Optimized hierarchical radial basis function neural networks by developing coronavirus herd immunity optimizer for solid oxide fuel cells

Abstract

SummaryA new blackbox technique has been presented in this article for model estimation of solid oxide fuel cells (SOFCs) for providing better results. The proposed method is based on a hierarchical radial basis function (HRBF). The presented method is then developed by a new modified metaheuristic called developed coronavirus herd immunity algorithm (DCHIA). The suggested model has been named DCHIA‐HRBF. The proposed model is then trained by some data and prepared for identification and prediction. The model is then analyzed and put in comparison with several latest techniques for validation of the efficiency of the technique. It is also verified by the empirical data to prove its validation with the real data. The results show that the best cost for the performance index which is the network error, is achieved by the proposed developed coronavirus herd immunity algorithm with about 119.442, which is satisfying for the considered function and target against the other state‐of‐the‐art methods. As a result, the simulation results specified that the suggested DCHIA‐HRBF delivers high effectiveness as an identifier and prediction tool for the SOFCs.