A general radial basis function neural network assisted hybrid modeling method for photovoltaic cell operating temperature prediction

Abstract

Accurate and reliable prediction of photovoltaic (PV) cell operating temperature is vital for performing accurate output power prediction. Although numerous mathematical models have been developed to capture the effect of environmental variables on PV cell temperature, the prediction accuracy needs to be further improved and a relatively general modeling framework needs to be developed to enhance adaptability for different PV cell types. In this study, a novel radial basis function (RBF) neural network assisted hybrid modeling strategy is proposed to predict the PV cell temperature. The hybrid model is designed as an explicit mathematical formulation combined with an RBF neural network assisted correcting factor. The known function formulation is induced from prior knowledge and the unknown correcting factor is modeled by the RBF neural network. The hybrid cell temperature model is combined with the equivalent circuit model and the effectiveness of cell temperature and output power prediction is evaluated. The results illustrate that the proposed method can perform accurately cell temperature and output power prediction for both laboratory and commercial plants. It is thus indicated that the proposed hybrid modeling strategy could provide a potential general solution framework of cell temperature and output power prediction for different PV cells.