Renewable Energy Systems in Remote Sites
A geographical-information system (GIS) is a computerized data base which allows one to integrate and to process information coming from different sources, to elaborate strategies for the development of developing countries. GIS was developed in the late 1960s but has been installed in very few places because, the hardware was very expensive and the amount of software available on the market was limited. However, with the decrease in the hardware prices and the increase in computer performance, the GIS has become a more useful tool for political or economic policy.
As a tool box, a GIS allows planners to perform spatial analysis by using its geoprocessing or cartographic modeling functions, such as map overlay, selection and selection SQL (structured query language) and thematic analysis. Among all the geoprocessing functions, the map overlay is probably the most useful where the different layers could represent the relief of the area, the electric network, the remote sites etc.
During the last 20 years, GIS applications have increased in several domains: scientific, economic and political. Moreover, this range is expanding due to the capabilities offered by this method for regional and urban partners. GIS has shown its usefulness in several situations, for example in rural and urban planning, data importation for regional development, and specific site-selection.is the originator of GIS utilization for the integration of renewable-energy systems with an early study of the wind-energy potential. Two systems (extension of the electric network and wind generators) were compared on the basis of an entropy minimization. More recently, have developed a cartography of grid-connected systems potentialities in the USA. The coupling of these results with solar data using a GIS will permit the evaluation of the real benefits for the American energy market. Other recent works apply GIS methodologies to the study of water desalination and renewable-energy systems.
The purpose of this paper is to apply the GIS methodology to renewable-energy systems providing electrical energy in stand-alone conditions. First, the methodology of the project is explained, then a regional base is built containing information, characterizing the area studied such as the solar potential, the existing electrical networks (medium and high voltage grids), and the remote site locations. From this regional base, computer tools are developed to size the energy systems in relation to the electricity needs of the site. The various decentralized electricity systems are compared on the basis of a minimization of the energy cost. Finally, the main results of the study are described and an integration plan for renewable-energy systems is presented.
The area studied is delimited by the south Corsica department shown in Fig. 2. The area of the square is 8100 km2 and the terrestrial surface is about 3800 km2. With about 250,000 inhabitants, the island of Corsica has a mean population density of 27.7 inhabitants/km2. But in rural areas, this density falls to 7?10 inhabitants/km2. Corsican culture is mainly pastoral and many sheepfolds are scattered over the country representing an important potential of remote sites because these locations are without electrical supplies.
The flowchart shown in Fig. 3 illustrates the complete methodology of our study. To elaborate this GIS, the first step consists of a digitalization of the data maps (RASTER mode) to create the regional data base, which must be as complete and accurate as possible. Each information layer represents a particular characteristic data set, such as solar potential, electrical network, topographic data and radar records, satellite and meteorological data and land availability. The data are spatially referenced. They are processed by several computer tools to design the electricity-supply systems: determination of the component sizes for the conventional and photovoltaic systems and estimation of the distance from a remote site to the nearest transformer for grid extension. These various systems are compared in terms of energy production cost using an economics tool integrated into the GIS.
These data represent the natural constraints of the studied area (lakes, ponds, military areas, parks …) which are not digitized by the GIS; so the pixels corresponding to these areas are not computed by the software.
The French Utility, EDF (Electricité de France), supplied us with paper maps representing both the high (scale 1/200000) and medium (scale 1/50000) voltage grids in the studied area with a resolution of 500 m. These maps, updated in December 1995, and visualized on IGN (Institut Géographique National) background maps, also show the locations of the medium and low-voltage transformers. The medium-voltage grid is shown in Fig. 4.
Topographic maps IGN TOP25 (scale 1/25000) indicating the land suitability were used. These maps, updated in 1988, have an accuracy estimated at 50 m and are used in the LAMBERT ZONE II ETENDU system representing the projections and coordinate system used by the IGN Institute to characterize the kilometric grid.
By correlating the previous layers, we have computed the locations of remote sites by performing the following sequence:
• combination of “electric network” and maps of suitability;
• construction of N layers (1, …, i, …N) considered as buffer areas computed from the “electric network” layer. The ith layer represents the locations of sites i×0.25 km away from the electrical grid. As an example, in Fig. 5 the layers i=2–8 are drawn for one sector of the existing medium-voltage electric network;
• determination of remote-site locations, when the distance between the site and the grid exceeds 500 m (conventional maximum limit for electrical transmission at low voltages).
This algorithm has allowed us to determine the locations of 1046 individual remote sites representing 496 grouped sites (around a reference house in a circle of 500 m). The distribution of the distances of these sites from the grid is presented in Fig. 6. We note that 80% of the sites are located at a distance less than 2 km from the grid, which indicates that the majority of customers are near to the electrical network.
In a previous study, we have shown the inadequacy of the pyranometric Corsican network for calculation, by interpolation methods, of the total solar irradiation all over the island. Indeed, taking into account the bumpy relief, there is no relation from a solar radiation point of view, between two sites which are about 10 km apart. A model allowing the computation of daily global irradiation on a horizontal plane, based on the utilization of METEOSAT satellite-derived data, was used to map the solar energy potential. The spatial resolution of satellite WEFAX (Weather FACSimile) images defines a pixel with an area of about 25 km2. The validation of this model has shown that the relative root-mean-square error between the experimental and simulated daily horizontal global-irradiation is less than 13%. The annual mean value of daily global-irradiation varies over the range of 4.04 to 4.83 kWhm?2 corresponding to a variation of 20% over the island.
The various solutions envisaged to electrify remote sites are:
• photovoltaic systems, characterized by their PV modules area and battery storage capacity;
• hybrid systems consisting of a PV system with an auxiliary engine generator; its sizing parameters are PV modules area, battery-storage capacity and nominal power of the engine generator;
• engine generator directly connected to the load via an AC/DC converter, characterized by its nominal power; and
• an extension of the electrical grid, for which the lengths of the medium and low voltage lines must be known.
The sizing of these systems required meteorological data as input and provided electrical consumption data as output.
Tags: renewable energy
