Futuristic deep learning algorithms for long-term solar power prediction

Abstract

Abstract Solar power(SP) prediction using a futuristic deep-learning(DL) algorithm is an important piece of research. The SP data is accumulated in real-time with a 1 kW capacity that is available in our laboratory. The collected data is pre-processed via, initialization, normalization, and validation for accurate prediction. Data sets are normalized in order to fill in missing values using the K-nearest neighbor(KNN) algorithm and the interpolation method. Then, the data is validated using a newly proposed deep long short-term memory(DLSTM) with a recurrent neural network(RNN) algorithm for solar power prediction(SPP). To express the superiority of the DLSTM-RNN model, it is contrasted to other exciting models. All the DL algorithms are trained and tested using three different activation functions: Sigmoid, ReLU, and tanh with varying values of an epoch. Finally, the precision is evaluated in terms of different performance error indexes, such as the basic error index(BEI) and promoting percentage error index(PPEI).