Intelligent Solar Irradiance Forecasting Using Hybrid Deep Learning Model: A Meta-Heuristic-Based Prediction

Solar PhotoVoltaic is one among the majority of key techniques for moving away from fossil fuels and toward renewable energy. Solar prediction is an efficient approach for improving the process of an electrical method for combining a huge number of solar power production systems, and it seeks to extend a novel empirical approach to represent the solar irradiance prediction uncertainty. The major highlight of this work is to design and develop a new solar irradiance forecasting using a hybridized deep structured architecture method. Initially, the benchmark data is gathered that incorporates the “numerical weather forecasting data” like “temperature, dew point, humidity, visibility, wind speed, and other descriptive information”. After collecting the data, the feature extraction is performed by ICA, PCA, and LDA. Moreover, the optimal feature selection is performed for selecting the important features. The features are used for the solar irradiance forecasting using the hybridized deep structured arhitectures. The suggested deep learning model consists of Deep Neural Network (DNN) and Recurrent Neural Network (RNN). The training algorithm of both DNN and RNN is modified by the Slow Rider-based Rider Optimization Algorithm (S-ROA) so known as Weight Optimized Deep Recurrent Neural Network. The suggested hybrid deep learning could be extremely successful for solar irradiance prediction, outperforming conventional learning algorithms, according to the comparison study.