A Hybrid PSO-ANFIS Approach for Horizontal Solar Radiation Prediction in Nigeria

For efficient and reliable hydrogen production via solar photovoltaic system, it is important to obtain accurate solar radiation data. Though there are equipment specifically designed for solar radiation prediction but are very expensive and have high maintenance cost that most countries like Nigeria are unable to purchase. In this study, the accuracy of a hybrid PSO-ANFIS method is examined to predict horizontal solar radiation in Nigeria. The prediction is done based on the available meteorological data obtained from NIMET Nigeria. The meteorological data used for this study are monthly mean minimum temperature, maximum temperature, relative humidity and sunshine hours, which serves as inputs to the developed model. The model accuracy is evaluated using two statistical indicators Root Mean Square Error (RMSE) and Coefficient of determination (R²). The accuracy of the proposed model is validated using ANFIS, GA-ANFIS models and other literatures. Based on the statistical parameters used for the model evaluation, the results obtained proves PSO-ANFIS as a good model for predicting solar radiation with the values of RMSE=0.68318, R²=0.9065 at the training stage and RMSE=1.3838, R²=0.8058 at the testing stage. This proves the potentiality of PSO-ANFIS technique for accurate solar radiation prediction