Dunlop, E. D., Šúri, M., Huld, T. A. (2003). Photovoltaic Potential Assessment to Support Renewable Energies Growth in 10 EU Candidate Countries. In Gottschalg, R.. (ed.) Proceedings of the Conference C79 of the Solar Energy Society CREST “Photovoltaic Science, Applications and Technology”, Loughborough University (UK) 3.-4.4.2003, pp. 1007-1016.
Photovoltaic Potential Assessment to Support Renewable Energies Growth in 10 EU Candidate Countries E. D. Dunlop1, M. Šúri1,2, T. A. Huld1 European Commission Joint Research Centre, Institute for Environment and Sustainability, Renewable Energies Unit, TP 450, I-21020 Ispra, Tel: +39 0332 789090, Fax: +39 0332 789268, Italy 2 Institute of Geography, Slovak Academy of Sciences, Štefánikova 49, 814 73 Bratislava, Slovakia
[email protected],
[email protected],
[email protected] 1
Abstract: We present a GIS database of solar radiation and photovoltaic (PV) potential estimations of 10 European Union Candidate Countries created to support. The database was integrated with a web application to provide access also for a broad public. An application was developed to browse and query GIS maps and to do a simple calculation for any location in the region. The established web site provides access also to analyses and relevant documents (http://iamest.jrc.it/pvgis/pv/index.htm).
1
Introduction The European Commission is committed to develop strategies for renewable energy sources for both Member States and Candidate Countries (CC). Under the development of the initiative of the European Research Area, the JRC is responding to this commitment through integrated research and knowledge transfer instruments. The aim of the JRC in this domain is to enhance the integration of European activities (both EU and CC) in the field by creating a critical mass of researchers and establishing a scientific and technical reference system to support the energy strategy. To formulate effective national PV-support strategies in the EU CC, the basic estimations of the solar energy resource have to be made. These elaborations have to be available for all desired locations in a simple and understandable way. Our approach to this is aimed at developing a solar radiation database and standard maps of potential PV power generation using the technology of geographical information systems. The databases and maps were integrated within a web application to provide access also to the general public. The considered region covers 10 EU candidate countries: Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia and Slovenia.
2
Developing GIS solar radiation database Solar radiation data are essential for planning and designing a PV installation. These data are measured at a limited number of climatic ground stations and before using they need to be processed by climatologists. To provide the information as spatially continuous (i.e. in the form of maps) the geographical information systems (GIS) are used. The GIS is a combination of computer hardware and software that is designed to manage, process, analyse and visualise georeferenced data. The GIS database of solar radiation was developed using a solar radiation model and interpolation techniques, both developed within the GIS GRASS [1]. The model equations are conceptually based on the results published in the European Solar Radiation Atlas (ESRA) [2]. It estimates beam, diffuse and reflected components of the clear-sky and overcast global irradiance and irradiation on a horizontal and inclined surfaces. The total daily irradiation values [kWh.m-2] are computed by the integration of the irradiance values [W.m-2] calculated at a selected time step and summarised between the sunrise and sunset. The model also accounts for sky obstruction (shadowing) by the local terrain features. The main input parameters, used in the computation, were as follows: • Solar irradiation measured at 182 ground stations over the region, representing the time period of 1981-1990 (available in the ESRA database); • Linke turbidity values derived from the SoDa web service [3]; • Digital elevation model with a raster resolution of 1 sq. km derived from the USGS GTOPO30 data; The developed GIS database contains raster layers with resolution of 1 sq. km of monthly and annual means of global irradiation [Wh.m-2.day-1] that were calculated for PV modules at a horizontal position as well as for those inclined southwards at angles of 15, 25, 40 and 90°. The database includes time series of Linke atmospheric turbidity [dimensionless] and a ratio of diffuse to global irradiation. The database also includes monthly and yearly averages of optimum inclination angle [degrees] of PV modules to harvest a maximum of the available irradiation. Finally solar radiation was computed for the yearly average of the optimum inclination angle of PV modules. In the calculation also the shadowing effect of the terrain features (considering the spatial resolution of the data) was considered. The solar radiation model using the digital elevation data improved the radiation
Dunlop, E. D., Šúri, M., Huld, T. A. (2003). Photovoltaic Potential Assessment to Support Renewable Energies Growth in 10 EU Candidate Countries. In Gottschalg, R.. (ed.) Proceedings of the Conference C79 of the Solar Energy Society CREST “Photovoltaic Science, Applications and Technology”, Loughborough University (UK) 3.-4.4.2003, pp. 1007-1016.
estimations, especially in regions with lower density of ground measurements. The details about the used methodology are described in the project web site are in [4].
3
Map-based estimation of PV power production The solar radiation database was used to produce maps of annual potential electricity generation E [kWh] assuming technical parameters of a household grid-connected 1 kWp system with modules inclined at different inclination angles. The estimation was calculated as: E = 365 Pk rp Gi,h where Pk (kW) is the peak power installed, rp is the system performance ratio (typical value is 0.75) and Gi,h is the monthly or annual mean of daily global irradiation on the horizontal or inclined solar module. The PV potential maps were overlaid with a data layer of residential areas, mapped in CORINE Land Cover database [5] as discontinuous urban fabric (class 112) to exclude areas outside of settlements. An average potential generation of a 1kWp system was calculated on the level of administrative regions and on a national level. Lastly, we have elaborated estimations showing a total possible energy production per region and per country, assuming a theoretically homogeneous dispersion of 100 PV systems per 1 km2 in the residential areas, each system assuming to have installed capacity of 2.5 kWp. The results of this stage consist of maps of the annual average potential PV production (kWh), considering the defined array geometries of the grid-connected household installations.
4
Access to the data via a web application A set of maps was created for the web application to demonstrate spatial heterogeneity of PV generation within the regions. The maps of irradiation, Linke turbidity and diffuse/global coefficient represent yearly averages. The map of optimal angle represents the optimal inclination of solar modules to harvest maximum solar irradiation over a year considering local shadowing. The PV maps show the yearly electricity generation. The web interface consists of two parts. The solar radiation data utility allows browsing the climatic GIS database and querying the actual values for a point clicked on the map or explicitly defined by user. The monthly and yearly values are displayed in a separate window as a table and as graphs. The PV potential estimation calculator enables to browse the solar radiation database and to calculate the power output from a PV installation defined by nominal installed power, angle of PV modules and performance ratio of the system. The integrated calculator enables also to calculate for a given location the optimal inclination of PV modules as well as their optimal east-west orientation. The calculator uses the GIS climatic database that includes also terrain shadowing effects. The web application has been written almost exclusively using server-side scripting, using PHP. While this does increase the load on the server it demands very little of the client-side computer. Thus the site is accessible also for users with less powerful hardware. The application has been written so that the actual source code is separate from the data for a given region, using a few configuration files to inform the application about the location of the data. Therefore it should be very easy to adapt the web application for use on a different region.
5
Results The maps reveal significant regional differences in available global irradiation, determined by latitude, terrain and local climatic conditions. The optimum inclination angle of PV modules to receive maximum available global irradiation in the 10 EU candidate countries (not assuming the shadowing effects) is in interval of 30-41°. If shadowing of the local terrain is also assumed (often the case in the Carpathians and the mountains in Balkan) the optimum angle of the modules inclination decreases up to 15° (see Figure 1). Figure 2 presents the yearly total of global radiation for an optimally inclined PV module. The values [kWh/m-2/year] in the region are within an interval 1140 in Estonia to 1790 in the Danube delta in Romania (this makes about 35 % less of irradiation in Estonia when compared to Romania). Comparing inclination of modules, the annual average power generation of a PV system with modules inclined at angles of 15, 25 and 40°, respectively increases about 7-12, 10-17 and 9-20%, respectively, when compared to the horizontal irradiation. The maps of potential PV power production present average values on a regional level, as this information is more understandable for decision-makers. The Figure 3 shows an average PV power generation [kWh.year-1] of a 1 kWp system installed in a residential areas, with modules inclined at an optimum angle. In the calculation, the equation cited in part 3 was used, with the performance ratio rp assumed to be 0.75. It is noticeable that although the potential for Bulgaria, Romania, Hungary, Slovenia, Slovakia and some regions in the Czech Republic are more favourable, even in the more northerly countries a small 1 kWp system can produce more of 800 kWh of electricity per year.
Dunlop, E. D., Šúri, M., Huld, T. A. (2003). Photovoltaic Potential Assessment to Support Renewable Energies Growth in 10 EU Candidate Countries. In Gottschalg, R.. (ed.) Proceedings of the Conference C79 of the Solar Energy Society CREST “Photovoltaic Science, Applications and Technology”, Loughborough University (UK) 3.-4.4.2003, pp. 1007-1016.
Figure 1 Optimum inclination angle of a PV module to receive maximum global irradiation (assuming also terrain shadowing) (°)
Figure 2 yearly total of global irradiation on an optimally inclined surface (kWh.m-2.year-1); the dots represent the ground measured data used in the calculation
Figure 3 Average annual PV power generation of a 1 kWp system installed in a residential area, with panels inclined at an optimum angle (kWh.year-1)
Figure 4 Total sum of annual PV electricity generation per region. This scenario assumes 2.5 kWp PV systems inclined at an optimum angle, homogeneously distributed in residential areas with density of 100 installations per 1 km2 (GWh.year-1).
Dunlop, E. D., Šúri, M., Huld, T. A. (2003). Photovoltaic Potential Assessment to Support Renewable Energies Growth in 10 EU Candidate Countries. In Gottschalg, R.. (ed.) Proceedings of the Conference C79 of the Solar Energy Society CREST “Photovoltaic Science, Applications and Technology”, Loughborough University (UK) 3.-4.4.2003, pp. 1007-1016.
The total potential electricity yield per year was calculated for the residential areas in individual administrative regions. This scenario assumes a theoretical assumption of having homogeneously distributed 2.5 kWp PV systems at optimum inclination within the residential area of regions with density of 100 PV installations per 1 km2. The regional differences in the potential annual PV production are given by the solar irradiation as well as a density of urbanized residential area. Therefore the most productive administrative regions are generally those with high population density and favorable climatic conditions (Figure 4). The summarization on a national level (Figure 5) gives an overview of the potential PV electricity production provided the installations are dispersed within homogeneously.
3500
3500
installed capacity [MWp]
2500 [GWh]
3000
PV power generation potential [GWh]
2500
2000
2000
1500
1500
1000
1000
500
500
0
[MWp]
3000
0 BG
CZ
EE
HU
LT
LV
PL
RO
SK
SI
Figure 5: Total potential annual PV power generation (GWh) on a country level and the potential installed capacity (MWp). A theoretic assumption of 2.5 kWp systems, homogeneously dispersed in residential areas (density 100 PV systems/km2) with modules at optimum inclination is assumed. The visualization of the regional differences by the means of maps provides a better insight into the problem and can provide a significant support for formulation of the state policies and in a regional planning. The energy yield is not the only factor taken into consideration in the state policy of support to PV installations. The prioritising of the less-favoured areas can help them to overcome social and economic limits and promote their development based on the sustainable energy production. Assuming a small 1.5 kWp configuration (approx. 12 m2), even given the low yields of 1200 kWh per year in the Baltic states, this can cover needs of a typical 4person house (in average about 3 kWh per day).
6
Conclusion The presented work is intended as a concise reference and information source to support decision-makers, researchers and industry in addressing national/regional PV development plans. The data are available also via internet and can contribute to education and promote an awareness of the general public that is still suffering from a lack of information. More information, including support pages explaining the used input data, methodology and obtained accuracy can be consulted at http://iamest.jrc.it/pvgis/pv/index.htm.
Dunlop, E. D., Šúri, M., Huld, T. A. (2003). Photovoltaic Potential Assessment to Support Renewable Energies Growth in 10 EU Candidate Countries. In Gottschalg, R.. (ed.) Proceedings of the Conference C79 of the Solar Energy Society CREST “Photovoltaic Science, Applications and Technology”, Loughborough University (UK) 3.-4.4.2003, pp. 1007-1016.
7 1.
2. 3. 4.
5.
References Hofierka J, Šúri M. The solar radiation model for Open source GIS: implementation and applications. In Ciolli M., Zatelli P. (eds), Proceeding of "Open Source GIS - GRASS Users Conference 2002", Trento, Italy, 11.-13.9.2002, www.ing.unitn.it/~grass/conferences/GRASS2002. Scharmer K, Greif J (eds). The European Solar Radiation Atlas, Volume 2: Database and Exploitation Software. Les Presses de l' École des Mines: Paris, 2000. Wald L. SODA: a project for the integration and exploitation of networked solar radiation databases. European Geophysical Society Meeting, XXV General Assembly, Nice, France, 25-29 April 2000. Šúri M, Dunlop ED, Jones AR. GIS-based inventory of the potential photovoltaic output in Central and Eastern Europe. Proceeding of Conference "Photovoltaics in Europe. From Photovoltaic Technology to Energy Solutions", Rome, Italy, 7.-11.10.2002. Heymann Y, Steenmans Ch, Croisille G, Bossard, M. CORINE Land Cover. Technical guide. Office for Official Publications of the European Communities: Luxembourg, 1994.
