A differential evolution proposal for estimating the maximum power delivered by CPV modules under real outdoor conditions

• Published in: Expert systems with applications Vol. 42; no. 13; pp. 5452 - 5462
• Main Authors: García-Domingo, B., Carmona, C.J., Rivera-Rivas, A.J., del Jesus, M.J., Aguilera, J.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2015
• Subjects: Atmospheric models; Concentrating photovoltaic technology; Differential evolution; Evolution; Mathematical models; Maximum power; Modules; Photovoltaic cells; Soft computing; Solar cells; Spectra; Differential evolution; Concentrating photovoltaic technology; Soft computing; Maximum power
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2015.02.032

•Multivariable model to calculate the maximum power given by concentrating modules.•Spectrum additional indexes in the equation proposed by the standard methodology.•Calculation of regression coefficients by a differential evolution algorithm.•Applicability to the forecasting of the energy produced by big power plants. Concentrating photovoltaics is an innovative alternative to flat-plate module to produce cost-competitiveness electricity. It is based on the use of optical system of reduced cost which is able to concentrate the solar light on a very small surface (high efficiency solar cell). At present, this technology has a marginal position in photovoltaic market and to take off needs to increase the confidence of the public and private sector. A better understanding of the concentrating photovoltaics technology electrical performance under real meteorological conditions would improve this situation. Because the bankability of a concentrating photovoltaics plant is addressed through the modelling of its energy production, an accurate estimation of the maximum power of the these modules is crucial to achieve it. Accordingly, the commercial evolution of concentrating photovoltaic technology demands prediction models for estimating the maximum power delivered by a concentrating photovoltaic module under real atmospheric conditions. Until now the only established standard method for outdoor power rating of this type of modules (ASTME-2527-09, defined by the American Society for Testing and Materials) does not consider the impact of the direct normal irradiance spectral distribution. The solar spectrum has an important influence on the electric performance of multijunction solar cells which composes concentrating photovoltaic modules. In this work, an analysis of the inclusion in the prediction model of the solar spectrum by means of two indexes (spectral matching ratio and the average photon energy) and different spectral intervals is performed. Then, a differential evolution proposal for the estimation of regression coefficients for the two multivariable regression models is described. The accurate calculation of the model parameters reveals relations among the atmospheric conditions very useful for the experts. The multivariable regression models have been applied to two different concentrating photovoltaic modules, obtaining mean absolute percentage error values within the range 1.91–3.94%. The use of these accurate models for the estimation of the maximum power would allow to estimate the electric production of a concentrating photovoltaic power plant and the analysis of its costs and profitability, with the consequent benefits for the commercial development of this technology.