Prediction and estimation of solar radiation using artificial neural network (ANN) and fuzzy system: a comprehensive review

The immense capability of artificial intelligence is to improve solar and wind speed prediction capability using various algorithms and models. The author has reviewed many papers on both ANN and fuzzy logic for the estimation of solar radiation. The energy sector heavily relies on the prediction and optimization for energy production. One of the biggest problems in renewable energy market uptake is uncertainty, and for that, AI/ML plays a crucial role. Artificial neural network is used for predicting the result from given input parameters called meteorological parameters such as sunshine duration, relative humidity, temperature, atmospheric pressure, and so on. It provides the output from the computation part. In contrast, fuzzy logic also used for estimation but with different algorithms and models compared to ANN. The proposed models with varying combinations of input are tested with the help of statistical indicators like mean absolute percentage error (MAPE), root means square error (RMSE), mean bias error (MBE), and coefficient of determination (R2). We can say that ANN gave better accuracy and a low error rate compared to fuzzy logic.