Blackbox-based model identification of solid oxide fuel cells by hybrid Ridgelet neural network and Enhanced Fish Migration Optimizer. (November 2022)

• Record Type: Journal Article
• Authors: Yang, Guihua; Ma, Junchi; Deng, Yuwei; Sun, Shujia; Fu, Baohong; Fathi, Gholamreza
• Is Part Of: Energy reports. Volume 8(2022)
• Page Start: 14820
• Page End: 14829
• Publication Date: 2022-11
• Subjects: SOFC -- Model estimation -- Ridgelet neural network -- Enhanced Fish Migration Optimization Algorithm; Power resources -- Periodicals; Energy industries -- Periodicals; Power resources; Periodicals; Electronic journals; 621.04205
• DOI: 10.1016/j.egyr.2022.11.020
• Languages: English
• ISSNs: 2352-4847

Abstract

Abstract: - An effective energy conversion tool is solid oxide fuel cell that converts chemical energy to heat and electricity by oxidant reactions and electrochemical fuel. Efficient modeling of this device can decrease the analyzing and design costs a lot. Thus, a new blackbox-based identification system is designed and defined for the solid oxide fuel cells herein. The presented method is a modified Ridgelet neural network. The method has been modified by a new enhanced design of the Fish Migration Optimizer. The new designed model, Enhanced Fish Migration Optimizer-Rigdelet neural network, is then performed on some data to train and validate. The model is finally put in comparison with several latest models to authenticate the model. Final results show that the proposed method with 1.78 MSE for 2/12/1 network topology provides the best accuracy to the others. The model is also confirmed by experimental data and the final achievements indicate that the suggested technique gives satisfying results for model identification of the solid oxide fuel cells.