Global cost advantages of autonomous solar – battery – diesel systems compared to diesel-only systems

(1)

Global cost advantages of autonomous solar – battery – diesel systems compared to diesel-only systems

C. Cader

^a,

⁎ , P. Bertheau

^a

, P. Blechinger

^a,b

, H. Huyskens

^a

, Ch. Breyer

^c

aReiner Lemoine Institut gGmbH, Ostendstr 25, 12459 Berlin, Germany

bDepartment of Engineering, Berlin Institute of Technology, Fasanenstraße 89, 10623 Berlin, Germany

cLaappenranta University of Technology, Skinnarilankatu 34, 53850 Lappeenranta, Finland

a b s t r a c t a r t i c l e i n f o

Article history:

Received 31 May 2013 Revised 31 July 2015 Accepted 29 December 2015 Available online xxxx

Isolated diesel systems are the main electricity generation method in many rural areas nowadays and represent a viable option to supply un-electriﬁed villages in the Global South. However, this generation scheme leads to a dependency on fossil fuels and their price volatility on a global market with a projected increase of costs in the future. At the same time, high carbon dioxide emissions increase environmental costs. Up to date, many hybrid mini-grid pilot projects and case studies were performed globally to assess how the inclusion of renewable energy in these systems can enhance technical and economic performance. This provides insights in local characteristics and challenges of that approach on a case by case basis. This study, on the other hand, takes a look at the overall global potential for solar–battery–diesel mini-grids for rural electriﬁcation and derives a comparative analysis of the respective regions. The introduction of a GIS-based analysis in combination with a sophisticated mini-grid simulation allows a highly automated approach to draw global conclusions with the option to down- scale to local regions. The results of the methodology show that in many regions substantial LCOE reductions are achievable by introducing solar–battery–diesel systems compared to pure diesel systems. Furthermore, the crucial role of spatial varying of diesel fuel prices over different regions and the impacts on the feasibility of solar–battery–diesel systems can be observed.

Keywords:

Rural electriﬁcation geo-spatial analysis Off-grid

Spatial cost modeling

Introduction

Access to electricity is still a huge challenge for more than 1.3 billion people globally. Highly affected regions are located in rural remote regions in Sub-Saharan Africa, Southeast, and South Asia (IEA, 2014a).

Rural remote regions are facing the great challenge that the population densities are often low, the demand is difﬁcult to assess and the reliabil- ity regarding the ability to pay for electricity is unclear(Zvoleff et al., 2009; Winkler et al., 2011). Low local demand for electricity makes the installation of transmission lines expensive and less efﬁcient and hence the hybrid renewable system more feasible. Centrally organized power supply systems are not designed for small settlements, and their costs are often underestimated (Cader, 2015; Chaurey et al., 2004). These are reasons for utilities and power companies to be careful about investing in that area and opening the space for decentralized energy solutions (Narula et al., 2012).

Electricity is a key requirement for development: education, health sector, and economic sectors beneﬁt from it. A clear relation between the gross national product (GDP) and the national energy consumptions can be shown (Doll and Pachauri, 2010). In addition, the human

development index (HDI) is also positively influenced by having access to electricity (Kanagawa and Nakata, 2008). Electricity access is not given because the national electricity transmission and distribution infrastructure does not reach far enough, the grid is facing regular power outages, or the electricity from the grid or alternatives such as small fossil fuelled generators are not affordable for the respective in- habitants (Kaundinya et al., 2009). In particular, the latter option– small diesel generators–is a frequently used method of electricity generation for remote locations (Bertheau et al., 2014). With the international agenda of universal access to electricity by the UN accompanied by other global institutions and emerging national development policies and the new goals and efforts regarding low carbon development, renewable mini-grids are becoming more and more important. The Inter- national Energy Agency publishes statistics regarding urban, rural, and total un-electrified population per country (IEA, 2014b). In addition, a newly developed multitier framework also includes the quality of electrification, such as the duration of availability of electricity per day and their affordability (Worldbank/ESMAP, 2014).

Solar–battery–diesel mini-grids are a possible solution for providing high quality access to electricity in different regions, such as Laos (Blum et al., 2015), Burkina Faso (Ouedraogo et al., 2015), or Nigeria (Dada, 2014; Ohiare, 2015) and Cameroon (Nfah et al., 2010; Nfah, 2013).

Hazelton et al. (2014)provide a comprehensive overview of beneﬁts

⁎ Corresponding author.

E-mail address:catherina.cader@rl-institut.de(C. Cader).

http://dx.doi.org/10.1016/j.esd.2015.12.007

Contents lists available atScienceDirect

Energy for Sustainable Development

(2)

and risks of hybrid mini-grids.Bhattacharyya (2013)discusses challenges of decentralized options for rural electriﬁcation from a multi- disciplinary perspective. The expansion or so-called hybridization of existing diesel grids with renewable energy systems could signiﬁcantly reduce the electricity costs and the emission of air pollutants (Dekker et al., 2012). IEA's New Policies Scenario suggests the generation of 26 TWh for only Sub-Saharan Africa by mini-grids until the year 2040.

The two most prominent technologies to achieve that are solar photovoltaic (PV) and oil, followed by hydro power (Fig. 1).

Within this study, a comparable analysis is carried out to evaluate where mini-grids powered by solar PV and diesel will be the most economical solution. This paper presents a novel methodology on how to scale up a local mini-grid potential analysis at one speciﬁc site to a global comparative assessment for solar–battery–diesel mini-grids by combining a GIS-based approach with an energy system simulation. The paper provides an overview to the following discussions:

1. Where will solar–battery–diesel systems be more economical than pure diesel systems under actual technology costs in the developing world?

2. Which percentage of solar power should be installed in the optimized hybrid systems and which parameters inﬂuence that?

3. What is the optimum size of batteries in the systems and how is it re- lated to low carbon development?

Decentralized systems allow electricity generation without the need for a national grid. In this study, the economic performance of solar– battery–diesel systems is compared to diesel-only systems, as the latter are frequently used in many regions (Bertheau et al., 2014).

Advantages of diesel generators are low investment costs, simple operation, and a well-known technology. Recent calculations of an analysis in Nigeria show that they are the optimum choice in some locations if compared to grid extension or small-scale PV solutions (Ohiare, 2015). Furthermore, fossil fuelled generators can be run flexibly and therefore allow varying electricity demands. However, diesel generators contribute to local and global environmental pollution (Ramanathan and Feng, 2009). In addition to purefinancial ben- efits, also the aspects of CO2savings and independency of fossil fuel supply are reflected with an establishment of hybrid mini-grids.

Solar–battery–diesel mini-grids are chosen as an alternative to pure diesel powered generations as many areas with low electriﬁcation rates are located in areas where a high photovoltaic potential can be expected (Breyer et al., 2011). Solar irradiation as a local resource holds the advantage of overall spatial availability, which is only limited by climatic factors as well as the day and night rhythm. With regard to the steep learning curves and achieved grid parity of small renewable

energy systems under certain scenarios (Schleicher-Tappeser, 2012;

Breyer and Gerlach, 2013), investigations in the potential of solar energy (photovoltaic) are the key to achieve global electriﬁcation. Due to the progress in the technical development, generation costs of renewable energy drop, thus becoming more and more economically feasible. In addition, battery costs are declining fast, which makes storage options much more viable today (Juelch et al., 2015; Nykvist and Nilsson, 2015).

In the past few years, more and more pilot projects and detailed mini- grid case studies were developed, implemented, and analyzed. This provides helpful insights in local environments and performance of in- dividual systems. However, to get an overview of the overall potentials of mini-grids, a broader research framework is necessary. This global comparison of potentials is especially helpful when it comes to business model development for different regions. The utilization of GIS allows an in-depth study of a large spatially disaggregated data set. In particular, when it comes to renewable energies, local resource variation needs to be accounted for. As diesel prices vary on a national level and are also subject to transport costs, these local characteristics need to be taken into consideration. GIS tools allow the processing of large data sets to account for the preparation of automation to optimize LCOE-based mini-grids for many different locations at once.

Methodology and input data

The work described in the paper is based on the approach described inFig. 2. Data sources are listed inTable 1. In thefirst step of the analysis a GIS-based approach is chosen to select all countries with rural electrification rates of 90% or below based on 2013 data (IEA, 2014b). For these countries, a high spatial resolution raster is defined. The raster cell size (length of sides) is set to 1/6 degree. The pixel size corresponds to about 18 km at the equator. This value is chosen to reflect the recom- mended grid resolution which compromises between the coarsest and finest legible grid resolution (Hengl, 2006). It is adopted to correspond to the spatial properties of the input data and still allow a simulation in a conceivable time frame. The sample countries are classified into 131,147 pixels in total. For each of these raster cells (pixel) the input parameters for the simulation (hourly solar irradiation, travel time, national diesel price, and transport costs) were extracted for the centre coordinates from global data bases. This automation allows the bulk processing of the spatially dissolved simulations.

The calculations aim at the distinction of two different scenarios:

1. Areas where the use of renewable energies applying a solar–battery– diesel system is more cost-effective than the use of conventional diesel generators and which would produce less emissions.

2. Areas where diesel-generated electricity is the cheapest option because here the investment in renewable energy technologies might be difﬁcult due to a lower potential and existing cheap alternatives.

Model formation to calculate electricity costs for the two scenarios, diesel stand-alone systems and hybrid solar–battery–diesel systems, is carried out using a modeling tool developed by the RLI (Fig. 3). This tool simulates the techno-economic optimization based on LCOE. The design of the approach is based on the modeling approach bySzabo et al. (2013, 2011). Their spatial concept is adopted and extended from an African perspective to a global scale. Furthermore, input parameters and model assumptions were changed and updated. Diesel transport costs are calculated using a global travel time raster. The global travel time raster (Nelson, 2008) is analyzed and validated regarding the infrastructure data, which can be drawn from the VMap0 data derived from the Digital Chart of the World compiled by the Na- tional Imagery and Mapping Agency (NIMA) of the United States. The global travel time raster is a product of spatial infrastructure assess- ments, land cover analysis, and a digital elevation model among other input parameters. It therefore reﬂects the impact of missing in- frastructural elements, which is especially important for electricity Fig. 1.Projection of the role of mini-grids in the IEA New Policy scenario. Source (IEA,

2014a, 2014b).

C. Cader et al. / Energy for Sustainable Development 31 (2016) 14–23 15

(3)

generation and electricity access in remote regions. We assume that diesel is purchased at the national diesel price (Ebert, Metschies, Schmid, and Wagner, 2009) in the next major settlement.

The accessibility of each region within the countries is considered to reﬂect the additional costs for transporting oil to the remote regions for the use in generators (Szabo et al., 2011). The transport costs are calculated as a function of the respective national diesel price for the transport vehicle, the amount of transported diesel, and travel time (Eq.(1)). The variables are as follows:Pt, transport cost;Pd, national diesel price;c, diesel consumption of transport vehicle (l/h) = 12;t, travel time (h);V, volume of diesel transported (division to get the value per kilowatt hour); and rt, factor for the return travel of the

transporter = 2, reducing the weight of the transport costs to the value (k) = 0.4:

Pt¼ððrtP_dctÞ VÞ k ð1Þ

In the calculation, further costs like charges for the transport vehicle and labor costs for the service are not considered. In many cases, the fuel is transported by large companies that shift the incidental costs internally to achieve affordable prices. However, these costs still occur from the macroeconomic perspective.

Fig. 2.Workﬂow diagram: combining GIS with mini-grid simulation.

Table 1

Data sets and respective sources.

Data set Source

Rural electriﬁcation rates IEA World Energy Outlook 2014: Electricity data base (http://www.Worldenergyoutlook.org/resources/energydevelopment/energyaccess database/(accessed 24/05/2014).

Country borders Global administrative areaswww.gadm.org(accessed 24/05/2014).

Accessibility Nelson, A., 2008. Estimated travel time to the nearest city of 50,000 or more people in year 2000. Global Environment Monitoring Unit—Joint Research Centre of the European Commission, Ispra Italy. 2008,http://forobs.jrc.ec.europa.eu/products/gam/index.php(accessed 24/05/2014).

National diesel prices World Bank, Pump price for diesel fuel. Available online:http://data.worldbank.org/indicator/EP.PMP.DESL.CD(accessed 20/02/2015).

Solar irradiation Stackhouse P.W., Whitlock C.H., (eds.), 2009. Surface meteorology and Solar Energy (SSE) release 6.0 Methodology, NASA SSE 6.0, Earth Science Enterprise Program, National Aeronautic and Space Administration (NASA), Langley,http://eosweb.larc.nasa.gov/sse/

(4)

For the analysis, global solar irradiation values for a year on an hourly basis are prepared as these are required for calculating the PV yield from the irradiation value. The potential PV yield is estimated from the site-speciﬁc global horizontal irradiation (GHI) for optimallyﬁxed tilted mono-crystalline modules (Huld et al., 2008). An LCOE-based optimization is carried out for both options: the simulation of diesel-only mini-grids and solar–battery–diesel mini-grids (Fig. 3). The optimization yields at minimizing LCOE for both of the systems.

Global calculation of electricity costs for a solar–battery–diesel system and pure diesel system is carried out using a generic algorithm. Ini- tial PV costs are set to 1500€/kWp, battery costs are assumed to be 350€/kWh. Weighted cost of capital (WACC) is set to 10% per annum.

For the diesel component a genset efﬁciency of 0.33 l/kWh_elis adopted, the operational expenditure is set to 0.01 l/kWhel(Table 2).

The calculation of LCOE is based on Eq.(2), which is drawn from Short et al. (1995). Abbreviations are as follows: CAPEX, capital expen- ditures; CRF, capital recovery factor; WACC, weighted average cost of capital; N, project lifetime; OPEX, operation and maintenance expendi- tures per year; Costsfuel, cost of diesel per liter; Fuel, consumed diesel per year; and Elconsumed, consumed electricity per year. It accounts for CAPEX, OPEX, and the fuel price over the project life time and considers a capital recovery factor, which is represented as a function of weighted cost of capital and project life time (Eq.(3)). As a result, all occurring costs are annualized for the project duration:

LCOE¼CAPEXCRF WACCð ;NÞ þOPEXþCosts_fuelFuel

Elconsumed ð2Þ

CRF WACCð ;NÞ ¼WACCð1þWACCÞ^N 1þWACC

ð Þ^N‐1 ð3Þ

An hourly load proﬁle is necessary to account for electricity demand estimation.Zeyringer et al. (2015)) show that detailed input data can be

utilized to develop load demand estimations. This study uses a simpli- fied approach of using one standardized hourly load profile for all locations to achieve a better comparability of results (Fig. 4). The load profile is characterized by a strong evening peak, which is typical for rural regions with predominant agricultural activities. For the simulation, the 24-h load profile is concatenated 365 times to account for the annual consumed electricity in hourly resolution. Boundary conditions are that the load demand and the stability criterion, the spinning reserve, have to be met in every hour of the year.

Results Country selection

The country selection leads to a sample of 76 countries for the mini- grid potential analysis. A summary of results for each country is at- tached in the Annex.Fig. 5displays the relation between national and rural electrification rates as defined byIEA (2014a, 2014b)and indicates the absolute number of non-electrified population of each of the 76 countries. The country sample shows a great variety regarding the rural electrification rates of almost 0 up to 0.9 of the total population.

Lowest values are mostly found in Sub-Sahara Africa, with some countries having rural electrification rates less than 10%. Other countries are reaching higher overall electrification values (N0.5), whereas their huge population size still leaves many people in darkness. In particular, this applies for India, with more than 300 million non-electrified people.

Other countries with a high absolute number of people without access to electricity are Nigeria, Ethiopia, Bangladesh, Indonesia, and the Dem. Rep. Congo (Fig. 5). All sample countries account to 1245 million un-electriﬁed people. This number is close to the internationally stated number of 1.3 billion, which may include more than the chosen 76 countries.

Fig. 3.Schematic simulation approach.

Table 2

Technical parameters and cost assumptions.

Parameter Unit Value

Diesel generator efﬁciency % 30

Annual fuel price increase % 3

Battery round cycle efﬁciency % 90

Battery max. depth of discharge % 50

Battery life time Y 8

Battery C-rate kW/kWh 1/3

Capital expenditure diesel generator €/kW 0

Operational expenditure diesel generator—variable €/kWh 0.01

Capital expenditure PV €/kW 1500

Operational expenditure PV—ﬁxed % of CAPEX/y 2

Capital expenditure battery €/kWh 350

Operational expenditure battery—ﬁxed €/kWh/y 10

Project duration Y 20

WACC % 10 Fig. 4.Standardized load proﬁle to estimate the electricity consumption of rural areas.

Characteristic for that is the prominent evening peak.

(5)

Comparative assessment of solar–battery–diesel systems and pure diesel systems

Signiﬁcant potentials of solar–battery–diesel systems exist globally.

The overallﬁnding is that in most locations power supply with solar– battery–diesel mini-grids is economically feasible over pure diesel- based power generation. LCOEs are reduced by 14€ct/kWh on global average (Annex). Countries with the highest average LCOE reductions are the Central African Republic, Niger, Chad, Mali, and Malawi with

more than 40€ct/kWh.Fig. 6displays local LCOE reductions, showing clearly national imbalances resulting from varying national diesel prices. Sub- national differences in LCOE reduction, such as in Columbia, Chad, or Indonesia are derived from varying diesel prices due to a high variance in transport costs reﬂecting a varying accessibility within the respective country. Under the assumed national diesel prices and technology costs, pure diesel systems only dominate where diesel fuel prices including transport are below 50€ct/kWh. Also increasing prices in remote areas due to higher transportation costs are observable (Fig. 6).

Fig. 5.Relation of rural and national electriﬁcation rate and accumulated number of un-electriﬁed population per country.

Fig. 6.LCOE reduction of solar–battery–diesel mini-grids compared to pure diesel systems. Given are the results for the three geographical regions South America, Sub-Saharan Africa, and Asia. The histogram shows the respective frequency distribution.

(6)

The latter is especially recognizable in large remote areas such as Peru or Mongolia. Local diesel prices (national diesel prices plus transport costs) are in most locations between 1 and 2€/l. Adding respective transport costs makes systems with high shares of PV and batteries attractive. Good solar resources in the case study countries make PV feasible in most locations. All analyzed countries except North Korea and Mongolia show higher solar yield values than 1500 kW/m/year. 24 countries are showing values higher than 2000 kW/m/year (Annex).

Generally, without batteries, about 30%–45% solar energy proportion in the hybrid systems are optimal (Fig. 7). When increasing the solar energy share due to high diesel prices and good solar resources, a higher battery capacity becomes necessary to shift the required energy from diesel to PV or to the battery, as PV is not available in the evening and at night. With the strong peak load in the evening (Fig. 4), batteries are indispensable and become competitive if diesel prices are high. As a consequence, the battery size is clearly correlated to the PV share in the system.

Fig. 8illustrates that large battery capacities are only installed in regions with high PV penetration. Regarding the PV share and batteries, a general rule of thumb is that batteries are getting important when the PV share reaches more than 45% of the installed capacity.

With the assumed load and battery costs, the transition point where batteries become necessary to allow load shifting into the evening hours can also be drawn fromFig. 8. The majority of regions do not require the installation of large batteries to be competitive to the pure diesel systems. The results of this study can be verified with thefinding from (Narula et al., 2012). Their research focus on the electrification of Southern Asia and conclude that the cost of electricity is decreasing when more and more distributed electricity generation is included in the overall system.

Discussion

Inﬂuence of national diesel prices

The global results show the impact of national diesel prices, which are inﬂuenced by national taxes and subsidies. Oil producing countries, such as Colombia, Nigeria, and Angola, do not achieve high cost savings because their local costs of renewable hybrid solutions do not compete with the low diesel prices. If the diesel costs are subsidized by the gov- ernment, the costs for the national economy are increasing. High subsidies are generally not sustainable over the long term and the operating life of a generation genset (Zerrifﬁ, 2011). There can be differentiated between subsidies on CAPEX or OPEX. When aiming at a possible self- sustainable system, subsidies for CAPEX should be favoured over subsidies for OPEX. The overall use of subsidies has to be carefully assessed, also considering the future development. Import tariffs on solar panels in certain countries augment CAPEX. The strong dependency and signif- icance of the diesel price show that the system is connected to political decisions and fossil fuel pricing. In particular, in regard to a future outlook with assumed raising diesel prices the cost advantages of the solar–battery–diesel systems become apparent.

Impact of travel time

In very remote locations, the advantage of being independent from fossil through systems with almost 100% renewable share en- sures a stable energy supply, as these regions are difﬁcult to access and hence are prone to fuel shortages and delays on fuel deliveries.

The transport costs, which are added to the national diesel prices, may be overestimated in the study, as often large companies sell

Fig. 7.Solar energy share of the produced energy. Given are the results for the three geographical regions South America, Sub-Saharan Africa, and Asia. The histogram shows the respective frequency distribution.

(7)

their diesel fuel for lower prices in these regions and compensate for the higher costs internally.

Two concepts:“fuel saver”and renewable energy systems

Regarding the optimum share of photovoltaics in the system, two main concepts can be differentiated: With low shares of PV (below 40% of installed capacity), the design can be described as“fuel saver” (Davies, 2014). Here batteries are mainly included for stability reasons and powerﬂuctuations of photovoltaic, such as through a sudden cloud cover. In addition, these systems reduce fuel consumption during the day as the demanded energy is directly produced by the photovoltaic panels up to their installed capacity. If larger shares of solar energy in the system are desired, larger batteries are required to allow the usage of PV generated energy in the evening and at night. Hereby larger CAPEX occur for large battery installations. OPEX on the other hand will decrease, as less diesel fuel is required.

Limitations

The study is limited by focusing only on solar–battery–diesel and pure diesel systems as technology. Smaller solutions, such as solar home systems or pico hydro are not included just as the calculation and feasibility of grid extension. Furthermore, a generic comparison from a bird's eye view does not consider local characteristics as much as a local case study will do. That means the next steps for building mini-grids require in-depth feasibility studies. For example, detailed load estimations are required for exact load estimation at each location.

Market potential analysis on a global scale allows the inclusion of other global indices in the analysis. Thus, the attractiveness of a market can be estimated for international investors or companies looking for regions where they can expand their services. Innovative business models for

the application of small distributed energy generation systems in rural areas are required (Zerrifﬁ, 2011). Possible challenges regarding the op- portunities of small power producers need to be addressed and policies need to be designed by energy regulations in the respective countries (Tenenbaum et al., 2014). Transmission grid development will be the main electriﬁcation pathway in total numbers, especially in urban areas. Therefore, a next step will be a grid extension cost modeling—to contrast grid extension with decentralized options like solar home systems or mini-grids, comparable to the work ofZeyringer et al. (2015)for the example of Kenya. With such an approach on a global scale and with including detailed electricity infrastructure data, the calculation of the absolute potential of solar–battery–diesel mini-grids is possible.

The strong pilot project nature of mini-grid development necessarily needs up scaling; therefore, the development of business models is required (Chaurey et al., 2012). By designing them, strong focus should be set to the sustainability of the project. As an example, maintenance structures are needed. Remote accessibility is also a problem for transport of technology for decentralized approaches. Many companies are afraid to consider these due to logistic risks. Future population growth rates will increase and aggravate the situation of many un-electriﬁed areas.

Conclusion

The developed methodology allows the depiction and comparison of different electriﬁcation pathways over a large geographical scale. GIS as tool for the input parameter preparation and visualization of results on the one hand and the simulation for comparing the costs of the two options in a detailed way on the other hand allow this complimentary approach. After a careful consideration of local resources, accessibility, and technology costs a large potential of hybrid solar–battery–diesel systems compared to diesel-only systems is evaluated. In particular, the Fig. 8.Installed battery capacity. Given are the results for the three geographical regions South America, Sub-Saharan Africa, and Asia. The histogram shows the respective frequency distribution. Regions without installed battery capacity (pixel number 88,000) exceed diagram dimensions.

(8)

combination of about one third PV and two thirds diesel generation is the LCOE-optimized system conﬁguration. Only with very high diesel prices in highly remote areas a larger share of PV in combination with batteries becomes attractive. Pure diesel systems are economic in countries with low national diesel prices today. With the dynamic development in renewable energy and storage technology, these results may change in future towards more renewable energy and less diesel fuel consumption.

Acknowledgment

The present study was only realizable with funding from cdw Stiftungsverbund gGmbH (formerly SMA Stiftungsverbund gGmbH) and Reiner Lemoine-Stiftung. The authors would like to thank these two organizations for supporting and enabling the current research.

Also, the authors would like to thank Markus Hlusiak for the initializa- tion of theﬁrst version of the simulation tool.

Appendix A. Annex

Indicator Population without

electricity

National electriﬁcation rate

Rural electriﬁcation rate

Local diesel price

Average LCOE

Average LCOE reduction

Solar irradiation

Unit Million % % ¤ct/l ¤ct/kWh ¤ct/kWh kWh/m/y

Source IEA IEA IEA GIZ model model NASA/DLR

Angola 15.0 0.30 0.06 0.69 0.20 0.03 1992

Argentina 1.5 0.96 0.61 1.50 0.38 0.13 1658

Bangladesh 62.0 0.60 0.48 0.91 0.26 0.05 1665

Benin 7.0 0.28 0.06 1.41 0.35 0.12 1905

Bolivia 1.2 0.88 0.66 0.63 0.19 0.02 1853

Botswana 1.0 0.66 0.51 1.05 0.27 0.08 2034

Burkina Faso 14.0 0.16 0.02 1.38 0.34 0.12 2017

Burundi 9.0 0.10 0.07 1.89 0.44 0.19 1793

Cabo Verde 0.0 0.94 0.84 1.39 0.35 0.12 1781

Cambodia 10.0 0.34 0.18 1.58 0.38 0.15 1833

Cameroon 10.0 0.54 0.17 1.59 0.37 0.16 2130

Cen. Afr. Rep. 4.0 0.03 0.01 2.77 0.46 0.46 1964

Chad 12.0 0.04 0.00 2.30 0.40 0.37 2180

Colombia 1.4 0.97 0.89 1.78 0.39 0.20 1704

Comoros 0.0 0.45 0.35 1.77 0.39 0.20 2176

Congo 3.0 0.35 0.05 1.38 0.35 0.11 1746

Dem. Rep. Congo 60 0.09 0.01 1.25 0.33 0.09 1798

Djibouti 0.0 0.50 0.14 1.51 0.36 0.15 2168

Dominican Republic 0.4 0.96 0.90 1.35 0.34 0.11 1851

DPR Korea 18.0 0.26 0.11 0.43 0.15 0.00 1663

Ecuador 0.9 0.94 0.84 1.18 0.30 0.10 2057

El Salvador 0.5 0.93 0.82 1.51 0.38 0.12 1609

Equatorial Guinea 0.0 0.66 0.48 1.93 0.40 0.24 2106

Eritrea 4.0 0.32 0.17 1.17 0.30 0.10 2058

Ethiopia 70.0 0.23 0.10 1.17 0.31 0.08 1612

Gabon 1.0 0.60 0.34 1.26 0.32 0.10 1982

Gambia 1.0 0.35 0.02 0.97 0.27 0.06 1797

Ghana 7.0 0.72 0.52 1.15 0.3 0.09 1856

Guatemala 2.2 0.86 0.75 1.29 0.33 0.10 1889

Guinea 10.0 0.12 0.03 1.54 0.38 0.13 1929

Guinea-Bissau 1.0 0.20 0.06 0.98 0.26 0.07 1960

Haiti 7.3 0.28 0.08 1.27 0.32 0.10 1804

Honduras 1.1 0.86 0.75 1.04 0.28 0.07 1784

India 304.0 0.75 0.67 1.50 0.36 0.14 1740

Indonesia 60.0 0.76 0.59 1.13 0.29 0.09 2014

Ivory Coast 15.0 0.26 0.08 1.36 0.34 0.12 2043

Jamaica 0.2 0.93 0.87 1.40 0.36 0.11 1578

Kenya 35.0 0.20 0.07 1.42 0.35 0.13 1858

Laos 1.0 0.78 0.70 1.38 0.36 0.10 1635

Lesotho 2.0 0.28 0.17 1.77 0.39 0.20 2014

Liberia 4.0 0.02 0.00 2.24 0.42 0.32 2020

Madagascar 19.0 0.15 0.04 2.22 0.41 0.33 2078

Malawi 15.0 0.09 0.05 1.98 0.40 0.26 2027

Mali 11.0 0.27 0.12 1.70 0.41 0.16 1440

Mauritania 3.0 0.21 0.02 1.23 0.31 0.10 1942

Mongolia 0.0 0.90 0.73 0.89 0.25 0.05 1661

Mozambique 15.0 0.39 0.27 1.34 0.32 0.13 2115

Myanmar 36.0 0.32 0.18 1.34 0.32 0.13 1810

Namibia 2.0 0.30 0.17 1.32 0.33 0.11 1847

Nepal 7.0 0.76 0.72 2.28 0.38 0.37 2242

Nicaragua 1.6 0.74 0.50 1.05 0.28 0.08 1952

Niger 15.0 0.14 0.04 1.40 0.37 0.11 1381

Nigeria 93.0 0.45 0.35 1.25 0.31 0.11 1866

Pakistan 56.0 0.69 0.57 1.14 0.31 0.08 1684

Panama 0.4 0.89 0.63 1.72 0.39 0.19 1851

Peru 2.7 0.91 0.65 1.08 0.30 0.07 1748

Philippines 29.0 0.70 0.52 0.25 0.09 0.00 2037

Qatar 0.0 1.00 0.69 2.13 0.46 0.25 1695

(continued on next page) C. Cader et al. / Energy for Sustainable Development 31 (2016) 14–23 21

(9)

(continued)

Indicator Population without

electricity

National electriﬁcation rate

Rural electriﬁcation rate

Local diesel price

Average LCOE

Average LCOE reduction

Solar irradiation

Unit Million % % ¤ct/l ¤ct/kWh ¤ct/kWh kWh/m/y

Source IEA IEA IEA GIZ model model NASA/DLR

Réunion 0.0 0.99 0.87 1.86 0.38 0.24 2192

Rwanda 10.0 0.17 0.05 1.58 0.39 0.14 1773

Sao Tome &Principe 0.0 0.59 0.40 1.55 0.38 0.14 1865

Senegal 6.0 0.55 0.28 1.77 0.41 0.18 1981

Sierra Leone 6.0 0.05 0.01 1.27 0.34 0.09 1706

Somalia 9.0 0.15 0.04 1.29 0.32 0.11 2099

South Africa 8.0 0.85 0.82 1.30 0.32 0.11 1921

South Sudan 11.0 0.01 0.00 2.29 0.45 0.31 1957

Sri Lanka 2.0 0.89 0.88 0.98 0.27 0.06 1814

Sudan 24.0 0.35 0.21 0.63 0.18 0.03 2155

Swaziland 1.0 0.27 0.24 1.50 0.38 0.13 1720

Syria 1.6 0.93 0.84 0.60 0.18 0.03 1818

Tanzania 36.0 0.24 0.07 1.39 0.34 0.12 2008

Togo 5.0 0.27 0.21 1.60 0.39 0.14 1851

Uganda 31.0 0.15 0.07 1.36 0.34 0.11 1953

Yemen 13.8 0.42 0.29 0.58 0.17 0.03 2231

Zambia 10.0 0.26 0.14 1.78 0.40 0.19 2065

Zimbabwe 8.0 0.40 0.14 1.06 0.28 0.08 1986

All countries ∑1,245 Ø0.49 Ø0.34 Ø1.39 Ø0.33 Ø0.14 Ø1891

References

Bertheau P, Cader C, Mueller H, Blechinger P, Seguin R, Breyer C. Energy storage potential for solar based hybridization of off-grid diesel power plants in Tanzania.

Energy Procedia 2014;46:287–93.http://dx.doi.org/10.1016/j.egypro.2014.01.

184.

Bhattacharyya SC. To regulate or not to regulate off-grid electricity access in developing countries. Energy Policy 2013;63:494–503.http://dx.doi.org/10.1016/j.enpol.2013.

08.028.

Blum NU, Bening CR, Schmidt TS. An analysis of remote electric mini-grids in Laos using the technological innovation systems approach. Technol Forecast Soc Chang 2015;

95:218–33.http://dx.doi.org/10.1016/j.techfore.2015.02.002.

Breyer C, Gerlach A. Global overview on grid-parity. Prog Photovolt 2013.http://dx.doi.

org/10.1002/pip.1254.

Breyer C, Werner C, Rolland S, Adelmann P.Off-grid photovoltaic applications in regions of low electriﬁcation: high demand, fastﬁnancial amortization and large market potential. Proceedings of 26th EU PVSEC; 2011. p. 5–9.

Cader C.Is a grid connection the best solution? frequently overlooked arguments assessing centralized electriﬁcation pathways. Micro Perspectives for Decentralized Energy Supply: Proceedings of the International Conference (2015, Bangalore);

2015. URL:https://opus4.kobv.de/opus4-tuberlin/frontdoor/index/index/docId/

6281.

Chaurey A, Krithika P, Palit D, Rakesh S, Sovacool BK. New partnerships and business models for facilitating energy access. Energy Policy 2012;47:48–55.http://dx.doi.

org/10.1016/j.enpol.2012.03.031.

Chaurey A, Ranganathan M, Mohanty P. Electricity access for geographically disadvan- taged rural communities—technology and policy insights. Energy Policy 2004;32:

1693–705.http://dx.doi.org/10.1016/S0301-4215(03)00160-5.

Dada JO. Towards understanding the beneﬁts and challenges of smart/micro-grid for electricity supply system in Nigeria. Renew Sustain Energy Rev 2014;38:1003–14.http://

dx.doi.org/10.1016/j.rser.2014.07.077.

Davies D. Reducing costs via solar fuel savers. Renew Energy Focus 2014;15:26–7.http://

dx.doi.org/10.1016/S1755-0084(14)70042-0.

Dekker J, Nthontho M, Chowdhury S, Chowdhury S. Economic analysis of PV/diesel hybrid power systems in different climatic zones of South Africa. Int J Electr Power Energy Syst 2012;40:104–12.http://dx.doi.org/10.1016/j.ijepes.2012.02.010.

Doll CN, Pachauri S. Estimating rural populations without access to electricity in developing countries through night-time light satellite imagery. Energy Policy 2010;38:

5661–70.http://dx.doi.org/10.1016/j.enpol.2010.05.014.

Ebert S, Metschies DrGP, Schmid D, Wagner A. International Fuel Prices 2009 6th Edition.

Eschborn, Germany: GIZ; 2009.https://www.giz.de/expertise/downloads/gtz2009- en-ifp-full-version.pdf.

Hazelton J, Bruce A, MacGill I. A review of the potential beneﬁts and risks of photovoltaic hybrid mini-grid systems. Renew Energy 2014;67:222–9.http://dx.doi.org/10.1016/j.

renene.2013.11.026.

Hengl T. Finding the right pixel size. Comput Geosci 2006;32:1283–98.http://dx.doi.org/

10.1016/j.cageo.2005.11.008.

Huld T, Suri M, Dunlop E. Geographical variation of the conversion efﬁciency of crystalline silicon photovoltaic modules in Europe. Prog Photovolt Res Appl 2008;16:595–607.

http://dx.doi.org/10.1002/pip.846.

IEA. Africa Energy Outlook. 2014. URL:https://www.iea.org/publications/freepublications/

publication/WEO2014_AfricaEnergyOutlook.pdf.

IEA. Weo 2014 Electricity access database. 2014. URL:http://www.worldenergyoutlook.

org/resources/energydevelopment/energyaccessdatabase/.

Juelch V, Telsnig T, Schulz M, Hartmann N.A holistic comparative analysis of different storage systems using levelized cost of storage and life cycle indicators. Conference paper for the 9th International Renewable Energy Storage Conference (IRES); 2015.

p. 1–12.

Kanagawa M, Nakata T. Assessment of access to electricity and the socio-economic impacts in rural areas of developing countries. Energy Policy 2008;36:2016–29.http://

dx.doi.org/10.1016/j.enpol.2008.01.041.

Kaundinya DP, Balachandra P, Ravindranath N. Grid-connected versus stand-alone energy systems for decentralized power—a review of literature. Renew Sustain Energy Rev 2009;13:2041–50.http://dx.doi.org/10.1016/j.rser.2009.02.002.

Narula K, Nagai Y, Pachauri S. The role of decentralized distributed generation in achiev- ing universal rural electriﬁcation in south Asia by 2030. Energy Policy 2012;47:

345–57.http://dx.doi.org/10.1016/j.enpol.2012.04.075.

Nelson A. Travel time to major cities: a global map of accessibility. Technical Report Global Environment Monitoring Unit—Joint Research Centre of the European Commission, Ispra Italy; 2008. (URL:http://forobs.jrc.ec.europa.eu/products/

gam/index.php.

Nfah EM. Evaluation of optimal photovoltaic hybrid systems for remote villages in far north Cameroon. Renew Energy 2013;51:482–8.http://dx.doi.org/10.1016/j.renene.

2012.09.035.

Nfah EM, Mucho NJ, Godpromesse K. Economic evaluation of small-scale photovoltaic hybrid systems for mini-grid applications in far north Cameroon. Renew Energy 2010;

35:2391–8.http://dx.doi.org/10.1016/j.renene.2010.03.005.

Nykvist B, Nilsson M. Rapidly falling costs of battery packs for electric vehicles. Nat Clim Change 2015.http://dx.doi.org/10.1038/nclimate2564.

Ohiare S. Expanding electricity access to all in Nigeria: a spatial planning and cost analysis. Energy Sustain Soc 2015;5.http://dx.doi.org/10.1186/s13705-015- 0037-9.

Ouedraogo BI, Kouame S, Azoumah Y, Yamegueu D. Incentives for rural off grid electriﬁ- cation in Burkina Faso using LCOE. Renew Energy 2015;78:573–82.http://dx.doi.org/

10.1016/j.renene.2015.01.044.

Ramanathan V, Feng Y. Air pollution, greenhouse gases and climate change: global and re- gional perspectives. Atmos Environ 2009;43:37–50.http://dx.doi.org/10.1016/j.

atmosenv.2008.09.063.

Schleicher-Tappeser R. How renewables will change electricity markets in the next ﬁve years. Energy Policy 2012;48:64–75.http://dx.doi.org/10.1016/j.enpol.

2012.04.042.

Short W, Packey DJ, Holt T.A manual for the economic evaluation of energy efﬁciency and renewable energy technologies; 1995.

Szabo S, Bodis K, Huld T, Moner-Girona M. Energy solutions in rural Africa and mapping electriﬁcation costs of distributed solar and diesel. Environ Res Lett 2011;6.http://

dx.doi.org/10.1088/1748-9326/6/3/034002.

Szabo S, Bodis K, Huld T, Moner-Girona M. Sustainable energy planning: leapfrogging the energy poverty gap in Africa. Renew Sustain Energy Rev 2013;28:500–9.http://dx.

doi.org/10.1016/j.rser.2013.08.044.

Tenenbaum B, Greacen C, Siyambalapitiya T, Knuckles J. From the bottom up:

how small power producers and mini-grids can deliver electriﬁcation and renewable energy in Africa. Washington, DC: World Bank; 2014.http://

dx.doi.org/10.1596/978-1-4648-0093-1 (URL: http://hdl.handle.net/10986/

16571).

Appendix A(continued)

(10)

Winkler H, Simones AGF, Rovere ELL, Alam M, Rahman A, Mwakasonda S. Access and affordability of electricity in developing countries. World Dev 2011;39:1037–50.http://

dx.doi.org/10.1016/j.worlddev.2010.02.021.

Worldbank/ESMAP. Capturing the multi-dimensionality of energy access. Livere wire—a knowledge note series for the energy & extractives for global practice. 2014. URL:

http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2015/

02/27/090224b082b6d2b4/2_0/Rendered/PDF/Capturing0the00ity0of0energy0access.

pdf.

ZerrifﬁH. Rural electriﬁcation. Springer Netherlands; 2011.http://dx.doi.org/10.1007/

978-90-481-9594-7.

Zeyringer M, Pachauri S, Schmid E, Schmidt J, Worrell E, Morawetz UB. Analyzing grid extension and stand-alone photovoltaic systems for the cost-effective electriﬁcation of Kenya.

Energy Sustain Dev 2015;25:75–86.http://dx.doi.org/10.1016/j.esd.2015.01.003.

Zvoleff A, Kocaman AS, Huh WT, Modi V. The impact of geography on energy infrastructure costs. Energy Policy 2009;37:4066–78.http://dx.doi.org/10.1016/j.

enpol.2009.05.006.