Prediction of daily diffuse solar radiation using artificial neural networks

This study presents two optimization techniques, genetic algorithm (GA) and particle swarm optimization (PSO), to improve the efficiency and generalization ability of back propagation neural network (BPNN) model for predicting daily diffuse solar radiation. Seven parameters including month of the year, sunshine duration, mean temperature, rainfall, wind speed, relative humidity, and daily global solar radiation are selected as the evaluating indices. The predictions from the BPNN optimized by PSO model were compared with those from two models: BPNN and BPNN optimized by GA. The results show that the proposed BPNN optimized by PSO model has potential in accurately predicting the daily diffuse solar radiation.