To get an estimate of the cost of such a cable and its onshore substations other similar projects can be used. The NORDBALT cable has a total cost of 500 million e, it has a capacity of 700 MW and the cable is 400 km subsea and 50 km onshore. The cost of the onshore substations are estimated to be 300 million e, which means that the cost of the cable is 0,49 million e/km. A rough estimate for the cable connecting SE3 and NO2 would be around 440 million e, which would give it a payback time of 26 years. The life time of a HVDC cable is usually considered to be at least 30 years, with a high probability it will last longer than 50 years. This is a very simplified payback calculation, to see the feasibility of the cable. With any type of discount factor the payback time for the cable would be longer than its lifetime. There are however more benefits to increasing the transfer capacity between two nations that are not economic, like increased stability and the ability to use intermittent power more effectively. The power that is generated by the wind farm could also be sold in the market with the highest price. The fact that Norway is renewing their hydropower plants by expanding their output capacity, and Sweden is closing some of its nuclear power in the near future, trade between the nations will likely increase, something that a cable between these areas can accommodate. The cable would also provide power further south in Sweden, where power production is lacking. [12]

18

Economics of an offshore wind farm

5

Nils Jonsson Forsblad

Economics of an offshore wind farm

When considering the economics of an offshore wind farm, there are many different factors to take into account. As with most large-scale power generation projects, the initial cost can be very high. This makes it necessary to consider the project from a life cycle perspective, to accurately determine the cost of the energy created. The cost of different parts of a wind farm used in this thesis are taken from many different sources (see [9][18][19]) to get an overview of the cost. The information is hard to come by and often confidential, so numbers are presented as either a percentage or a rough estimate.

5.1

Capital Expenditures (CAPEX)

Figure 5: The distribution of CAPEX for the 600 MW reference wind farm.[18]

Capital expenditure (CAPEX) is a measure of the total investments needed to be made in the beginning of a project. This covers everything from planning to construction and installation. For the reference park (explained in detail in the section ”Scenarios”) of a 600 MW wind farm, the division of the cost can be seen in Figure 5. The two biggest expenditures are the foundation (anchor, mooring, substructure) and the turbine itself, which account for more than 75% of the total cost in the reference park. When the transmission is made with HVDC instead of HVAC, electric infrastructure is a slightly larger part of the investment. The total cost for a 600 MW (100 turbines) wind farm is over 3 billion e.

19

Economics of an offshore wind farm

5.1.1

Nils Jonsson Forsblad

Costs of substructures

Table 4: Production cost estimate for the floating substructures for a 5 MW turbine. [20]

Material Consumption [Tons] Material Cost [ke] Manufacturing Complexity Factor Manufacturing Cost [ke] Total production cost [ke]

TLB

Hywind

Windfloat

445 445 110% 489.5 934.5

1700 1700 120% 2040 3740

2500 2500 200% 5000 7500

The substructures for floating offshore wind vary in cost, mostly dependent on construction complexity and weight of the substructure. The TLB type of substructure use less steel than any other type and multi turbine platforms use the most. The price of the substructure will depend on the price of steel, which multi-turbine platforms will be the most sensetive to. The complexity factor shown in Table 4 is a measure of how expensive the production process is expected when mass manufacturing will be available for the technology, this also influence the price. 5.1.2

Cost of Mooring

Table 5: Cost of mooring for some floating turbine concepts, assuming medium clay seabed, 200 meter depth. [20]

Concept

TLB Hywind

Windfloat

Total Line Cost [ ke] Anchor Cost [ke]

874 1042

168.8 456

118.5 342

Mooring cost is both dependent on the chain or wire needed to hold the substructure in place, and of the anchor used. There are several different types for both, but in general the types of substructure that require tension in the line require both more expensive cables and anchors. The mooring cost is between 5 % - 17 % of total CAPEX cost which is about 0.5-1.8 million e/ turbine. [20] 5.1.3

Cost of Electrical infrastructure

The electrical infrastructure is a big part of the investment cost for offshore wind. It generally requires three parts, an offshore substation, an onshore substation and a

20

Economics of an offshore wind farm

Nils Jonsson Forsblad

cable connecting the two. Both HVDC and HVAC will be used, and the costs differ greatly with HVDC being the more expensive one, but with lower energy losses. To get an understanding of the price ranges there are some examples. the NORDBALT project connects the Swedish grid and the grid of Lithuania, it has a total length of 450 km (400 subsea 50 onshore) HVDC cable and two onshore substations, all rated at 700 MW. It has a total cost of 552 million e. [21] The estimated production cost distribution are around 300 million efor the substations and 250 for the cables, which gives a cost at 550 000 e/km, and around 214 000 e/MW for the onshore substations. These substations would be even more expensive if they were offshore, up to three times as expensive per MW according to some estimates. [18]

5.2

Operating Expenditures (OPEX)

The OPEX is the cost of maintaining the wind farm over the years. It is assumed to be about the same each year, and consists mainly of maintenance, service vessels, repair staff and major replacements. The total OPEX cost is usually around 30 % of the total lifetime cost of the wind farm. [9]

5.3

LCOE

To calculate the total cost of energy of the wind farm, the LCOE is calculated. This is simply based on the cost of energy production, and does not take variable income such as electricity price or subsidies into account, which makes the calculation easier. It avoids the need for impossible predictions like ”what is the electricity price in 15 years?” or ”how long will subsidies last?”. It is a measure of the life cycle cost of energy and is calculated by: Pn

Ct +Ot t=0 (1+r)t Et t=0 (1+r)t

LCOE = Pn

where Ct denotes CAPEX at time t, Ot the OPEX at time t, Et the energy generation at time t, r the discount rate and t denotes the time, ranging from zero to n. The discount rate is a reflection of the project risk and the market value of both equity and debt. This also represents the difference in value with both energy and money created in the future compared to today. This rate varies between different projects, and the rate used in the LCOE calculations here will be 9%, which is based on the rate which Vattenfall uses in their calculations for offshore wind. The high

21

Economics of an offshore wind farm

Nils Jonsson Forsblad

and low cases looked at in the sensitivity analysis will use 6% as a low rate and 12% as high (based on onshore wind, and offshore wind in other countries). LCOE is very useful in that it can be used to compare different energy sources with each other strictly financially speaking. It does not however take other important factors such as social acceptance, air pollution or carbon emissions in to consideration. An attempt to remedy this has been made by measuring Societal Cost of Energy, or SCOE. Siemens has made an example of this and lowered the ”cost” of energy from offshore wind from 140 e/MWh in 2013 to 60 e/MWh in 2025, by taking these factors in to account. [22]

22

Skagerrak

6

Nils Jonsson Forsblad

Skagerrak

Skagerrak is a part of the North Sea, which is located between Sweden, Denmark and Norway. There is much activity in the area from shipping lanes, fisheries, the public and the military. Much is needed to be taken into consideration when evaluating the possibility of building a offshore wind farm.

6.1

Natura 2000, national nature reserves.

There are several nature reserves and nature protection areas in and around Skagerrak. The map in Figure 6 shows the nature reserves and Natura 2000 areas of Sweden, Denmark and Norway in the vicinity of Skagerrak. The obstruction of the view is also an important factor to take in to consideration. The area covered by Natura 2000 and other nature reserves is extensive in Skagerrak, especially the two central areas north of Denmark, named ”Bratten” and ”Skagens gren og Skagerrak”. The Natura 2000 network has provided a document to simplify what needs to be taken in to consideration when constructing in these areas. [24] Both are habitat protected areas, Skagens gren for its sandy bot- Figure 6: Natura 2000 areas and nature retom habitat and the protected species serves. [23] Porpoise and Bratten for it Gorgonians (a type of hard bottom coral). [23] Construction in these areas could be done, but their natura 2000 status makes it harder, so they are best avoided to make the process as simple as possible. A wind farm does however also act as an artificial reef, and creates a safe zone for fish as less commercial fishing can be conducted in the area. Each case in Swedish waters is evaluated individually by the Swedish Enviromental Protection Agency.

23

Skagerrak

6.2 6.2.1

Nils Jonsson Forsblad

Risk areas Mines and Dumping Sites

Skagerrak contains a few risk areas, after and during World war I and II, some areas mines were set and ammunition dumped. These areas are indicated by The Swedish Armed Forces and The Swedish Coast Guard and can be seen in Figure 7. Any work on the seabed in these areas must be conducted with extreme care, and should therefore be avoided. [25] [26] When fishing in these areas, there have been cases when fisheries get unexploded mines and pieces of old solidified mustard gas in their nets, which then have to be handled by the authorities. 6.2.2 area

Armed

forces

training

Figure 7: Swedish and Danish military exercise Another type of risk area are military areas in Skagerrak, and areas with mines and training areas, These areas are indi- dumped ammunition. [25] [26]

cated in Figure 7 and are training areas for the armed forces of both Sweden and Denmark. When there are exercises in these areas the respective military forces issue a firing warning, which indicates that you should not enter the area and flying above it is prohibited. According to an official from the Swedish Armed Forces, these areas have become fewer and fewer over the years, and both nations’ armies would most likely try to keep these areas dedicated as training areas, which means that construction here would not be possible.

24

Skagerrak

Nils Jonsson Forsblad

Figure 8: Traffic in Skagerrak in 2013, the different colors represent different types of ships. [27]

6.3

Shipping lanes

Shipping lanes and other shipping activities can be another challenge. The smaller lines in the map indicate all shipping activities, both commercial and private. This shows that the area is highly trafficked, but ships can also take many different routes. The only area in the sea that is indicated as a permanent sea route is marked marked in red with the text ”TSS off Ris¨or”, as shown in Figure 8 which means that that is the only place where construction would be impossible. This part of the ocean is however also very deep, so construction of wind farms here would be very expensive in any case. A wind farm claims a big area of the ocean, which means that it could severely hinder boating activites. For the most part it is possible for small ships to travel through a wind farm, but bigger ships are usually not allowed to do so. This causes a ”squeezing” of available ocean for bigger ships, and force them to gather closer to each other either further out to sea or closer to shore. Since Skagerrak has seen no wind farm development so far, this should not be a big problem in this location.

25

Skagerrak

6.4

Nils Jonsson Forsblad

Commercial Fishing.

There are areas indicated for commercial fishing in Skagerrak. These areas are indicated by the government agencies The Swedish Agency for Marine and Water Management for Sweden and Directory of Fisheries for Norway. Denmark does not currently have dedicated fishing zones, but are working to designate them according to the EU directive (214/89EU).There could be more areas designated in the years to come, but most likely only in the Danish economic zone of the sea. These economic zones are primarily intended for making sure that commercial fishing activities have priority over other activities when challenged, including the construction of offshore wind farms. Since the area covered Figure 9: Areas indicated for commercial fishby offshore wind is often very large ing. [28] [29] (up to1 km2 for each turbine) they can be very disruptive for other activities. The green lines in Figure 9 represent commercial fishing acitivies, which also indicates that many of the fisheries utilize areas outside of the official fishing zones, but since these areas are not dedicated fishing zones, it is assumed that commercial fishing will not get priority over other activities in this area.

26

Skagerrak

Nils Jonsson Forsblad

Figure 11: Material on the seabed, based on a different source for each nation. [30] [31] [32]

Figure 10: Depth map of Skagerrak, based on a map from HELCOM (Helsinki Commission). [27]

6.5

Water depth & Seabed

Different designs of offshore wind farms are optimized for different depths. To indicate which areas are best for which, a depth map has been provided by HELCOM. The depths vary from 0-500m. The seabed of Skagerrak is mostly well documented, and both Norway and Denmark have good coverage of open source data, the Swedish is covered by a less precise Europe-spanning source, but they still indicate what type of seabed should be assumed in the Swedish area as well. The north,west and east parts of the ocean are covered by mud and sandy mud, and the south gradually go toward a more sand covered seabed. The type of seabed determines which type of anchor/mooring system that can be used when constructing in these areas. A seabed made primarily of medium clay is the one which is most preferable to the floating wind industry because they can use the cheapest (and easily recoverable) type of anchor, more on this in the anchor subsection of the technology section.

27

Skagerrak

6.6

Nils Jonsson Forsblad

Wind conditions

Figure 12: Wind speed map of Skagerrak provided by Kjeller Vindteknikk. [33]

One of the most important factors when considering a wind farm are the wind conditions. The speed is generally more stable offshore, with higher average wind speeds as well. As shown in Figure 12 the wind speeds in Skagerrak vary from 8-10.5 m/s average at a height of 80 meters. The wind speeds were measured and compiled by Kjeller Vindteknikk. [33] The data from Kjeller Vindteknikk is a few years old, and new towers are usually 100 meters tall. The wind speed at slightly higher heights can be calculated with the formula V = V0 ( HH0 )m where V is the wind speed at the new height, H0 is the measured height, H is the new height and m depends on the roughness of the terrain, which on the ocean is 0.11. This means that if a measured wind speed at H0 = 80m is V = 9m/s, the wind would be 9.22 m/s at 100 m. (Henrik Stiesdal, Personal correspondence, May 2016) The capacity factor for a wind farm using Siemens 6 MW turbines can be calculated with the formula cap = 8.5% ∗ V − 22%. which means that a wind using 6 MW turbines and has an average wind speed of 8.5 m/s will get a capacity factor of 50%.

28

Skagerrak

6.7

Nils Jonsson Forsblad

Overview of Skagerrak

To get a better overview of the area all of the previously mentioned areas are added in a new map. The area which is not covered in Figure 13 is the most suitable area to construct a offshore wind farm. Figure 13 also shows the economic zones of the three Scandinavian nations, since they each have differing subsidies for constructing offshore. The area which is most interesting is middle northern part, between Sweden and Norway, shown in Figure 13. The bottom here is mud to sandy mud, but the data in the Swedish part of the ocean not very precise. The depth varies from 50 - 500 meters, with most of it being of 100-200 meters depth. Figure 13: All of the previously mentioned fac-

The wind speed for 80 m height is 7.5- tors put together, Natura 2000 in yellow and the 9 m/s and the estimated capacity fac- rest in red, usable area in green. tor for the area is calculated to 0.36 0.4 by Kjeller Vindteknikk in 2009 [33]. The capacity factor and wind speeds that are considered here are at 100 meters, it is calculated for the new height. The capacity factor is also dependent on technology and since 7 years have passed since their study and wind turbine technology is evolving rapidly, is also a fair bit higher.

29

Result

7

Nils Jonsson Forsblad

Result

From the previous sections on technology, transmission and Skagerrak many different scenarios can be created for the location of the wind farm and its overall production and cost. The area considered for this thesis is places where depth is 100-200 m, and not covered by any of the other areas mentioned in the section ”Skagerrak”. This sums up to 1868 km2 and is indicated in green in Figure 13, and shown in Table 6. This information show several suitable locations for big wind farms, of the 100 turbine size. These are presented in Figure 14 as the wind farms A - E. Some of these can be scaled up to 200 turbines, and area F is smaller and can only accommodate 83 turbines. All of the areas are located in Figure 14: The locations of the different wind farm scenarios and their HVAC/DC cabes. only one Nations economic zone. A 6 MW wind farm would require approximately 1 km2 in area, based on the wingspan of the turbines (120 m) and the recommended 7-10 wingspan between the turbines. To keep the losses in the wind farm low, the cables connecting them all to a substation should be made as short as possible, therefore the parks are made as circular as possible. Table 6: The usable area in Skagerrak in km2 , based on Figure 13.

Usable area Sweden Area [km2 ]

7.1

Norway

Denmark

Total

806

483

1868

579

Reference park

The ”reference park” is based on a source given from DNV GL [18]. This data is used as a base because of its reliability, and its similarity to many of the scenarios

30

Result

Nils Jonsson Forsblad

listed below. The reference wind farm is of the spar-buoy type, with drag driven anchors and 100x6 MW turbines. The average depth in the reference park is 120 meters, which is lower than in the scenarios, and the mooring chain cost is adjusted accordingly. The distance to shore is 60 km, and HVAC technology is used. The CAPEX division can be seen in Figure 5, and most of the cases below have a similar CAPEX with the exception of park BC, which has a significantly higher electrical infrastructure cost. The discount rate, as mentioned in the LCOE section, is 9% for all of the cases. Other important factors such as losses and capacity factor is different between the areas. The capacity factor for the parks is calculated with the formula mentioned in section 5.6, but it is also lowered because of different losses, namely: • 5% availability loss • 7% wake loss • 2% grid loss • 1% high wind loss for a total of 14% lower capacity factor. A wind farm with an average wind speed of 9 m/s would have a capacity factor of 46.3% for example.

31

Result

7.2

Nils Jonsson Forsblad

Park A

This scenario uses area A which is shown in Figure 14. This location was chosen because of its proximity to the Swedish power grid. The location has a average depth of 150 meters, and is connected to the grid by a 69 km long HVAC cable. The average wind speed in the area is 9 m/s according to the map made by Kjeller Vindteknikk [33], this makes the assumed capacity factor for the whole park 46.4% . This mens the farm produces an average of 2440 GWh yearly.

Figure 15: Schematic picture of park A.

The LCOE cost for energy generated here will be 187 e/ MWh with data according to table 7. Table 7: Data for Park A.

PARK A

Avg depth

Wind Speed

Substations

Distance to grid

Capacity factor

600 MW

150 m

9 m/s

2

69 km

46,3%

7.3

Park D

D is similar to A, but its location make a bigger farm possible, up to 1200 MW. A bigger part means a lower CAPEX and OPEX per MW, which could lower the LCOE for the wind farm. The location closer to shore and further north makes the wind speeds slightly lower, which also lowers the capacity factor of the farm. The shorter cable makes the losses Figure 16: Schematic picture of park D. slightly lower. The higher average depth negates the CAPEX saved by shorter cable though. The short distance means that similarly to park A it uses

32

Result

Nils Jonsson Forsblad

HVAC cables. The LCOE for park D is 203 e/MWh for a 600 MW wind farm, and 176 e/MWh for 1200 MW based on the information from Table 8. Table 8: Data for park D

PARK D

Depth

Wind Speed

Substation

Cable length

Capacity factor

600-1200 MW

180 m

8.5 m/s

2

45 km

42.7%

7.4

Park E

Park E is located north of Denmark, and can support a 600-1200 MW wind farm. The seabed is much steeper, which means the area is wider and array cables within the park are longer. It is located 90 km from shore, which means that losses because of the distance is higher, and it is in the border region where a study is needed to determine whether HVAC or HVDC is the technology is more viable economically.

Figure 17: Schematic picture of park E.

Which transmission technology that should be used here is hard to determine, and many different sources claim differing conclusions in the matter. [34] [19] The western location of this park gives it the highest capacity factor of all farms considered, and will generate approximately 2370 GWh yearly for a 600 MW farm. The LCOE for this wind farm is 169 e/MWh for a 600 MW and 149 e/MWh for a 1200 MW wind farm, based on information in Table 9. The location in Danish water does however pose a problem. They already have the most offshore wind power of the Scandinavian nations, and some of them are located close to park E. They are yet to determine fishing areas, as mentioned in the section ”Commercial Fishing”. This area could fall under that category in the future. The Danish government have also halted construction of many new offshore wind farms recently.

33

Result

Nils Jonsson Forsblad

Table 9: Data for park E

PARK E

Depth

Wind Speed

Substation

Cable length

Capacity factor

600-1200 MW

150 m

9,7 m/s

2

90 km

51.6%

7.5

Park F

The wind Park F connects the power grids of Sweden and Denmark, enabling more trade between the nations and the electricity generated by the farm to be sold in the more expensive of the two markets. The limited area for use makes the size of this park smaller, at 83 turbines or 498 MW, generating an average of 2220 GWh yearly. This location of a wind farm is close Figure 18: Schematic picture of park F. to an already existing HVDC cable between the two nations, and this fact is used in this scenario. If the export capacity of this connection were to be expanded with another cable, a detour of 71 km (for a total length of 162 km) would enable a wind farm to use a T-connection to the cable. This would mean that the farm only need to fund one substation and 71 km of cable. This type of tee-connection is currently not in use anywhere, and this one could be a proof of concept for the technology. Since Swedens’ and Jyllands’ (part of Denmark) power grids are not synchronized, a HVDC cable is needed. This is more expensive but has lower losses than its HVAC counterpart. The LCOE for park F is calculated to 175 e/MWh according to the data in Table 10. Table 10: Data for park F

PARK F

Depth

Wind Speed

Substation

Cable length

Capacity factor

498 MW

150 m

9.3 m/s

1

71 km

42%

34

Result

7.6

Nils Jonsson Forsblad

Park BC

Figure 19: Schematic picture of parks B and C.

Park BC combines two different parks and can therefore have a higher capacity, up to 1800 MW. It would also connect Norway and Sweden via HVDC cable, which would connect two previously unconnected electricity trading areas. This is discussed in detail in the section ”Trade between NO2 and SE3”. This would also enable the electricity to be sold in the more expensive market, whichever that is at the time. In this scenario, the HVDC substation is located in park B, with a HVAC substation and cable connecting park C to it. The HVDC substation itself uses a T-connection like the one mentioned in Park F to connect it to the HVDC cable between Norway and Sweden. In this case however, it is not assumed that the onshore substations and some length of the cable is paid for by other interests, this means that the costs for electric infrastructure is considerably higher in this scenario. Park B has a capacity factor of 46.4% and park C 44.2%, the depth is an average of 150 m at site B, and 180 m at site C. This gives an LCOE of 244 e/MWh for 1200 MW, and 223 e/MWh for 1800 MW. Table 11: Data for park BC

PARK B

Depth

Wind Speed

Substation

Cable length

Capacity factor

600-1200 MW

150 m

9 m/s

3

143 km

44.5%

180 m

9 m/s

1

17 km

39.5%

PARK C 600 MW

The Swedish nuclear power plant Ringhals is planned to close in the next couple of years, this will make it necessary to install new power production in the southern

35

Result

Nils Jonsson Forsblad

parts of Sweden. One solution to this could be a HVDC cable to Norway, who is currently expanding its hydroelectric power. In combination to this, a tee-in connection like the one discussed in ”Park F” could be used with wind farm BC as the tee-in park. When the wind isn’t blowing Norwegian power production could produce the power. When this scenario is considered similar assumptions as the ones made in Park F applies - some of the cable cost and all of the onshore substation costs are assumed to be paid by other stakeholders. This increases the profitability of the wind farm considerably, giving a LCOE cost of 222 e/MWh for the 1200 MW scenario, and 200 e/MWh for 1800 MW.

36

Result

7.7 7.7.1

Nils Jonsson Forsblad

Sensitivity analysis Discount factor

Figure 20: Changing the discount rate when calculating LCOE to 6%, 9% and 12 %.

The discount factor when calculating LCOE is very important, and is based on many different factors (like equity, risk, losses, and so on). To look closer at the impact of this it has been varied by +/- 3%, and the result can be seen in Figure 20. The impact on the resulting LCOE is about -17 to -19% when the discount rate is lowered to 6% and +19 to 21% when it is raised to 12%. This shows how sensitive the profitability is to the perceived risk of the investment.

37

Result

7.7.2

Nils Jonsson Forsblad

Lifetime

Figure 21: Adding 5 years to the lifespan of the parks, also with varying discount factor.

The life time of a wind farm is mostly a economic choice made by the constructors. 20 years have been chosen partly because of the rapid evolution of the wind turbine, which means that after 20 years, the technology used is so old that it would be better to simply replace them rather than continue repairing them. As the technology have aged, and turbines and investments grown, they may get a longer lifespan. The result of +/- 5 year life span is shown in Figure 21, and the discount factor is also varied between 6% - 12 %. The percentage difference is shown in Table 12, and is around 4.8 - 5.3 % for all of the cases. Table 12: The percentage difference in LCOE (9% discount rate) between a wind farm working for 20 and 25 years.

LCOE change

Park A

Park BC

Park D

Park E

Park F

-5.3%

-4.8%

-5.3%

-5.3%

-5.3%

38

Result

7.7.3

Nils Jonsson Forsblad

Capacity factor

Figure 22: Varying capacity factor by +/- 5% for all of the cases above.

The capacity factor is very important for the profitability of a wind farm, as is shown in Figure 22. It is therefore important to determine what it will be when considering a location. The values for it used in this thesis is based on wind speed measurements by Kjeller Vindteknikk [33], and then calculated according to a formula mentioned in section ”Wind speed” under Skagerrak. (Henrik Stiesdal, Personal correspondence, May 2016) Because of this uncertainty they are varied by +/- 5% and the result can be seen in Figure 22. The LCOE is lowered by around 10% in all cases when the capacity is raised by 5%, and raised by 8.6-13.2% when lowered by 5%. This shows that an overestimation or underestimation of the capacity factor changes the apparent profitability of the wind farm by a big margin.

39

Result

7.8

Nils Jonsson Forsblad

Cost Compression result

As shown in the section on cost compression earlier, the industry believes in big cost compression in the future because of maturing technology. 40% reduction by 2020 is envisaged by the industry, and similar numbers for floating offshore wind by Carbon Trust. The CAPEX cost has been reduced according to the values that carbon trust hope to achieve, which can be seen in Table 2. This gives a total reduction of CAPEX for the reference park at almost 40%, and when applied to all of the different parks mentioned in the previous section they get an LCOE according to Table 13. Table 13: The new LCOE when the cost compression envisaged by Carbon Trust and DNV GL has been applied to the CAPEX of park A-F.

[e/MWh]

Park A

Park D

Park E

600 MW 1200 MW

125 -

136 117

113 99

40

Park F

Park BC

117 (498 MW) 171 (1200 MW) 139 (1800 MW)

Discussion and Conclusion

8

Nils Jonsson Forsblad

Discussion and Conclusion

As shown by the results and sensitivity analysis, building floating offshore wind today will be very expensive, and the LCOE cost will be 149 e/MWh at the lowest with 9 % discount rate. This can be compared to other new energy sources, for example onshore wind which have a LCOE as low as 50 e/MWh. This technology is however very new, and the subsequent technology and increases in effectiveness have a potential to lower the CAPEX costs by 40%. Since CAPEX is such an important factor when calculating LCOE, this would lower the LCOE cost considerably, as shown in the cost compression sensitivity analysis. The value of building this wind farm can not only be considered in economic terms of the wind farm. This is a new industry which could grow considerably in the future. As of today Europe is the only part of the world with any offshore wind to speak of. This is partly because of the big investments in renewable technology made there, but also because of the favorable conditions with many suitable locations that aren’t too deep, unlike many other parts of the world. The societal benefit of more wind power is not measured by LCOE either, but it can be measured with Societal Cost of Energy (SCOE). This tries to take social factors in to consideration when evaluating power sources, factors like pollution, societal costs, geopolitical risks and employment are also important. Since offshore wind does not require any fuel, has no pollution and cannot be seen from land (if built far enough offshore), the cost to society is very low. Floating offshore wind would make construction in depths up to 900 m possible, which means that the offshore industry as a whole can grow considerably, and big nations such as India, Japan, USA and China can build in many more locations, since most of the ocean is unavailable to bottom fixed offshore wind. This also removes one big bottleneck of offshore construction today - the highly specialized ships that are needed for erecting the turbines. These ships can add years in lead time since there are few, and are often a bottleneck in the installation process. This also means that this industry could go the way of Danish bottom fixed foundation offshore wind. Denmark built many of the early offshore wind farms and are now one of the biggest players in the bottom fixed foundation offshore wind field, which increases exports and employs many people there. As shown in the case of F and BC, offshore wind farms can be built as tee-in connections to international HVDC cables that are planned. This increases the profitability of the wind farm (and lowers LCOE), since part of the investment is made by other stakeholders. The generated power can be sold in whichever nation

41

Discussion and Conclusion

Nils Jonsson Forsblad

has the higher price, and when the wind isn’t blowing, other power sources can make sure that the cable is still going at full capacity, potentially fixing the problem of intermittent power production from the wind farm. This also has the positive effect of making the power production more stable from the perspective of the energy grid. This could also remedy the fact that the big nuclear power plant Ringhals is closing soon in Sweden, and new production will be needed in the southern part of the nation. This also has the potential to avoid otherwise costly investments in infrastructure. The main point of building offshore wind power is ofcourse that it has one of the lowest carbon footprints of all available technologies, and carbon emissions need to be drastically reduced, the European Union for example has a goal that 97% of power production need to be renewable before 2050. A well developed floating offshore wind industry in Scandinavia would enable a big expansion of offshore wind in the developing world that are currently expanding their power production the most, and sadly mostly by using fossil fuels.

42

REFERENCES

Nils Jonsson Forsblad

References [1] R. James and M. Costa Ros. Floating offshore wind: Market and technology review. June 2015. Carbon Trust. [2] Intergovernmental Panel On Climate Change. Climate change 2007: Impacts, adaptation and vulnerability. Genebra, Su´ı¸ca, 2007. [3] A. Corbetta, G. Mbistrova and A. Ho. Wind in power 2015 european statistics. February 2016. Fuente: http://www.ewea.org/fileadmin/files/library/publications/ statistics/EWEA-Annual-Statistics-2015.pdf. [4] Nicolas Boccard. Capacity factor of wind power realized values vs. estimates. energy policy, 37(7):2679–2688, 2009. [5] Andrew. Capacity factors at danish offshore wind farms, 2016. http://energynumbers. info/capacity-factors-at-danish-offshore-wind-farms, Accessed 2016-05-10. [6] Variablepitch. Capacity factor offshore wind. http://www.variablepitch.co.uk/output/ off_shore_wind/. Accessed: 2016-05-15. [7] Hexicon offshore floating platforms. http://www.hexicon.eu/offshore-platform/ our-solutions.html. Accessed: 2016-04-26. [8] Det Norske Veritas. Anchoring the future offshore power. December 2013. [9] Anders Myhr, Catho Bjerkseter, Anders ˚ Agotnes, and Tor A Nygaard. Levelised cost of energy for offshore floating wind turbines in a life cycle perspective. Renewable Energy, 66:714–728, 2014. [10] Finn Gunnar Nielsen. Hywind. deep offshore wind operational experience. In 10th Deep Sea Offshore Wind R&D Conference, 2013. [11] ABB. Dolwin reference. http://new.abb.com/systems/hvdc/references/dolwin2. Accessed: 2016-05-06. [12] M. W¨ amundson. Gothia Power E. Isabegovic. Interview with employees from gothia power. Personal Correspondence, May 2016. [13] DNV GL. Offshore wind, a manifesto for cost reduction. 2013. [14] Thomas Ackermann. Wind Power in Power Systems. Wiley, 2 edition, 2012. [15] Jan De Decker, Paul Kreutzkamp, S Cowdroy, L Warland, J V¨olker, J Tambke, et al. Offshore electricity grid infrastructure in europe. OffshoreGrid Final Report, 2011. [16] Svenska kraftn¨ at. Sveriges stamn¨at. stamnatskarta/. Accessed: 2016-04-07.

http://www.svk.se/drift-av-stamnatet/

[17] Nordpool trading areas. http://www.nordpoolspot.com/How-does-it-work/ Bidding-areas/. Accessed: 2016-05-06.

43

REFERENCES

Nils Jonsson Forsblad

[18] Den Norske Veritas M. Ebbesen. Cost of an offshore windfarm. Personal Correspondence, March-April 2016. [19] Bram Van Eeckhout. The economic value of vsc hvdc compared to hvac for offshore wind farms. Offshore (Conroe, TX), 2008. [20] Catho Bjerkseter and Anders ˚ Agotnes. Levelised costs of energy for offshore floating wind turbine concepts. 2013. [21] Svenska Kraftn¨ at. Om projektet - nordbalt. http://www.svk.se/natutveckling/ utbyggnadsprojekt/nordbalt/om-projektet/. Accessed: 2016-05-30. [22] Siemens. Social cost of electricity (scoe). http://www.energy.siemens.com/hq/en/ renewable-energy/wind-power/SCOE.htm. Accessed: 2016-05-15. [23] Natura 2000 areas. http://ec.europa.eu/environment/nature/natura2000/index_en. htm. Accessed: 2016-03-07. [24] 2011 Publications Office of the European Union. Wind energy developments and natura 2000. http://ec.europa.eu/environment/nature/natura2000/management/docs/Wind_ farms.pdf. Accessed: 2016-03-07. [25] F¨ orsvarsmakten (eng: The Swedish Armed Forces). Riskomr˚ aden. http://www. sjofartsverket.se/sv/Sjofart/Sjokort/Ufs---Underrattelser-for-sjofarande-/ Minor/Riskomraden/. Accessed: 2016-03-07. [26] Kustbevakningen (eng: The Swedish Coast Guard). Riskomr˚ aden. https://www.kustbevakningen.se/hallbar-havsmiljo/miljoraddning/andra-skadligaamnen/oexploderad-ammunition-till-sjoss/riskomraden/. Accessed: 2016-03-11. [27] HELCOM. Baltic sea data and map service. http://maps.helcom.fi/website/mapservice/ index.html. Accessed: 2016-03-08. [28] Fiskeridirektoratet. Fishing zones. http://kart.kystverket.no/. Accessed: 2016-03-11. [29] Havs och vattenmyndigheten. Fishing zones. https://www.havochvatten.se/ hav/samordning--fakta/kartor--gis/karttjanster-fran-hav/karttjanster/ riksintresse-for-yrkesfisket.html. Accessed: 2016-03-11. [30] D. et al. Ottesen. Geological atlas of the southern part of the norwegian trench. NGU Rapport 2000.104, 2000. [31] Publications Office of the European Union. Wind energy developments and natura 2000. http://ec.europa.eu/environment/nature/natura2000/management/docs/Wind_ farms.pdf, 2011. Accessed: 2016-03-07. [32] 2011 Publications Office of the European Union. Wind energy developments and natura 2000. http://ec.europa.eu/environment/nature/natura2000/management/docs/Wind_ farms.pdf. Accessed: 2016-03-07. [33] Kjeller Vindteknikk. Wind data. http://www.vindteknikk.se/, 2013. [34] D´esir´ee Moberg. Transmission alternatives for grid connection of large offshore wind farms at large distance. 2013. Umea Universitet, Master Thesis [internet].

44