Precise modeling of PEM fuel cell using a novel Enhanced Transient Search Optimization algorithm

• Published in: Energy (Oxford) 2022-05, Vol.247, p.123530, Article 123530
• Main Authors: Hasanien, Hany M., Shaheen, Mohamed A.M., Turky, Rania A., Qais, Mohammed H., Alghuwainem, Saad, Kamel, Salah, Tostado-Véliz, Marcos, Jurado, Francisco
• Format: Article
• Language: English
• Subjects: Algorithms; Distribution functions; electric potential difference; Energy system modeling; Error analysis; Fuel cells; Fuel technology; Genetic algorithms; Hydrogen energy; Neural networks; nonlinear models; Optimization; Optimization algorithms; Optimization methods; Parameter extraction; Proton exchange membrane fuel cells; Searching; Sensitivity analysis; Statistical analysis; t-test; Weibull distribution; Weibull statistics
• ISSN: 0360-5442; 1873-6785
• DOI: 10.1016/j.energy.2022.123530

This paper introduces a novel enhancement to the Transient Search Optimization (TSO) algorithm to estimate an accurate electrical model of the proton exchange membrane fuel cell (PEMFC). The PEMEFC model is a non-linear model that includes seven unknown variables which cannot be calculated analytically. The TSO is enhanced by inserting two new factors, the Levy function and the Weibull distribution function. The proposed enhanced Transient Search Optimization (ETSO) and TSO algorithms are applied to estimate the seven variables by minimizing the sum of the squared errors (SSEs) between the measured and calculated voltages. The error is defined as the difference between the measured and the calculated output voltage of the PEMFC. Three different commercial types of PEMFCs are modeled: i) Ballard, Mark V 5 kW, ii) Horizon H-12, and iii) 6 kW Nedstack PS6 stacks PEMFC. The estimated seven variables and the minimum SSE of electrical PEMFCs using ETSO and TSO algorithms are compared with the results obtained by using other optimization algorithms like whale optimization algorithm, genetic algorithm, neural network algorithm and others. The results obtained by the ETSO are better than that obtained by TSO by more than 10% and this percentage increases with other algorithms. The accuracy of the proposed PEMFC model is verified by comparing the estimated V–I and P–I characteristics with the measured data. The effectiveness of the proposed ETSO based model is verified by an investigation of sensitivity analysis for design variables and the robustness of the ETSO algorithm via the statistical analysis and the parametric t-test. •This paper presents a novel improved TSO algorithm to extract PEMFC model parameters.•Three commercial PEMFC models are implemented in this paper.•The parameters of proposed model are compared with other recent models.•The proposed model is verified by comparing its results with the measured results.•The sensitivity and robustness analyses of the proposed method are tested.