Assessment of global solar resource development

0 Introduction

Climate change has become a challenge to the sustainable development of human society.In addition,there has been a global consensus that a green energy transition is needed to control carbon emissions and respond to this climate change.Solar energy resources are an important part of clean energy,and there is a need to analyze how to develop these to promote green energy transition.This study constructed a global renewable-energy exploitation analysis(GREAN)platform,which was comprised of a basic database,series of assessment models,and visual software tool.This platform included the global horizontal irradiance(GHI)and direct normal irradiance(DNI)produced by the solar geographic information system(GIS).The development of solar energy resources is generally based on the principle of prioritizing the development of highquality resources with good conditions and low average development cost(ADC).This study creatively established a refined model that can utilize the capacity factor and ADC of solar power energy as two indicators to accurately show the distribution characteristics of the technical installed potential capacity(TPIC)and development potential.

Analyzing the “location,” “quantity,” and “quality” of global solar energy resources is the foundation for the large-scale development and utilization of solar energy.Much work has been done on the analysis of renewable energy resources from a geographical point of view.For example,weather station data or regional simulation data from specific provinces,regions,cities,and photovoltaic(PV)bases have been analyzed to research the development of PV resources or the locations of PV power plants[1-5].Other studies[6-10]have assessed the resources in Brazil,Vietnam,India,and other countries to find the most suitable site locations for developing solar energy resources.However,the above research was focused on regional level solar energy development and power plant construction,with none including the quantitative assessment of global solar resources.

From the perspective of analytical accuracy,some studies[11-14]have used TRs to estimate the renewable resource potential.However,TRs only reflect the differences and distribution of renewable resources to a limited degree,because power generation technologies,land cover,terrain distribution,and other restrictive factors also affect the results of renewable energy resource assessments.Furthermore,the TPIC can more accurately reflect the development potential of renewable resources in a region,and it has practical value for strategy and planning research.Therefore,an increasing number of experts and scholars have used the TPIC to conduct resource evaluation research[15-19],but none have calculated the ADC of developing renewable energy resources.Some studies[20-23]have analyzed areas in different countries that are unsuitable for developing solar energy,including urban and cultivated areas.They also provided the evaluation results for PV development,but the development costs were not mentioned.Some scholars have made economic calculations for solar energy projects using indicators such as the levelized cost of energy(LCOE)[24-25],and the literature[28]includes an economic analysis of the crossregional parity of centralized PV utilization in western China.However,when that study calculated the LCOE,it only considered the network fee without the grid integration cost and off-site transportation cost.Moreover,only the centralized development mode was analyzed,without a coanalysis of the distributed development mode.

The main contributions of this paper on solar energy resource assessment include the following.First,a database was established that included solar energy resource information,global geographic information,and human activity information that could be used for quantitative evaluations and calculations,as well as the integration of data with different resolutions.Second,an assessment method for solar energy resources was established,and global solar energy resource assessment research was performed to develop a system that can compare assessment results,including the theoretical reserves(TRs),TPIC,and ADC.Third,the study focused on the two core indicators of the capacity factor and ADC to calculate the distribution characteristics of the global TPIC and realize a quantitative evaluation of the refined development potential based on the data.Finally,in order to verify the rationality of the assessment results,the locations of already-built power plants were compared to the assessment results of this study.It was found that the assessment results for the solar energy resources and development potential could provide support for decision-makers in the energy transition for various regions and countries.

1 Assessment method

1.1 Basic data and computing technology

Solar radiation and geographic data are necessary parameters for conducting a solar energy resource assessment.In order to realize a digital and multi-dimensional assessment of solar energy resources,geographic information data such as the global land cover distribution,as well as data related to human activities such as global conservation areas,the transport infrastructure,and the power grid infrastructure were used in this work.A renewable energy resource assessment database was established that contained 16 items of data in three categories covering the whole world,as listed in Table 1[29].

Table 1 Basic data for renewable energy resource assessment

Info TypeData DescriptionSpatial ResolutionData Type ResourcesGlobal solar energy resource data9 km × 9 kmRaster data Geographic information Global classification information 30 m × 30 mRaster data Global distribution of major conservation areas/Vector data Global distribution of major reservoirs/Raster data Global distribution of lakes and wetlands1 km × 1 kmRaster data Global distribution of major geological faults/Vector data Global distribution of plate boundaries/Vector data Global distribution of historical seismic activity frequency5 km × 5 kmRaster data Global distribution of main rock types/Vector data Global terrain satellite images/Raster data Global terrain elevation data30 m × 30 mRaster data Global ocean boundary data/Vector data Human activities Global population distribution900 m × 900 mRaster data Global distribution of transportation infrastructure/Vector data Geographic distribution of global power grid/Vector data Global power plant information and geographic distribution/Vector data

The data mentioned above were massive,complex,unstructured,multi-temporal,multi-dimensional,and multi-source,which produced a series of challenges to quantitative calculations.Therefore,data preprocessing was conducted,which included data collation,data fusion,and data reduction.Among these data preprocessing steps,a geographic information calculation related to the land surface area and slope was a basic algorithm of the global renewable energy resource assessment system.Because of the different spatial resolutions,types,and formats of the basic data,it was necessary to solve two key problems:multi-resolution data fusion(MDF)and the computation of different types of data,in order to establish a unified resolution basic data set.

This study adopted a normalization method to transformpixel rasters with different resolutions into pixels with the same resolution,which included the following main calculation steps.First,the best resolution was selected,unifying the global coordinate system and calibrating initial points.This means that based on the actual situation of the global large-scale calculation,the resolution was set to 500 m × 500 m,and the global land area was divided into more than 600 million rasters for calculation.Second,a bilinear interpolation method was used to transform lowresolution data into high-resolution data,such as resource data,and data for the distribution of the historical seismic activity frequency and global population distribution,Then,a weighted average method was used to transform the highresolution data into low-resolution data,such as global geographic elevation data.

The 16 items of global basic information data mainly included two types of data:vector and raster data.In the geographic information calculations,it was necessary to perform mixed calculations for different types of data to realize the data fusion and quantitative analysis.Specifically,the mixed calculations fell into two categories.The first category was fixed vector data such as global conservation area distribution data,global reservoir distribution data,and global stratum distribution data.By rasterizing the vector data,a mixed calculation between vector data and raster data was realized.The second category was the ever-changing vector data.In the process of selecting the target area,the vector polygon data in such an area will change,and the partitioning of vector data and raster pixels will lead to an irregular neighborhood,which can generally be calculated using the center point exclusion method.

In this study on the strength of the extensive,massive volume of computable data,a systematic and quantitative resource assessment method for global solar energy was established based on three main indicators:the TRs,TPIC,and ADC.

1.2 Assessment method for TRs

The TR of solar PV resources is the sum of the solar energy received by the surface of a certain area and completely converted into electric energy,usually regardless of the loss of power generation conversion efficiency.A TR assessment can be performed by calculating the sum of each raster multiplied by its corresponding solar global horizontal irradiance(GHI).The formula for calculating the TR of PV resources(ETRPV)is shown below.

Here,GHIis the annual GHI corresponding to the grid;Aiis the area of the ithgrid;and nis the total number of grids in the selected area.

1.3 TPIC assessment method

The TPIC refers to the PV installed capacity scale that can be developed and utilized under the technical level and conditions of the assessment year.The key to the assessment of the TPIC is to exclude an unavailable area resulting from restrictions.The assessment analysis mainly includes three important processes,which are the calculations of the available area,effective installed capacity area,and installed capacity density.

The available area for PV resource development refers to the land area,excluding areas unsuitable for development because of restrictions such as those involving resource shortages,conservation areas,altitude,slope,and land cover.The calculation formula is as follows.

Here,Aavailableis the available area;Asumis the total assessment area;Areserveijis the jthconservation area of the ithtype in the assessment area;and Alow_resourcei,Ahigh_altitudei,Ahigh_gradienti,and Aground_objectiare the areas of regions that are unsuitable for development because of the restrictions related to resource shortage,altitude,slope,and land cover,respectively.

The effective installed capacity area for PV resource development can be obtained by setting the landuse coefficient,considering how different land covers compromise the actual development conditions within the available area.The calculation formula is as follows:

Where Aeffectiveis the effective installed capacity area,and ηarea is the land-use coefficient(i.e.,the land-use coefficient corresponding to different land covers).The main indicators and recommended land-use coefficients are listed in Table 2.

Table 2 Main indicators and recommended coefficients used in solar energy resource assessment of TPIC

Type Restrictive FactorThreshold ValueCentralized Development Parameters Distributed Development Parameters Resource restrictionsGHI＞1000 kWh/m2——Technical development restrictionsLand elevation＜ 4500 m——Consecration area restrictions Natural ecosystemNot suitable for development0%0%WildlifeNot suitable for development0%0%Natural relicsNot suitable for development0%0%Natural resourcesNot suitable for development0%0%Other conservation areasNot suitable for development0%0%Land cover restrictions ForestNot suitable for centralized development0%10%Cultivated landNot suitable for centralized development0%25%Wetland and swampNot suitable for development0%0%Urban areasNot suitable for centralized development0%25%Ice and snowNot suitable for development0%0%ShrubsSuitable for development50%0%Herbaceous and vegetationSuitable for development80%0%Bare groundSuitable for development100%0%Terrain slope restrictions＞30°Not suitable for development0%0%Infrastructure restrictionsAirport＞2 km——

Specifically,in an available area calculation,the first to be excluded are areas with poor solar energy resources.Based on engineering construction practice and the current level of photovoltaic module technology,it is generally believed that areas where the GHI is less than 1000 kWh/m2 are resource-deficient areas with unsatisfactory lighting conditions and poor cost-effectiveness,and are not suitable for photovoltaic development.The second to be excluded are all types of conservation areas,including natural ecosystem conservation areas and wildlife conservation areas.The third to be excluded are areas not suitable for large-scale development under the current technical conditions,such as plateaus with an altitude of more than 4500 m.Such areas are mostly covered by glaciers and have frozen soil all year round,which will affect project construction and result in huge technical difficulties in photovoltaic development and poor cost-effectiveness.Moreover,the fragile ecological environment on a plateau would make it difficult to restore the surface vegetation after the construction of a largescale project.The fourth to be excluded are areas that are not suitable for development because of the land cover,which can include forest,cultivated land,wetlands and swamps,cities and urban areas,and ice and snow,making them unsuitable for centralized development.However,in areas with cultivated land and in cities and urban areas,a reasonable option is to use fields,fish ponds,building roofs,and open spaces in industrial parks for distributed photovoltaic development.The fifth to be excluded are areas with a terrain slope greater than 30°,which are difficult to develop and have poor cost-effectiveness at the current technical level,and thus are not suitable for development.

Based on the photovoltaic development experience gained in various countries,the different land cover types vary in the degree that they are suitable for photovoltaic development.Following the principles of comprehensiveness,hierarchy,operability,sensitivity,and scientificity,and under the premise of sustainable development and ecological balance protection,this report proposes that three types of areas,including areas with shrubs,herbaceous vegetation,and bare ground,are suitable for centralized photovoltaic development,and the corresponding recommended values for the landuse coefficient are 50%,80%,and 100%,respectively.The developable installed capacity scale of distributed photovoltaic generation at the current technical level can also be obtained by adjusting the relevant parameters in combination with the specific conditions of the region to be assessed.The report suggests that distributed development could be considered for cultivated lands and cities and urban areas,with 10% and 25% as the recommended values for the land-use coefficient,respectively.

The TPIC assessment should consider the installed capacity density,which can be obtained by calculating the total power capacity of the PV generation equipment arrangement array per unit area based on the equipment parameters and the optimal arrangement principle of PV generation modules under the current technical conditions.The current mainstream 300 W monocrystalline silicon PV module was selected as a typical example for research,and the typical vertical and horizontal arrangement parameters for the PV array were given,as shown in Table 3.According to the estimations,the total power capacity of the PV modules in a single array was 26.4 W.The arrangement principles for arrays are as follows:arrays in the Northern Hemisphere should face due south and those in the Southern Hemisphere should face due north.The inclination of the array should equal the terrain slope angle of the grid when the terrain slope angle of the grid is greater than the local latitude and equal to the local latitude when the terrain slope angle of the grid is less than the local latitude.Based on the annual hourly solar energy resource data,the total solar irradiance on the tilted plane was calculated using the Klein method in this study[30].The array spacing should meet the requirement that the front and back rows of the array are not blocked by each other between 9:00 a.m.and 3:00 p.m.on the local winter solstice.The array spacing is affected by the inclination of the array,terrain slope angle of the grid,latitude,etc.

Table 3 Parameters of typical PV modules and arrays

Type of ModulePower per ModuleModule LengthModule WidthNumber of Vertically Arranged Arrays Number of Horizontally Arranged Arrays Monocrystalline silicon300 W1.956 m0.992 m422

The TPIC of the PV power generation of the grid,PTPGPV,can be obtained by calculating the sum of the product of the effective installed capacity area and installed capacity density of each geographic grid.The calculation formula is as follows:

where Aeffectiveiis the effective installed capacity area,and Punitis the installed PV capacity per unit area.

The technical potential annual power generation can be obtained by calculating the power of PV generation per hour on the basis of calculating the TPIC of the solar PV resources and considering the PV generation output loss caused by factors such as blockage,equipment wear and tear,and temperature.The specific calculation formulae are as follows:

where Ppvis the PV output sequence after considering the PV output power loss caused by the temperature,cover,and inverter and other equipment losses.The annual PV power generation can be obtained by summing the Ppvicorresponding to 8760 hours; GTIis the global tilted irradiance;ξis the temperature coefficient of the module power,which is generally considered to be -0.3%/℃;kis the loss coefficient of the PV array;Tis the local ambient temperature of the assessment area;and ETPGPVis the technical potential annual power generation.

1.4 ADC assessment method

The development cost was based on the estimation of the economic level of PV equipment by 2035,considering the transportation and power grid infrastructure conditions.In this work,the LCOE was adopted as an indicator when assessing the ADC.In the assessment process,each geographic grid was regarded as an independent calculation unit,and the corresponding kilowatt-hour cost was calculated separately,and compared with the preset comprehensive reference electricity price.The installed capacity and annual power generation of the cost-effective grids were accumulated according to the assessment area to obtain the economic potential installed capacity of solar power in the region.The calculation formula is as follows:

where PEPGPVis the economic potential installed capacity(EPIC)of solar power;PTPGPViis the TPIC of the ithgrid;and ηis the economic judgment factor.When the grid’s calculated LCOE was greater than the comprehensive reference price,the development was considered to be noneconomic,with a value of ηequal to 0,otherwise the value was 1.

The ADC was related to the resource conditions,solar energy development technologies,and policy environment,which affect the cost of power generation.Therefore,the ADC assessment needed to comprehensively consider the technical parameters,operating parameters,financial parameters,cost parameters,and policy factors.The technical parameters mainly included the PV installed capacity and annual power generation,while the financial parameters included the base discount rate,proportion of capital funds for a loan,loan interest rate,and load terms.The cost parameters should focus on the initial investment such as the equipment cost,construction cost,and grid integration cost.Different transmission modes and voltage levels are required for the grid integration of power supplies with different scales and distances,which cause very different corresponding cost levels.This work was based on Chinese practical experience.After obtaining the costs of different transmission modes and voltage levels,these were combined with geographic spatial analysis results to complete the grid integration distance calculation[27-28].Fig.1 shows the heat distribution for a grid infrastructure of 220 kV and above.The influence of the grid integration conditions in different regions on the resource development could be obtained by combining the transmission mode selection and cost factor measurement.In the same way,the off-site transportation cost took into account the distribution of transport facilities such as highways to measure the impact of the off-site access roads necessary for resource development on its ADC.

Fig.1 Schematic diagram of thermal distribution of power grid facilities in South America

1.5 Global renewable-energy exploitation analysis(GREAN)platform

In order to implement technical and economic assessments of solar resources from a global perspective,an integrated analysis platform called the global renewableenergy exploitation analysis(GREAN)platform was established by the Global Energy Interconnection Development and Cooperation Organization(GEIDCO).It includes a basic database,series of assessment models,and visual software tool.The basic database contains 16 items in 3 categories covering the whole world,which include resource data,geographic information data such as the global land cover distribution and terrain data,and data related to human activities such as global conservation areas,power grids,and transportation infrastructure distributions[29].Among these,a global solar energy resource database that included the GHI and direct normal irradiance(DNI)was produced using the solar GIS[30].A series of fine assessment models such as above seven functions mentioned has been integrated into the platform to perform the systematic calculations of three indicators,which are the TRs,TPIC,and ADC of solar energy.Fig.2 shows the GREAN platform interface and its function modules.

Fig.2 Main function modules of GREAN platform

2 Assessment results

This research combined geographic information processing,spatial analysis,big data calculation,and other technologies to complete the assessment of the global solar TRs,TPIC,and ADC.With respect to the ADC,2035 was used as the level year to select relevant equipment parameters and project cost prediction parameters[33-34].

2.1 TR assessment results

According to the estimation of the global horizontal irradiance data for solar energy,the theoretical reserves of global solar PV resources total 208 EWh/a.The proportions for different continents are shown in Fig.3,which were basically determined by the geographic latitude and land area.

Fig.3 Comparison of TRs of solar resources

The country with the highest TR is Africa.In Africa,the TR of solar PV resources is 63,505.48 PWh/a,accounting for 31% of the global total,and parts of northern,southern,and eastern Africa have excellent PV resources.In addition,there are excellent PV resources in west Asia,with the TR of the solar PV resources in Asia equaling 66,617.03 PWh/a,accounting for 28% of the global total.In contrast,the European and Oceania PV resources account for relatively small proportions of 5% and 8%,respectively.

2.2 TPIC assessment results

Considering the resources and various technical constraints,the global scale of the PV energy TPIC suitable for centralized development was assessed and estimated to be approximately 2650 TW,while that for distributed development was approximately 112 TW.The proportions on different continents are shown in Fig.4.Africa has the best conditions for centralized development,with a total installed capacity of 1,394 TW,accounting for approximately 50.5% of the world’s total.In general,considering the impact of resource endowments,land cover,terrain,conservation areas,and other factors,most of the land in Africa,except for the central part,meets the conditions for centralized development and the construction of PV bases.

Fig.4 Comparison of TPIC values of solar PV energy resources of continents

Asian PV power generation is mainly concentrated in Mongolia in East Asia,northern and western China,and Pakistan in South Asia.Asia has limited conditions for centralized development,with a total installed capacity of 608 TW,accounting for approximately 23% of the world’s total.For distributed development,because of the presence of many rain forests in Southeast Asia,and much arable land in East Asia and India,it is appropriate to adopt a distributed development model to develop PV resources using vacant land in the cultivated areas and urban roofs.Its distributed value is approximately 48 TW,accounting for approximately 42.8% of the world’s total.

However,because of factors such as land cover,only 8% of the land in Europe has the conditions for the centralized development and construction of PV power bases.In Europe,it has been estimated that the PV TPIC suitable for centralized development is approximately 10.6 TW,with a distributed value of approximately 7.1 TW.Some European countries are more suitable to develop PV resources using the distributed mode.

The ratio of annual power generation per unit of land area to installed capacity(i.e.,the number of installed capacity full-load hours)is also a key parameter reflecting the advantages of regional PV resource development conditions.Fig.5 gives a comparison of the full-load hours with different development modes for the continents.It can be seen that all seven continents,except Europe,have more than 1,700 full-load hours.Among them,Africa ranks first out of all the continents.The number of full-load hours of centralized development on all the continents is higher than that of the distributed development.

Fig.5 Comparison of full-load hours with different solar PV energy development modes of continents

The solar power capacity factor can reflect the quality of the technical conditions in a region,and is equal to the ratio of the full-load hours to the total annual hours(8,760).The distribution of the PV technical available areas and capacity factors in the world is shown in Fig.6,including both the centralized and distributed development modes.In general,the global PV TPIC is approximately 1,880 hours(with an average capacity factor of approximately 0.215),and a global maximum exceeding 2,500 hours occurs near Antofagasta in northern Chile,where the resource conditions are extremely superior.

Fig.6 Distribution of global technical available areas for solar photovoltaic generation and their photovoltaic capacity factors

In terms of the distribution on different continents,the average full-load hours of the PV TPIC in Asia are approximately 1,802 hours(with an average capacity factor of approximately 0.206).The maximum exceeding 2,100 hours occurs near Tabuk in northwestern Saudi Arabia.In Europe,the average full-load hours of the TPIC are approximately 1,345 hours(with an average capacity factor of approximately 0.154).The maximum exceeding 1,700 hours occurs in the southeast of Andalucia in Spain.The average full-load hours of the TPIC in Africa are 1,941 hours(capacity factor is 0.221).The maximum exceeding 2,100 hours occurs near Karasburg in southern Namibia.

2.3 ADC assessment results

Based on the economic level of the PV power generation equipment,engineering construction,transportation and grid infrastructure by 2035,this research produced a comparison chart for the PV ADC in 2035 on all the continents in the world.Fig.7 shows the average ADC for solar PV generation,including the centralized and distributed development modes.

Fig.7 Comparison of average ADC values for solar PV generation on continents

The average ADC of global PV resources is 2.79 cents,moreover the average ADC for centralized development is 2.77 cents,and the average ADC for distributed development is 3.26 cents.In terms of different continents,the average ADC of centralized PV generation in Asia is 2.47 cents,and the average distributed ADC is 3.23 cents.In Europe,the average centralized and distributed ADC values are 3.16 cents and 4.47 cents,respectively,with a total ADC of 3.65.In Africa,the average centralized and distributed ADC values are 2.87 cents and 2.94 cents,respectively.In North America,the average centralized and distributed ADC values are 2.53 cents and 3.49 cents,respectively.In Central and South America,the average centralized and distributed ADC values are 2.34 cents and 3.17 cents,respectively,with a total ADC of 3.40.Meanwhile,the average centralized and distributed ADC values in Oceania are 2.77 cents and 3.26 cents,respectively.

In terms of countries and regions,those with excellent resource conditions and relatively good transportation and power grid infrastructures have relatively low development costs and better cost-effectiveness for PV development.However,there are certain differences in some areas.Despite the excellent PV resources,the hinterland of the Sahara Desert,the interior of Australia,and other regions are far away from the load center and have poor power grid infrastructure conditions,which cause a high PV ADC.Fig.8 shows the global distribution of the ADC values.In Asia,the highest ADC in most countries and regions is less than 8 cents,which indicates that Asia as a whole has good conditions for large-scale development.The lowest ADC for PV generation in most Asian countries is less than 2.5 cents,among which the lowest cost of 1.64 cents is in Tabuk,Saudi Arabia.On average the United Arab Emirates has the lower average ADC of any country,namely,1.94 cents.In Europe,the highest ADC values in most European countries and regions are higher than 8 cents,and the lowest ADC values in nine countries,including Spain,Italy,Malta,Portugal,and Greece are less than 2.5 cents,among which the lowest ADC of 2.1 cents occurs in the southeast of Andalucia in Spain.In Africa,the most costeffective development region,the lowest PV ADC values in most countries are less than 2.5 cents,among which the lowest cost of 1.72 cents is found in the south of Karasburg in Namibia.In North America,the lowest ADC values in Mexico and the United Sates are less 2.5 cents.In southern California in the United States,the ADC is 1.89 cents.The lowest ADC in Oceania is 1.77 cents,which occurs in the northwestern part of Western Australia.

Fig.8 Distribution of ADC values for global solar resources

The conditions for the PV ADC in South America,Asia,and North America are relatively low,as a result of the lower grid integration and transportation costs,which produce lower average ADC values than the global average level.There are vast deserts in Africa and Oceania,so photovoltaic resources are abundant.Poor grid integration conditions have affected the average economic level of PV resource development.PV resources suitable for distributed development in Africa have an ADC of less than 3 cents,and their economic efficiency ranks first out of all the continents.This is because the PV development is mainly limited by the impact of ground cover rather than the resource itself.

3 Refined assessment of development potential

Fig.9 shows the distribution characteristics of the capacity factor of the global PV TPIC.It is can be seen that the PV capacity factor of the centralized TPIC is mainly distributed in the range of 0.19-0.24,and it has a peak value in the range of 0.21-0.23.The capacity factor of the distributed TPIC is mainly distributed in the range of 0.17-0.22 and has a peak value in the range of 0.19-0.21.Approximately 25.2% of the total TPIC has a capacity factor reaching 0.23(full-load hours of more than 2,000),with an amount of approximately 695 TW,which means the solar PV resources are of good quality.Moreover,approximately 4.1% of the PV resources can reach a capacity factor of more than 0.24(full-load hours of more than 2,100),with a total TPIC of approximately 114 TW,which shows the excellent development potential.

Fig.9 Distribution characteristics of capacity factors of global PV resources TPIC

Fig.10 shows the distribution characteristics of the ADC of the global PV TPIC.The distribution characteristics under the centralized development mode are mainly concentrated in the range of 1.8-3.9 cents,and the histogram has a shape similar to a “double hump curve,” reaching peaks at 2.1 cents and 3.5 cents.The distribution characteristics of the TPIC under the distributed mode appear to have a normal distribution,with values mainly in the range of 2.3-3.9 cents,and a peak at 2.8 cents.There are approximately 1,675 TW of global PV TPIC with an ADC of less than 3 cents,accounting for approximately 61%.Out of this,the distributed PV TPIC accounts for approximately 50 TW.In addition,there is a global PV TPIC of approximately 1,374 TW with an ADC of less than 2.5 cents,accounting for nearly 50%,and these parts of the PV resources show outstanding economy and remarkable potential.

Fig.10 Distribution characteristics of ADC of global PV TPIC

Considering the most economic development method,the global low-carbon goal can be satisfied by developing 10,920 GW of TPIC with an average ADC as low as 1.63 cents.Based on this,only approximately 200,000 km2 of land would be used,which is equal to 0.6% of the total global desert area(including shrubs,bare ground,and grassland).

4 Verification of assessment

In this study,the global power plant database released by the World Resources Institute in 2018 was used to verify the rationality of the assessment results[28].This database offers information on 5,424 PV power plants in 42 countries around the world.It can be used as a reference to compare and verify global solar resource assessment results.

A comparison of 4,604 of the 5,424 PV power plants located in the areas suggested by the assessment results of this study showed that the rationality of the site selection was 85%.Based on the locations of the established PV power plants and the assessment results of global TPIC’s ADC distribution,the ADC distribution characteristics of the TPIC in the areas of the 5,424 already established PV power plants and around the glove could be obtained and are shown in Fig.11.

Fig.11 Distribution characteristics of ADC of TPIC in already established PV power plants and world

A comparison of the distribution of the global TPIC with the already established power plants showed that the average ADC in the power plants was 2.25 cents,which was 19% lower than the global average ADC of 2.79 cents.The accumulative curve of the established power plants was steeper,indicating that the locations of the plants always occupied a larger proportion of low-cost PV resources.Therefore,this result was consistent with the practical site selection principles for PV plants,and also verified that the assessment of global solar energy resources proposed in this article was correct and reasonable.

5 Conclusions

First,this study established a basic database of global geographic information,solar energy resource information,and human activity information,which could be used for quantitative evaluations and calculations.Furthermore,it expanded the assessment of solar energy resources from TRs to the TPIC and ADC using information about the physical geography,human activities,and economics.

Second,considering the resources and various technical constraints,it was estimated that the global TIPC suitable for centralized development was approximately 2,648.95 TW and that for distributed development was approximately 112.35 TW.The average ADC of the global TPIC was 2.79 cents.The average cost of the TPIC suitable for centralized development was 2.77 cents,and the average cost of the TPIC suitable for distributed development was 3.26 cents.The global energy transition could be accomplished by developing 10,920 GW of the most economical PV TPIC,with an ADC of 1.81 cents/kWh.To do that would occupy approximately 1.22 million square kilometers of land area,which is equal to 4% of the total global desert(including land with shrub cover and bare ground).

Third,for the first time,the results of an analysis and calculation showed agreement between the distribution data for the already build PV plants and solar resource assessment results.This showed that the rationality of the site selection was 85%.

Finally,calculations showed that in addition to the already built PV plants,there are many solar energy bases with lower ADC values.Therefore,expanding the construction of the power grid will not only meet the domestic electricity demand,but will also allow other countries on the continent to develop cheaper renewable energy power plants.

In general,the result of this solar resource assessment is a good reference for different countries and continents when planning PV development.In the next step,energy development policies and the actual conditions of different countries,such as protected areas,military zones,and sea passages,will be combined to improve the accuracy of the resource assessment results and support the energy transition.

Acknowledgement

This work was supported by National Science and Technology Major Project(2018YFB0904000).

Declaration of Competing Interest

We declare that we have no conflict of interest.