On the Market Potential of Modular Stirling CSP Systems With Storage in the MENA

Master of Science Thesis

KTH School of Industrial Engineering and Management Energy Technology EGI- TRITA-ITM-EX 2018:612

Division of Heat and Power Technology SE-100 44 STOCKHOLM

On the Market Potential of

Modular Stirling CSP Systems

With Storage in the MENA

-II-

Master of Science Thesis TRITA-ITM-EX 2018:612

On the Market Potential of Modular Stirling CSP Systems With Storage in the MENA

Youssef Benmakhlouf Andaloussi

Approved Examiner

Björn Laumert

Supervisor

Rafael Guédez

Commissioner Contact person

Abstract

Given the intermittent nature of renewable energy sources, integrated storage solutions are necessary to accomplish the energy shift necessary for sustainable development. In the case of solar, PV-BESS tend to be highly capital intensive, especially for long storage hours most needed to guarantee stable electricity production day and night. This study presents a methodology to quantify the market potential for a novel distributed CSP technology with cost competitive thermal energy technology, where the cost target is 30% cheaper than PV-BESS. The system in question is similar to the one developed by Cleanergy AB, where a 13 kW Stirling engine is powered by heat collected from a heliostat field and stored in an integrated latent heat storage unit. Morocco, Tunisia, Egypt, Jordan and Saudi Arabia are chosen as representative countries of the MENA for the study. The study is done by detailed investigation of the macro-environment of each country, developing a methodology to rank identified business opportunities. Said opportunities are restricted to companies within the industrial sector, based on the assumption that such customers would be interested in a solution guarantying stable electricity production. First, a techno-economical optimisation is done to find optimal plant configurations to service a particular energy need for each business opportunity. Second, the multi-criteria analysis scores and ranks the latter with respected to different criteria that can be conflicting. Finally, the top business opportunity identified by the MCA in each country are compared through a scenario analysis, assuming different rates at which the electricity generated by the system can be sold. With a global market potential above 40 GW in the whole MENA, industrial sectors such as mining and cement hold the best prospects in terms of market share. The achievable costs of generation vary depending on the DNI of the sites considered but prove to be lower compared with conventional distributed generation (diesel gensets or PV-BESS). However, several countries in the MENA, although having high DNI resource, still offer low electricity utility prices to industrial customers for distributed CSP to become competitive with on-grid electricity procurement. Hence, Jordan is ranked first with the MCA, both because of the high DN in the country, and its high electricity rates, despite having the smallest market share in terms of capacity to install. The amount of subsidies necessary for the technology to be profitable and cons competitive were found respectively. Except in Jordan where the system is competitive with utility rates, all other countries needs to implement feed-in-tariffs schemes for distributed CSP with storage to become viable. The observed trend of increasing electricity prices in the MENA however, coupled with decreasing LCOE values due to high volumes of production indicate that economic viability in the countries with low present rates can be achieved in the future.

-III-

Sammanfattning

Eftersom förnybara energikällor har en oförutsägbar energiproduktion krävs välutvecklade energilagringssystem för att samhället ska gå över till förnybara energi. Solenergi kräver PV-BESS, vilket tenderar att vara kapital intensivt, speciellt vid energilagring över lång tid som krävs för stabil energiproduktion under nattetid. Denna studie tar fram en metodologi för att kvantifiera marknadspotentialen för nya distribuerade CSP teknologier med termisk lagring. Kostnadsmålet för sådan termisk lagring är 30% lägre än för PV-BESS. Som exempel för CSP systemet används tekniken utvecklat av Cleanergy AB, vilket består av en 13 kW Stirlingmotor som är driven av hettan från ett heliostatfält och lagrat i en integrerad latent värmelagringsenhet. Marocko, Tunisien, Egypten, Jordanien och Saudi Arabien används för att representera länder från MENA i denna studie. Analysen består av en djupgående forskning av makroekonomiska faktorer som används för att identifiera och ranka affärsmöjligheter. Dessa affärsmöjligheter är begränsade till den industriella sektorn som kräver stabil energiproduktion. Först görs en teknologisk och ekonomisk optimering för att hitta den bästa konfiguringen av energianläggningen för kunden. För det andra poängterar multikriterieanalysen (MCA) och rankar kunderna med respekt för olika kriterier som kan vara motstridiga. Slutligen jämförs de bästa affärsmöjligheterna som identifierats av MCA i varje land genom en scenarioanalys, förutsatt att det är olika priser för elektricitet. Med en global marknadspotential på över 40 GW i hela MENA, har industrisektorer som gruv och cement de bästa utsikterna när det gäller marknadsandelar. De uppnådda LCOE varierar beroende på de undersökta platsernas DNI men är ändå lägre jämfört med alternativa distribuerad generation (dieselgeneratorer eller PV-BESS). Men flera länder i MENA , trots att de har en hög DNI-resurs, fortfarande erbjuda låga elverktygspriser till industrikunder för distribuerad CSP för att bli konkurrenskraftiga med elförsörjning på nätet. Därför rankas Jordan först med MCA, både på grund av den höga DN i landet och höga elpriser, trots att den minsta marknadsandelen. ängden subventioner som är nödvändiga för att tekniken ska vara lönsam och konkurrensbegränsad hittades. Förutom i Jordanien där systemet är konkurrenskraftigt med nyttjandepriser måste alla andra länder genomföra inmatningstullsystem för distribuerad CSP med lagring för att bli lönsam. Den observerade trenden med att öka elpriserna i MENA, i kombination med minskande LCOE-värden på grund av stora volymer av produktion tyder på att ekonomisk lönsamhet i länder med låga priser kan uppnås i framtiden.

-IV-

Acknowledgements

My first line of acknowledgments is bound to be addressed to Rafael Guédez, my supervisor, who played a central part in giving me the opportunities and experience I have today. Such small recognition cannot get right how grateful I am to him. To that effect, an additional appendix is needed in this report, so I can enumerate all the things he taught me. Naturally, Jonas Wallmander comes next in this thank you note, who directed me during the internship, and played a big role in the professional opportunities which came with it. Very much thanks also to Monika Topel, who always opened (literally) the door of her office to me. The laughter and good discussions we had there contributed a lot to this modest work. Special mention also to Osama Zaalouk in whom I found a precious ally against the Venezuelan mafia of the Energy department. Various reasons almost pushed me not to pursue this double degree master in KTH, but at the end, I am glad I went with it. Some of my closest friends now are people I met during these two years, and I am grateful to all one of them. Although Sweden is known for its cold and dark winters, my overall experience was one of warm memories. Finally, thanks to my family and mother most notably, whose unconditional love and words of wisdom will always resonate with me.

ةملك

ركش

نم

بجاولا

يلع

نأ

صصخأ

،ةرقف

ىتح

و

ول

تناك

،ةريغص

ركشل

يبأ

و

يمأ

.

نإف

مل

نكي

مهناكمإب

مهف

ام

يف

اذه

تحبلا

مكحب

،ةغللا

،مهلاولف

لاولو

مهمعد

مهتايحضتو

يف

،يتيبرت

امل

ناك

يناكمإب

،هرشن

وأ

ىتح

مهف

هلثم

.

كلذل

مهف

هوفلؤم

ردقب

ام

انأ

هكلام

.

لك

لضفلا

عجري

مهل

يف

يتناكم

،مويلا

يأوأ

ةناكم

يف

لبقتسملا

ىلإ

،يمأ

يتلا

تناك

ةقابسلا

يف

معد

يسردمت

يف

،اسنرف

يتللاو

تناك

يبناجب

يف

بعصأ

تاظحل

اذه

راوشملا

.

يف

لك

،ةرم

اهبح

عطق

لوطأ

تافاسملا

يتسناؤمل

ىلإ

،يبأ

يذلا

ينملع

ةيفيك

ربدتلا

يف

ةايحلا

تبثو

لك

تاوطخ

يتسارد

.

يمامتها

لاجملاب

يملعلا

عجري

،هيلإ

يتبغرل

يف

هلعج

اروخف

،يب

كلذو

دنم

،يرغص

نيح

تنك

لواحأ

لوصحلا

ىلع

ىلعأ

تاجردلا

يف

ةدام

تايضايرلا

هشاهدلإ

-V-

Abbreviations

AFEX Arab Future Energy Index BESS Battery Electricity Storage Systems CAPEX Capital Expenditure(s)

CSP Concentrated Solar Power CT Central Tower

DNI Direct Normal Irradiance DS Dish Stirling

GDP Gross Domestic Product GIS Geographic Information System HTF Heat Transfer Fluid

IPP Independent Power Producer

IRENA International Renewable Energy Agency IRR Internal Rate of Return

LCOE Levelized Cost of Electricity LF Linear Fresnel

MCA Multi Criteria Analysis MENA Middle East North Africa MGT Micro Gas Turbine NPV Net Present Value O&M Operation and Maintenance OPEX Operational Expenditure(s) PCM Phase Changing Material PT Parabolic Through PV Photovoltaic RE Renewable Energy RES Renewable Energy Systems SAM Serviceable Achievable Market SOM Serviceable Obtainable Market STEALS Investor-Owned Utility TAM Total Addressable Market TES Thermal Energy Storage

-VI-

List of figures

Figure 1 : Cleanergy Stirling dish demonstration plants, Dubai (right), Mongolia (left) [3] ... 2

Figure 2 DESERTEC project map. The red squares represent the area need for solar power plant to power the whole world, Europe and Germany [7] ... 4

Figure 3 PV installed capacity growth [10] ... 5

Figure 4 CSP installed capacity growth [10] ... 5

Figure 5 Flow diagram of a typical CSP plant [11] ... 6

Figure 6 CSP technologies [14] ... 6

Figure 7 CSP market trends [15] ... 8

Figure 8 247Solar Plant [18] ... 9

Figure 9 Vast Solar CSP system [19] ...10

Figure 10 Design configuration of STEALS [22]...11

Figure 11 Cleanergy's Alpha type Stirling engine [27] (adapted)...12

Figure 12 Cleanergy's initial target market (2021-2025) for the TES system design ...13

Figure 13 Model of one modular Cleanergy CSP unit with the three main components: concentrator, receiver with storage (10 hours) and Heat Engine (Stirling) ...14

Figure 14 PV-BESS LCOE in 2021 ...15

Figure 15 Market size estimation [34] ...16

Figure 16 Selection methodology for business opportunities [36] ...17

Figure 17 AFEX Renewable Energy 2016 [37] ...18

Figure 18 Techno-economical analysis process ...22

Figure 19 LCOE vs Reflective area ...23

Figure 20 NES 2030 targets [59] ...28

Figure 21 Electricity market Morocco [62] ...29

Figure 22 RE National Program 2017-2020, Tunisia [86] ...32

Figure 23 Electricity market, Tunisia [85]...32

Figure 24 Government power generation expansion plans [91] ...35

Figure 25 Egypt power market structure ...36

Figure 26 RE projects in Jordan 2016 [108] ...39

Figure 27 Jordan's electricity market [110] ...40

Figure 28 Long-term renewable energy targets, Saudi Arabia [117]...42

Figure 29 Power market structure, Saudi Arabia [25] ...43

Figure 30 RE private Investment Increase (2013-2016) [37] ...48

Figure 31 LCOE vs TES size (Cleanergy’s cost functions) ...49

Figure 32 LCOE vs TES size (STEALS cost data) ...50

Figure 33 Site positioning map, Morocco ...51

Figure 34 company positioning map MENA (Industry)...52

Figure 35 Scaling up the SAM to the MENA ...54

Figure 36 SAM in the MENA region by country ...55

Figure 37 Cleanergy's CAPEX breakdown ...55

Figure 38 LCOE sensitivity analysis ...56

Figure 39 NPV(k€) sensitivity analysis ...57

Figure 40 MCA country score (1-10) ...59

Figure 41 Multi Criteria Analysis – Ranking of business opportunities (1-10) ...60

Figure 42 Expected utility electricity price for industry 2021 (€/MWh) ...66

Figure 43 IRR vs Power price ...67

Figure 44 NPV vs Power price ...68

Figure 45 NPV vs Power price (zoom) ...68

Figure 46 Normalized LCOE evolution ...69

-VII- Figure 48 Case 2 ...99 Figure 49 Case 3 ...99 Figure 50 Case 4 ... 100 Figure 51 Case 5 ... 100 Figure 52 Case 6 ... 100 Figure 53 Case 7 ... 101 Figure 54 Case 8 ... 101

List of tables

Table 2 Entry modes categories [31] ...19

Table 3 Financial model inputs ...21

Table 4 Industry electricity rates ...25

Table 5 Attractiveness scoring table ...25

Table 6 Morocco generation units 2015 [58] ...27

Table 7 High voltage industry general rate, Morocco [51] ...30

Table 8 Identified industry companies, Morocco ...30

Table 9 Identified industry companies, Tunisia...33

Table 10 Identified industry companies, Egypt ...37

Table 11 Identified industry companies, Jordan...41

Table 12 Electricity rate, Saudi Arabia [121] [120] ...44

Table 13 Identified industry companies, Saudi Arabia ...45

Table 14 Countries Performance under International Indices [124] [125] [126] ...46

Table 15 Business model country comparison ...46

Table 16 Country score ranking ...48

Table 17 Market potential for the MENA (industry), with optimum configuration ...52

Table 18 SAM in the MENA, industry (grid connected, VHV-HV-MV) ...53

Table 19 Market potential for the MENA (industry), with optimum configuration using STEALS cost data ...56

Table 20 Most competitive business cases under the MCA (per country)...58

Table 21 Country score, by criterion (1-10) ...61

Table 22 Country score, additional criteria ...62

Table 23 Weighting factors case definition ...63

Table 24 MCA sensitivity (top 5 business opportunities) ...63

Table 25 Scenario analysis results, Morocco (MM31), WACC = 4,5% ...64

Table 26 Scenario analysis results, Tunisia (TC11), WACC = 4,8% ...64

Table 27 Scenario analysis results, Egypt (EM21), WACC = 4,9% ...64

Table 28 Scenario analysis results, Jordan (JCh31), WACC = 4,8% ...65

-VIII-

1 Contents

1 Introduction ... 1

1.1 Cleanergy AB ... 1

1.2 Objectives ... 2

1.3 Thesis structure ... 3

2 Theoretical Framework ... 4

2.1 Solar energy overview ... 4

2.2 Solar CSP technologies ... 6

2.3 Stirling-based CSP systems ... 9

2.3.1 Small scale CSP ... 9

2.3.2 Cleanergy CSP systems ...11

2.3.3 New design with TES ...12

3 Methodology ...16 3.1 Geographical limitation ...17 3.2 Market analysis/entry ...18 3.3 Financial model ...19 3.4 Techno-economical analysis ...21 3.5 Multi-criteria analysis ...24 3.6 Scenarios definition...26

4 National environment, industry and firms’ specifics ...27

4.1 Morocco ...27 4.1.1 Energy context ...27 4.1.2 Electricity market ...28 4.1.3 Industrial companies ...30 4.2 Tunisia ...31 4.2.1 Energy context ...31 4.2.2 Electricity market ...32 4.2.3 Industrial companies ...33 4.3 Egypt ...34 4.3.1 Energy context ...34 4.3.2 Electricity market ...35

-IX- 4.3.3 Industrial companies ...37 4.4 Jordan ...38 4.4.1 Energy context ...38 4.4.2 Electricity market ...39 4.4.3 Industrial companies ...41 4.5 Saudi Arabia ...42 4.5.1 Energy context ...42 4.5.2 Electricity market ...43 4.5.3 Industrial companies ...44 4.6 Country comparison ...45 5 Comparison/Analysis ...49 5.1 Optimum configurations ...49

5.2 Potential/Serviceable achievable market ...52

5.2.1 Results ...52

5.2.2 Sensitivity analysis ...55

5.3 Multi Criteria Analysis ...57

5.3.1 Results ...57 5.3.2 Sensitivity analysis ...61 5.4 Scenario analysis ...63 5.4.1 Results ...63 5.4.2 Sensitivity analysis ...66 6 Conclusions ...70 7 Appendixes ...72 Bibliography ... 102

-1-

1 Introduction

Being one of our era’s most critical problems, securing electricity access in a cost-effective way and without harming the environment is a big challenge that countries worldwide are trying to tackle. Consequently, electricity generation based on renewable sources is on the rise and has seen rapid growth in the last decade globally. Investments and technological innovations are the drivers of the shift in the current energy system, going away from centralized generation stations to decentralized and distributed units that can accommodate different configurations. Among the multiple technologies and renewable sources that can be used, solar energy ranks in the top due to its huge potential in providing electricity access on a global scale. The DESERTEC initiative suggested that in only six hours, the African desert receives as much energy from the sun as humankind consumes in a whole year [1].

However, and by definition, renewable and sustainable energy sources are intermittent, which hinders their full adoption and penetration into the grid. Solar energy is no exception to that, being available only when the sun shines, and unavailable after, period where the demand profiles are the highest. Specifically, intermittent solar production cannot fit around the clock the clock load profiles. The solution is then to couple the solar technology with a storage system that would generate electricity on demand, later in the day when there is no sun. Storage systems can refer either to chemical batteries in the case of PV plants, or Battery Energy Storage Systems (BESS), or thermal energy storage (TES) in the case of CSP systems. BESS are highly capital intensive and are only viable when considering large scale projects, but are not suitable, nor competitive to smaller distributed systems with large storage requirements [2] . Likewise, CSP systems, such through or tower with TES are proven to cost competitive in large scale sizes, and do not fit for distributed generation as their efficiencies drop when dealing with smaller systems.

Cleanergy AB, a privately owned Swedish company, is in the crossroad of all these considerations, offering a new modularized designed on-demand electricity production technology with distributed energy storage that can provide a high efficiency solar power plant and be built cost efficient in any size from the range of 10kW to hundreds of MW. This Master thesis, done in the form of an internship at Cleanergy AB, aims at elaborating business strategies for the company that wishes to develop its new technological solar innovation in the MENA and Sub-Saharan African regions where the solar resource is abundant and perfect for its product (DNI greater than 2000 kWh/m²/year)

1.1 Cleanergy AB

Cleanergy is a privately held Swedish company that was founded in 2008 and that focuses on Stirling engine-based renewable energy solutions. The company is a small and medium sized enterprise that has two sites of production, Uddevalla and Åmål, alongside offices in Stockholm and Gothenburg [3]. Its core expertise is the production and manufacture of Stirling engines which convert heat into electricity. The first target segment market was gas-fuelled power production that started in 2008 with the GasBox. This product is fully commercialized today in different countries (United Kingdom, Norway, Sweden…) [3]. In parallel, Cleanergy started developing solutions for CSP systems with the modified version of the Stirling engine called SunBox. The company proposes a modular CSP dish Stirling unit, comprising of a parabolic dish capable of tracking the sun that uses SunBox for electricity generation. While the generation is caped at 13kW per unit, the Cleanergy SunBox unit is well suited for large utility scales ranging from kW to MW scales thanks to its modular and autonomous design. No water is consumed in the power production cycle, which is a key competitive advantage over other CSP technologies such as linear trough and tower systems, especially in areas of high ambient temperature where high levels of Direct Normal Irradiation are usually to be found and water resources are usually scarce. It also holds the highest conversion efficiency from sun-to-electricity among all solar technologies, reaching 30% [3]. Currently, Cleanergy has commissioned 3 demonstration plants: 110 kW installed capacity in both Mongolia and Dubai seen in Fel! Ogiltig s jälvreferens i bokmärke., and a 13kW unit in Ouarzazate, Morocco. The company is now focused on

-2-

developing a thermal energy storage system of 10 hours to be coupled with its CSP unit, allowing for on-demand electricity production, increasing the grid stability, its flexibility and thus its attractiveness.

1.2 Objectives

The CSP system the company is developing is best suited for locations with a DNI greater than 2000 kWh/m²/year, which is why Cleanergy wishes to study the potential opportunities laying in the MENA. Indeed, most of the countries in the MENA are considered to be part of the Sunbelt, where solar radiation is well above the threshold mentioned [4]. More specifically, the countries studied in this thesis are: Morocco, Tunisia, Egypt, Jordan and Saudi Arabia. The choice of these 5 countries is strategic, as they all have enough resemblance to draw conclusions and strategies to be applied to MENA in general, but present in the same time enough differences for them to stand out and challenge the company in its way of doing business. The main goal of this thesis is to identify how a small scale distributed CSP system with TES, such as Cleanergy’s, can enter a specific marketplace and propose strategies the company should develop for its modular technology to successfully penetrate the MENA. Detailed prospective customer profiles are identified in each country, and for each customer-type a techno-economic analysis has been carried out to determine best configuration (in terms of key component size and operation) that would minimize the generation costs of the system in order to ultimately assess the competitiveness of a business case based on such a technology when compared to the current way of procuring electricity. At last, specific recommendations for technology developers and potential off-takers are provided.

As follows, this work needs to:

• Present a comprehensive and detailed market profile for each of the countries selected. This includes description of the electricity sector, generation capacities, regulatory framework in place for the power sector, future trends in terms of energy policy and generation.

• Identify potential customers in each of the countries selected, by validating the company’s assumptions on the market application, sizing the total addressable market, in terms of MW to install, in each one and identifying which customers are in need of dispatchable renewable electricity.

• Carry out a techno-economic analysis for each customer to determine best configuration (in terms of key component size and operation) that would minimize the generation costs of the system. • Set up business cases for selected companies/customers based on cash flow calculations to outline

the attractiveness of Cleanergy’s technology for them and its competitiveness when compared to current way of procuring electricity, but also provide proof of business profitability for Cleanergy.

-3-

• Suggest business development strategies to achieve successful market entry in the MENA region. The go-to market strategies to be developed will be based on the analysis of the countries selected and comparison between the customer identified. It will also include suggestions on how to build communication channels and strategic partnerships with the most promising customers.

1.3 Thesis structure

The report is constructed around the above objectives and follows the structure detailed below:

• Section 1, Theoretical framework: An overview of all the theory behind the concepts to be developed and used throughout the report, describing in detail Cleanergy’s value proposition and outlining the different solar technologies that compete with it, reviewing the market research theory and detailing the financial metrics that will be used as key performance indicators to rank and qualify the market.

• Section 2, Methodology: A step by step explanation of the methods followed to perform this research work, ranging from the data mining approach used, description of the economic model built to the actual market analysis and comparative approach applied.

• Section 3, National environment, industry and firm’s specifics: Acts as market analysis of the different countries reviewed, by setting up country profiles detailing market status and regulations, understanding how macro-environmental factors will help or hinder Cleanergy’s business, quantifying the total addressable market by identifying potential industrial customers, and carrying out techno-economical assessment for each.

• Section 4, Comparison/Analysis: Cross-country, cross-customer comparison and analysis to identify the most promising and profitable market for Cleanergy. The analysis will be based on individual business cases built for selected industrials and firms to understand the market behaviour and need, by considering different market shares scenarios for the future and different localization estimates for Cleanergy’s product.

• Section 5, Go-to market strategies: This section draws strategies and suggestions for Cleanergy to develop its business in each of the countries selected based on the market analysis done and delimits risks inherent in each one of them.

• Section 6, Conclusions: A summary of all the findings of the research work, and suggestion on how to proceed in the future.

-4-

2 Theoretical Framework

2.1 Solar energy overview

With regards to the shift in energy systems currently deployed around the world, solar energy technologies are well prioritized thanks to their numerous advantages. The old-fashioned way of producing electricity relying on fossil fuel brings many problems, from environmental concerns such as damage to the earth, pollution of the atmosphere and water, but also socio-political issues when considering a country’s need to secure energy access and the shrinking availability of fossil fuels and volatility of their prices. Relying on the sun for electricity production solves above issues, as the yearly received energy from the sun is 1500 times largen than the world energy use [5]. In fact, the DESERTEC project suggested that solar power plants located in the African desert covering 0,3% of its area, could power up all nations around the world given the right transmission infrastructure built [6]. Figure 2 shows the area requirement of such a project

Figure 2 DESERTEC project map. The red squares represent the area need for solar power plant to power the whole world, Europe and Germany [7]

While solar energy is usually associated with PV power generation, the sun’s irradiance also delivers its energy in the form of heat that can be used for power generation in CSP systems. In 2016, 76,6 GW of solar were installed, making the total solar capacity reach 306,5 GW, representing a 33% increase compared to 2015. Asia leads the solar market, dominating with 48% of the total install capacity making it the largest solar powered region. The global forecasted capacity to be installed in 2017 is 387 GW, while this figure will surpass 700 GW after 2030 [8]. The evolution of the global installed solar capacity throughout the years from 2006 till 2016, for both PV and CSP can be seen in Figure 3 and Figure 4. Solar CSP power plants are capital intensive, needing huge investments for their erection compared to PV plants, which explains the difference between the installed capacities of both [9].

-5-

Figure 3 PV installed capacity growth [10]

-6-

2.2 Solar CSP technologies

Unlike photovoltaic cells or flat plate solar thermal collectors, CSP power plants do not use the global solar irradiation, namely disregarding the diffuse part which results from scattering of the direct sunlight by clouds, particles and molecules in the air because they cannot be concentrated. The CSP technology is based essentially on the direct solar radiation, which is collected through a concentrator to a receiver. This makes CSP power plants best suited for locations with high percentage of clear sky days, which do not have smog or dust. The concentrated heat is then used to run a power conversion cycle to produce electricity. A schematic of a conventional CSP plant is given in Figure 5.

Figure 5 Flow diagram of a typical CSP plant [11]

The most common CSP systems are shown in Figure 6 .A brief overview of each is given below, while Table 1 compares the technological specifications of each [12] [13].

Figure 6 CSP technologies [14]

• Parabolic trough (PT): It is the most deployed CSP technology. Trough-shaped mirrors concentrate sunlight into a linear focus on a receiver tube that follows the parabola’s focal line. The mirror and receiver tube structure are mounted on a frame that follows the daily sun movement on one axis, while the seasonal movement of the sun are tracked with lateral movements of the line focus. The heat is collected from the receiver tubes via a heat transfer fluid and is used to feed a power block for electricity generation.

-7-

• Linear Fresnel reflectors (LF): Variation of the parabolic trough collectors. Their main difference from parabolic trough collectors is that, instead of using parabolic bent mirrors to concentrate sunlight, they use several parallel flat mirrors to concentrate it onto one receiver, which is located several meters above the primary mirror field. The secondary mirror structure is necessary to account for the astigmatism distortion caused by the optical principles of Fresnel collectors. • Centrale receiver tower (CT): This design contains an array of heliostats (large mirror structures

with double axis tracking) that concentrates the solar radiation into a central receiver mounted on the top part of a tower. This configuration gives high efficiency energy conversion into the large receiver point, yielding higher concentration ratios compared to linear focusing systems. It permits the power cycle to work at higher temperatures with reduced losses.

• Parabolic dishes (DS): Similarly to the trough design, dish systems rely on the geometric properties of a three-dimensional paraboloid to concentrate direct solar radiation to a point focus receiver, reaching in optimum condition temperatures over 1,000ºC, similar to tower systems. The latter gives them the advantage of having the highest solar conversion efficiency, since they always have the aperture facing the sun and avoid the cosine loss effect. These systems have a power conversion unit, namely Stirling engine, that transforms the concentrated heat into electricity. This will be further explained in the following sections, as it is Cleanergy’s key product.

Table 1 CSP technology comparison [15]

Technology PT LF DS CT Typical size (MW) 10 – 280 1 – 125 1 10 – 135 Concentration Factor 70 – 80 25 – 100 600 – 4000 600 – 1200 Capacity Factor (%) 30 – 50 20 – 30 20 – 30 40 – 70 Operation Temperature (ºC) 293 – 393 140 – 275 250 – 700 290 – 565 Sun-to-Electricity efficiency (%) 16 – 18 9 – 11 12 – 25 16 – 20 Installed worldwide (MW) 4336 319 3 689 Use of land (MWh/(ha·year)) 600 – 1000 600 – 1000 400 – 800 400 – 800

Maturity Commercial Commercial Demo Commercial

Reflector Parabolic _mirror Flat/curved _mirror Paraboloid _mirror Curved mirror

Receiver Absorber tube w/vacuum cover Absorber tube w/concentrator Stirling engine/Gas turbine External / Cavity HTF _{Thermal oil Saturated steam Air} Molten salt / _Water-steam

TES Molten salts, indirect

Steam

accumulator N/A

Molten salts, direct / steam

accumulator TES capacity 4 – 12 hours < 1 hour N/A 6 – 14 / < 1 _hours

-8-

In terms of market deployment, parabolic through systems dominate the CSP plants in operation globally as of 2016, followed by central tower systems. This is due to the huge solar investments Spain made in the past regarding solar CSP, considering that at that time trough design was the most developed and proven technologically. However, and as it is seen in Figure 7 , the majority of planned projects, as well as the ones under development are central tower systems, proving that there is a shift in the market justified by better efficiency due to the higher temperatures achieved in the receiver, compared to through systems. Central tower systems capitalize on that as well with regards to heat storage in molten salts, as the high temperatures reached in the receiver allow for a reduced cost of energy storage unlike trough systems [16]. Figure 7 also shows which players are investing the most in solar energy. While Spain lead the market in the early 2000s, near zero projects are planned or developed there in 2016. This is mainly due to the fact that the country suffers from a huge electricity tariff deficit that pushed officials to halt capital intensive projects. MENA countries, on the other hand, are investing heavily in solar and renewable energy in general, acknowledging their need for sustainable energy access and security, and capitalizing on the perfect renewable resources they have. It is suggested that PV alone has a potential of 7GW by 2020 in the MENA, and 27 GW by 2030 [4]. Nevertheless, the increasing number of large-scale solar projects planned and developed in that region make it a promising market for companies like Cleanergy, as its countries show strong commitment to energy transition.

-9-

2.3 Stirling-based CSP systems

Whereas conventional CSP technologies are competitive in large-scale parks, they have yet to become attractive for small-scale distributed and dispatchable generation (< 5MW). On the other hand, PV-BESS plant can be seen as best suited for that type of use, as they are more competitiveness to diesel generators [2], which they mostly compete with. Both alternatives can offer on-demand electricity production and represent a stable input to the existing grid generation. But PV-BESS joys from several advantages, in the sense that they are not subject to volatile oil prices, or CO2 abatement policies and restrictions, that often put extra financial hurdle on fossil fuel generation. In spite of that, the current cost of chemical batteries for PV systems is too high and render it not competitive, nor its 2030 future projections [2].

As a result, a number of companies are currently developing modular and distributed cost-effective CSP systems with TES. For instance, 247Solar develops a dispatchable 300 kWe system consisting of a heliostat field-tower system, a micro gas turbine (MGT), and a brick-based dry TES. Specifically, a small heliostat field concentrate solar power into a receiver mounted in a 35 m tower. The receiver heats air passing through it to about 980 °C that in turn warms up turbine's compressor air. The microturbine is then powered by the super-heated compressor air, thus spinning a generator to produce electricity. The system uses no water/steam, salts, oils, hydrogen or helium. Not all the hot air from the receiver is used from power generation, and serve as heating source to the TES, in the form of firebricks or small pieces of ceramic.). When the sun isn't shining, air is blown through the hot TES to heat the turbine's compressor air. Natural gas or biofuels (e.g., from landfills) provides backup power when there is not enough solar power, or during nighttime [17]. An illustration of the 247Solar system is given in Figure 8.

Figure 8 247Solar Plant [18]

The same idea is reprised by Australian company Vast Solar, which already constructed three CSP research and demonstration facilities, where a small heliostat field concentrates sunlight into a 30 m tower and receiver, for electricity to be generated through a small steam turbine. Storage is achieved thanks to molten salts. A pilot demonstration plant (6MWth, 1,1 MWe with 3 hours of storage) was commissioned in 2016, and later connected to the Australian grid, making it the first CSP plant with storage connected in Australia. The plant consists of five solar array modules. Each module consists of one tower of approximately 30m, a thermal energy receiver and about 700 heliostats. Modules connect to a central energy storage tank with molten salts, and from there the stored thermal energy is passed through a steam generator to make steam for a small (1.1MWe) turbine and electricity generator [19]. The schematic of the plant is given in Figure 9.

-10-

Figure 9 Vast Solar CSP system [19]

A similar concept was proposed by AORA-Solar, a developer in solar-biogas hybrid power technology that specializes in small-scale off-grid solutions. The design relies on the same heliostat field/tower configuration, but the electricity is produced by a Stirling engine instead of a MGT. AORA-Solar however do not propose TES, relying on natural gas, biofuels or diesel to offer around the clock electricity generation. The lack of updated information and literature about the company and its projects means most likely that it went bankrupt, as of today, its website is shut-down [20] [21].

Subsequently, promising alternatives are based on solar-powered Stirling engines integrated with TES, leveraging from both the modularity (10-40 kW) and the high efficiency of the engine (e.g. when compared to MGTs, and even to conventional cycles in large CSP plants). Indeed, a recent study performed by several U.S. research institutions and funded by the U.S. Department of Energy [22] studies the performance modelling and economical viability of a CSP tower small scale system with latent storage, referred to as STEALS. The research study reprises the heliostat field design, that reflect sunlight on top of a tower, where the entire thermal system is located. The cavity receiver heats up the bottom of the TES tank (and its PCM), while sodium heat pipes extend vertically from the bottom of the tank to the top, distributing heat in the storage material. A thermosyphon-based thermal valve acts as an interface between the TES tank and the Stirling engine used for electricity generation. The heat flow from the storage to the power block is controlled by a valve, as the Stirling engines considered in the study range from 0.1 to 1 MW. This modular design results in minimal balance of system costs and enables high deployment rates with a rapid realization of economies of scale, where generation costs reach values well below 100 €/MWh [22]. A schematic of the whole system is given in Figure 10.

-11-

Figure 10 Design configuration of STEALS [22]

Against this background, previous research have investigated the potential small distributed CSP systems can have, be it with storage or without [23] [24]. These studies provide summary information about market application for such a design, and are narrow in scope as they don’t go beyond a general classification of sector application: industries, villages…, or give a quantification of the potential, in terms of MW to install. Neither have these proposed a business model that would allow small scale CSP systems with TES to secure its market niche, based on the risks hindering its market entry. On the other hand, when it comes to analysis work that measures the potential of a new technology, and specifically CSP in different countries, one can name the HYSOL project work package 2 [25], which studies the economic feasibility and market penetration of an innovative configuration for a fully renewable hybrid CSP plant. The study is carried out for four selected countries: Kingdom of Saudi Arabia, Chile, Mexico and South Africa. It first assesses regulatory and policy framework regarding renewable energies in each to understand the power market, renewables energy targets…, to then carry out a corporate economic assessment for the decision-making process based on metrics such as LCOE, NPV and IRR for the different countries. Based on the mentioned analysis, it finally draws conclusions on the potential of the new technology in each prospective. Another project tutored by Apricum, the strategy consultancy firm specialized on renewable energies deals with the assessment of business opportunities present in the solar industry for Saudi Arabian companies. It represents a prefeasibility analysis for a Saudi Company to enter the solar industry, by analysing the solar market’s value chain and performing a multi criteria analysis of different business opportunities. Two representative business cases are presented afterwards to showcase the value of the best identified opportunities [26].

2.3.2 Cleanergy CSP systems

2.3.2.1 Legacy product

The company’s CSP solution “SunBox” is based on one core competence and component, which is the Stirling engine. Cleanergy modified this two-hundred-year-old technology to better suit solar applications, converting concentrated solar radiation to electricity with the engine [27]. The Stirling cycle is a closed cycle that contains a unique and fixed volume working fluid (gas) that is heated, expanded, cooled, and compressed, thus driving a piston for power generation [28]. Cleanergy’s system relies on an air cooled, Alpha type Stirling engine. This configuration, shown in Figure 11, is characterized by two different pistons in two separated cylinders. They are connected in series through a heater, a regenerator and a cooler. By doing so, one piston acts as the “hot” part of the engine, and the other as the “cold” part. The heat coming from the parabolic dish and receiver heats up the working fluid (hydrogen), causing it to expand, pushing

-12-

the hot piston and driving the crankshaft to create momentum. As a result, the cold piston is compressed, moving the working gas into the cold heat exchanger and regenerator. The generator connected to the moving crankshaft generates electricity, while warm air is rejected through the air cooling system [27] [3] [28].

Figure 11 Cleanergy's Alpha type Stirling engine [27] (adapted)

As mentioned before, Cleanergy is currently developing a Stirling based CSP solution with TES. However, the company had a previous design based on the Dish Stirling configuration. The latter will be referred to as “legacy product” in this report, since the company no longer focus on it. Dish Stirling systems use parabolic shaped mirrors to concentrate solar radiation into a receiver. The heat in the receiver feeds the Stirling engine to produce electricity as explained above. “SunBox” refers to the unit Cleanergy produces, which contains its modified version of the Stirling engine alongside the receiver. With a nominal capacity of 13kW, the SunBox unit is mounted onto the parabolic dish structure that can track sun movement throughout the day. The main advantage of this system is its modularity, that allows it to operate individually in remote locations such as small villages or off-grid locations. It also offers the possibility for medium to large scale application by associating multiple Dish Stirling systems. With regards to the technical specifications and comparison made in Table 1, the demonstration unit installed in Dubai reached a conversion efficiency of 30%, record value for any solar technology. The system is best suited in hot arid climate zones as mentioned before, where there is high level of direct normal solar irradiance. The water scarcity characterizing these regions does not hinder Cleanergy’s system operation, as no water is required for power generation. Moreover, the majority of the components of the system have near zero degradation over a period of twenty-five years, rendering Cleanergy’s product sustainable, efficient, robust and the perfect alternative for power generation during the day [3] [29] [27]. The addition of a thermal storage unit would allow for dispatchable electricity, which is being developed at Cleanergy, and will be further detailed below.

2.3.3 New design with TES

As seen in both Figure 3 and Figure 4, the installed capacity of PV power plants exceeds by far CSP. The reason for this is the cheap price of PV cells and modules compared to solar thermal. As a point of reference, PV power plants’ bidding price reached 30$/kWh in 2016, while CSP tower bidding price was 94.5$/kWh [30]. However, while PV is attractive for its cheap and simple design, it becomes less appealing when the sun falls, period when electricity demand is on the rise and PV output fading. The solution is to couple the photovoltaic panels with storage batteries to produce reliable electricity like CSP systems do with TES. Li-ion battery is most often the best choice, but offsets PV’s greatest advantage of being cheap. As of 2017, the cost of Li-on battery pack was around 230$/kWh, but degradation considerations make it even more costly: it may be needed to replace batteries 4 to 5 times during a PV power plant lifetime [30]. TES does

-13-

not present these limitations, and Cleanergy plans to capitalize on that to compete with PV and other CSP technologies.

Cleanergy’s target market and strategy can be seen in Figure 12. In a nutshell, while PV technology is definitely cheaper than any other renewable alternative from small scale installations(residential) to utility scale plants, dispatchability and storage features make it way more expensive than the other solar systems, especially in medium and large-scale projects needing storage capabilities greater than 4 hours. On the other hand, CPS tower and through systems with TES systems are commercially viable for large installed capacities of over 50 MW as was seen previously. Cleanergy then positions itself in markets of installed capacity from 100kW to 50 MW with long hours of storage above four hours. This market segment is a niche, meaning that it is a blank spot where no product with competitive added value has been proposed yet.

Figure 12 Cleanergy's initial target market (2021-2025) for the TES system design

To do so, the company started the R&D on a new CSP Stirling based system that incorporates TES of 10 hours or more. Figure 13 shows the proposed design for the new system. In this configuration, a solar field of small heliostats concentrate sunlight into a receiver mounted onto the top of a 10 m tower. The receiver is linked to the TES system, which uses a phase changing material (PCM). The PCM transfers the heat collected with a heat transfer fluid (HTF) that powers up the Stirling engine through a heat exchanger.

-14-

This new design is currently being researched and developed with the following technical specifications: • Each modular system (solar field, tower and Stirling engine) will have a rated power of

13kW. This way the technology can still be used for distributed configurations but allows for large scale deployment as well.

• The TES should have a capacity of 10 hours or more, storing enough heat for electricity production when the solar resource is not available, but to also contribute to grid stability by providing a firm electrical output. A storage utilization fraction of 90% is targeted. • The Stirling engine’s performance depends to a great extent on the input temperature, the

change operated in the system (compared to the old dish configuration) should not affect the thermal to electricity efficiency of 30%.

Based on these technical objectives, the new product must be competitive with other technologies on the market, namely PV-BESS, but also conventional power generation system (Diesel based generators). Cleanergy’s value proposition, especially in the niche mentioned above, must have the lowest cost of electricity generation. More specifically, the company aims for a 25-30% lower cost of electricity production compared to PV-BESS and Diesel Gensets based on forecasted generation costs for said technologies in 2021. Figure 14 shows the LCOE evolution of PV-BESS with storage hours placed in Ouarzazate, Morocco with a DNI of 2630 kWh/m²/year. The costs used are 2021 projections based on several forecast sources (Lazard [32] , IRENA [2] , NREL [33]). The target LCOE of the company should be lower than a similar PV-BESS or diesel generator system with the same amount of storage hours, i.e.. 10 hours or more for a similar location. The long-term goal is to reach a LCOE of around 35€/MWh in the year 2030 with the setting (location, storage size, rated power). The reduction in cost is expected to be driven by higher volumes

Figure 13 Model of one modular Cleanergy CSP unit with the three main components: concentrator, receiver with storage (10 hours) and Heat Engine (Stirling)

-15-

of production, as well as reduced costs of installation, engineering and O&M. The reported learning rate for CSP technologies with storage is 30% for the period 2010-2022 [31].

Figure 14 PV-BESS LCOE in 2021

0 20 40 60 80 100 120 140 2 3 4 5 6 7 8 9 10 11 12 13 14 L C OE (E UR /MW h) Storage (h)

-16-

3 Methodology

The present work aims at estimating the market potential of a dispatchable small scale CSP system with TES, such as the one Cleanergy is currently developing, limited in the spatial boundary of the MENA. In other words, the size of such a market must be quantified in relevant figures, such as MW to install, or units (solar field, tower and Stirling engine) to deploy. Market size determination relies generally on three key concepts as visually depicted in Figure 15 [34] :

• Total Addressable Market (TAM): represents the size of the market if the product analysed were to meet all the demand, disregarding any kind of competition. In Cleanergy’s case, the TAM would be the number of plants to install in the MENA region to meet all of its electricity demand.

• Serviceable Achievable Market (SAM): represents the portion of the TAM that the product assessed is actually targeted to and geographically reachable, excluding any competition. In the scope of this work, business opportunities to potential industrial companies (or any other local company) that would be interested in the product. Industry accounts for 42% of the electrical consumption globally [35], needing most often continuous electricity supply for its processes and activities. It is then a perfect sector Cleanergy targets to seek business opportunities.

• Serviceable Obtainable Market (SOM): is the selected business opportunities within the SAM that a company targets first to grow and develop its product.

Figure 15 Market size estimation [34]

At this stage, Cleanergy is most concerned with understanding its SAM, sizing its magnitude and numbering all business opportunities within that area. From that, the company can clearly sees which of them are the most promising, and that will represent its SOM. Thus, this work needs to present a way to accurately estimate the SAM for Cleanergy, and propose a way to sort the best cases forming its SOM. To do so, the approach depicted in Figure 16 is followed. The business opportunities are investigated in representative countries of the MENA: Morocco, Tunisia, Egypt, Jordan and Saudi Arabia. Business opportunities refer here to potential industrial companies (or any other local company) that would be interested in the product. Industry accounts for 42% of the electrical consumption globally [35], needing most often continuous electricity supply for its processes and activities. The methodology revolves around 2 steps:

• Filtering: In each country, a market analysis is done, where the main electricity intensive industrial companies are identified, and regrouped by sector: mining, cement, chemical, metallurgy, agriculture. Based on publicly available data, the electricity consumption of each company is estimated and broken down to each of their consumption sites, for which exact location coordinates and respective weather data are gathered. For each of these sites, a techno-economical analysis is carried out based on the company’s simulation model. Different plant configurations, in terms of installed capacity, storage size, and mirror area were evaluated, from which the optimal configuration able to reduce the LCOE is selected. The knock-out criteria to filter the business opportunities in this step is LCOE, and the lowest value it can reach for each site. Through this

-17-

step, a quantification of the SAM is made in each country, alongside identification of companies that could populate the SOM of Cleanergy. The choice of which will Cleanergy should effectively engage with will be the result of the second step.

• Scoring: The above-mentioned optimal configurations are regrouped by country in order to estimate the country-specific potential in terms of installed capacity. With such information, a multi-criteria analysis (MCA) is performed in order to be able to compare amongst the different markets (by country), considering not only the potential for installed installations in MW, but also the lowest cost at which parks could be built, macro-environmental factors in the country, and existing infrastructure, among others. The MCA is used to score each business opportunity, to finally select the top ones. The latter represent the lower bound of the SOM, while its higher bound is the SAM which is also sized using global industry electricity consumption.

• Scenario analysis: Once top business opportunities are identified with the MCA, three scenarios are defined to further the comparison between the opportunities and countries, in terms of profitability for Cleanergy.

Figure 16 Selection methodology for business opportunities [36]

Cleanergy aims at positioning itself in the solar energy market as technology provider. In other words, its business model will repose solely on its ability to find customers interested in owning and operating their own power generation facilities. However, for such a novel technology like Cleanergy’s, it will be hard to garner the desired market interest, due to it being unknown and not yet proven. Consequently, taking an active part in the first projects to be deployed, e.g. being a co-developer can be more strategic. Doing so will help introduce the product to the market, by showcasing that Cleanergy is equity shareholder in the power plants based on its technology, thus build general trust in its product. In that sense, the analysis to be performed will assume a different business model for Cleanergy, which will act as Independent Power Producer (IPP) and operate the first projects under a Build-Own-Operate scheme. For such models, producers compete in a liberalized power market, where off-takers decide freely the source of their electricity procurement, which can be the national electricity utilities, other IPPs (wind, solar PV…) or even invest in their own generation capacity. Hence, all calculations and discussions are done assuming the IPP model for Cleanergy, rather being just a supplier, to assess the economic viability of the projects (primarily on a LCOE basis). Indeed, although Cleanergy’s interest as suppliers is to sell as much systems as possible, the project perspective matters in reality to the company as only a viable/competitive project will make the technology be chosen. Conversely, while the IPP models encompasses the opportunities laying in the grid-connected market, it lacks including off-grid users, as by definition that business model is not valid for such setting. For that purpose, and not to disregard the potential business opportunities of users not connected to the grid, the highest scores/weight are given to such projects in the MCA.

3.1 Geographical limitation

As mentioned previously, the study revolves around the MENA, with a special focus on Morocco, Tunisia, Egypt, Jordan and Saudi Arabia. The choice of these 5 countries is motivated by is considered to be strategic, as they all have enough resemblance to draw conclusions and strategies to be applied to MENA in general, but present in the same time enough differences for them to stand out and challenge the company in its

-18-

way of doing business. Beyond that, those specific markets were included in the AFEX (Arab Future Energy Index) Renewable Energy survey, ranking among the top 10 countries with most potential in RE investments as seen in Figure 17. The AFEX is a policy assessment and benchmark tool that provides a detailed comparison of renewable energy development in 17 countries of the Arab region on more than 30 different indicators. Such indicators include market structure, policy framework, private investment regulations… [37].

Figure 17 AFEX Renewable Energy 2016 [37]

Hence, Morocco, Jordan and Egypt are representative of countries with the most potential when it comes to renewables. Saudi Arabia, though ranked 10th_{, is also investigated in this work considering its market size,} and the strategic role it plays in the region, and in the world as a dominant oil producer. Tunisia is included as well, as a representative under-developed country transitioning to renewables.

Once the SAM in these 5 markets is estimated, a scaling up process is necessary in order to have the same figure for the whole MENA region. This can be achieved by considering the contribution of the industrial sector as added value in each country’s GDP and interpolating the corresponding SAM as a function of the latter. This is made possible because the SAM sizing deals exclusively with the industrial sector electricity consumption. The data obtained for the 5 countries mentioned above serve as the basis for the interpolation. The choice to include Saudi Arabia as country to research gets then even more meaning as it is the number one ranking industrial economy in the MENA [38] [39], thus facilitating the scaling up process and make it somewhat more precise.

The MENA is a region encompassing approximately 22 countries in the Middle East and North Africa. While there is no standardized list of which countries are included in the MENA, the following are typically included in MENA: Algeria, Bahrain, Egypt, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Qatar, Palestine, Saudi Arabia, Syria, Tunisia, United Arab Emirates, and Yemen [40]. Subsequently, the SAM in these locations needs to be estimated to give a general figure about the MENA.

3.2 Market analysis/entry

To complete the task defined above, a comprehensive study of the five countries is carried out to assess their macro-environment, but also to identify potential industrial customers. The latter are individually researched by analysing their electricity need in terms of power capacity Cleanergy can install to cover their demand. This data mining step is crucial, as it is the basis of the analysis to be carried out later on and that will contribute in building the go-to market strategies. The information for each company is gathered through research of publicly available studies, governmental report, utility annual activity reports, journal articles… However, the data looked for, namely electricity consumption figures, load profiles, is most often not made public, or hard to find, and was estimated based on several assumptions. This may lead to multiple inaccuracies in the analysis done. As a result, finding ways to validate it is also primordial and is investigated.

-19-

Appendix 2 represents one way of doing so: a questionnaire made for the Moroccan companies, and that was designed to help assess and validate the information collected.

Beyond that, once an optimal prospective list of customers defined, there is a need to understand the environment around those customers, and how the company can do business in the desired markets. Renewable policy targets, regulatory framework, business risk and fraud are all factors that need to be considered and understood. Thereafter, an entry mode needs to be chosen to actively penetrate new markets. The choice of entry mode into a foreign market is a very important decision for companies whose activities are directed toward the international and wishes to expand there. Entry modes are based on the firm’s involvement level, or the degree of influence and confidence it has over its operations but are also based of equity of ownership and control [41]. Different motives explain a company’s choice of an entry mode over another. For example, a firm will find that entry mode yielding the higher percentage of return on investments the best suitable, while another would prefer an entry mode guaranteeing close to zero risk [42]. Entry modes fall into three different categories as detailed in Table 2. A company can choose between them when entering a new foreign market. Those categories depend on the firm’s level of control, and are differentiated by their types of arrangements [43]

Table 2 Entry modes categories [31]

Categories Entry modes Arrangements

High Control Modes Wholly Owned Subsidiaries

The owner of the parent company has full control over the business in the new market.

Intermediate Modes

Strategic Alliances Partners agree to share technology, jobs, and resources & provide support to each other during the agreed time.

Joint Ventures

Low Control Modes

Indirect Export

The parent company use independent organization located in the home country/ third country

Direct Export

The parent company sells directly to a distributor, agent or importer based in the market

3.3 Financial model

The power price that is set during RES tenders is the key point for a project to be awarded. These prices are based on the cost of energy and quantifies the profits the projects stakeholders will make over the lifetime of the power plant. They are determined by doing financial analysis that include all factors and risk predictions to yield the best profits. The key metrics used to rank electricity production technologies are the Levelized Cost of Electricity (LCOE), Net Present Value (NPV) and Internal Rate of Return (IRR). It is important to be rigorous on their definition as various methodologies (inflation, nominal rates, taxes issues…) are used in the industry and academia, and it can mislead decision makers when comparing projects whose indicators do not follow the same method of calculation. Appendix 1 is a report done prior to this thesis work, which compiles and explains in detail the methodologies and calculations of different metrics used to financially valuate solar power plants. Below is a brief description of the above-mentioned metrics, but the extensive explanations, definitions, terminology and symbols used in Appendix 1will be referenced in this research work.

-20-

Levelized Cost of Electricity

The levelized cost of electricity is the most frequently used economic performance metric for power generation plants. LCOE is used to assess/compare the performance and profitability of any form of generation technology, and not only concerns solar or renewable sources [44]. It is defined as the constant per unit cost of energy which over the system’s lifetime will bring all the project cash flows to zero. In other words, it is the ‘break even’ constant sale price of energy [45]. Another way to view the LCOE is it being the price at which the electricity must be sold to recover all the costs incurred during the lifetime of the project. Equation 1 gives the general formula for calculating the LCOE. All used symbols and terms are explained in Appendix 1 𝐿𝐶𝑂𝐸 = ∑ 𝐶𝐴𝑃𝐸𝑋∗𝐸𝑞% 𝑁𝑐𝑜𝑛𝑠×(1+𝑅𝑂𝐸)𝑡 𝑁𝑐𝑜𝑛−1 𝑡=0 −∑ 𝐷𝐸𝑃×𝑇 (1+𝑅𝑂𝐸)𝑡 𝑁𝑐𝑜𝑛+𝑁𝑑𝑒𝑝−1 𝑡=𝑁𝑐𝑜𝑛 +∑ 𝐼𝑁𝑇𝑡×(1−𝑇)(1+𝑅𝑂𝐸)𝑡 +∑ 𝑃𝑅𝐼𝑁𝑡 (1+𝑅𝑂𝐸)𝑡 𝑁𝑐𝑜𝑛+𝑁𝐿−1 𝑡= 𝑁𝑐𝑜𝑛 𝑁𝑐𝑜𝑛+𝑁𝐿−1 𝑡= 𝑁𝑐𝑜𝑛 ∑ 𝐸𝑡 (1+𝑅𝑂𝐸)𝑡 𝑁 𝑡=0 + ∑ 𝑂𝑃𝐸𝑋×(1−𝑇) (1+𝑅𝑂𝐸)𝑡 𝑁𝑐𝑜𝑛+𝑁𝑜𝑝−1 𝑡= 𝑁𝑐𝑜𝑛 +∑ 𝐷𝑒𝑐𝑜 𝑁𝑑𝑒𝑐×(1+𝑅𝑂𝐸)𝑡 𝑁−1 𝑡= 𝑁𝑐𝑜𝑛+𝑁𝑜𝑝 ∑ 𝐸𝑡 (1+𝑅𝑂𝐸)𝑡 𝑁 𝑡=0 (1)

Net Present Value and Internal Rate of Return

The Net Present Value (NPV) of a proposed project is most often used as the primary absolute metric to compare/assess investments and serves as a base for decision making [46]. The NPV is the sum of the discounted cash-flows over the lifetime of the project using an appropriate discount rate as discussed above. The cash-flows represent the yearly difference between the revenues and costs incurred each year. It is then linked primarily to the CAPEX, OPEX, decommission costs, the yearly energy yield or output and finally the price at which the electricity is sold. Equation gives the general formula for calculating the NPV. All used symbols and terms are explained in Appendix 1.

𝑁𝑃𝑉 = ∑(𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝑠 − 𝐶𝑜𝑠𝑡𝑠)𝑡 (1 + 𝑟)𝑡 𝑛

𝑡=0

= 0 (2)

Another fundamental economical metric that is used to rank projects and get a hold of their profitability is the internal rate of return or IRR [44]. The internal rate of return is the discount rate that would be used in an NPV calculation and would make it equal to zero, as seen in Equation 3.

𝑁𝑃𝑉 = ∑(𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝑠 − 𝐶𝑜𝑠𝑡𝑠)𝑡 (1 + 𝐼𝑅𝑅)𝑡 𝑛

𝑡=0

= 0 (3)

The IRR is then the interest rate that would break even the project accounting for the costs incurred and revenues generated during the lifetime of the plant. It is a measure of the profitability of a project and is used mainly by developers and financial institutions to base their investments decisions. Each company has its own predictions on how much profit can be made of a project and has usually a target return on investments. If the IRR is higher than that required target, the project is financially acceptable. To compare different projects and financing opportunities, the higher the IRR the better [44].

The input data for the financial model are described in Table 3 . The specific component costs (solar field, receiver, TES, Stirling engine) are confidential to Cleanergy, but cost values from a similar technology concept, known as STEALS and described in a recent study [47]. The latter are given, and used as a reference to compare the results, in terms of LCOE values and so on. It should be noted though that cost projections for the STEALS project are optimistic, and represent idealized future system cost, rather than the cost of a system that could be built today. Not all manufacturing costs were considered for example and consider large scale production rate. Moreover, heliostats are the equipment that weigh the most in the project’s CAPEX, and the STEALS study considers a low value compared to existing plans. As underlined by the

-21-

project report, it is assumed that improvements will be made by the greater CSP community to reach such low values, especially considering that heliostats are used in many systems [47].

Table 3 Financial model inputs

Parameter Value

Stirling engine cost 808 €/kWe

Heliostat cost 61 €/m²

Tower/receiver cost 90 €/kWe

TES cost 25 €/kWh

OPEX cost 25 €/kWe

Project lifetime 30 years

Inflation 0%

Power price escalation factor 0%

Equity financing 25%

Equity IRR 8%

Debt financing 75%

Cost of debt 5%

Debt amortization 15 years

Depreciation 25 years

Corporate tax rate

Morocco: 31%, Tunisia: 25% Egypt: 22,5% Jordan: 25% Saudi Arabia: 20%

Naturally, LCOE figures inherent to Cleanergy’s system cannot be shown due to confidentiality reasons. However, since the analysis done revolves to great extent on that metric, normalized values of the calculated LCOE will be shown later in the report. In each data set to be calculated, the normalization will be done taking as a reference the minimum value of said data set.

3.4 Techno-economical analysis

As explained previously, the analysis to be carried out consists of identifying potential off-takers in the form of industrial companies and trying to assess the electricity consumption on each of their sites, exact location coordinates and respective weather data, which serve as input data for the simulation model the company developed. The latter calculates the system size (MW) of Cleanergy’s product needed to service that electricity consumption. This in turn serves as input data for the financial model described above, that calculates the LCOE. Different plant configurations, in terms of installed capacity, storage size, and solar field area are evaluated, from which the optimal configuration able to reduce the LCOE is selected. A schematic of the process is shown Figure 18.