A Novel Intelligent Technique Based on Metaheuristic Algorithms and Artificial Neural Networks: Application on a Photovoltaic Panel
Noamane Ncir, Saliha Sebbane, Nabil El Akchioui
Abstract
This article presents a novel methodology of optimization based on metaheuristic algorithms for optimization including Particle Swarm Optimization (PSO), Grey Wolf Optimization (GWO), Whale Optimization Algorithm (WOA), and Imperialist Competitive Algorithm (ICA), and Artificial Neural Networks. The optimization method is mainly based on reducing the percentage of error before training the created neural network. For that, after the collection of dataset of the chosen system, those algorithms identifies the best configuration of weights and bias to train the Artificial Neural Network (ANN). However, metaheuristic methods work in a different way than classical methods, i.e. the mathematical modeling of these algorithms takes into consideration stochastic parameters and decisions. In this paper, all algorithms are validated by simulation using MATLAB software.