A multi-model evaluation of Enhanced Tunicate Swarm Optimization for parameter identification

Parameter identification for a proton exchange membrane fuel cell (PEMFC) entails employing optimisation techniques to discover the best unknown parameter values required to generate an accurate fuel cell performance prediction model. This technique, known as parameter identification, is important since manufacturers' datasheets do not usually disclose these values. To address this, the manuscript examines five optimisation strategies, including the suggested algorithm, Enhanced Tunicate Swarm Optimizer (ETSO), for predicting these parameters in PEMFCs. Each technique uses the six unknown parameters as decision variables, aiming to reduce the sum squared error (SSE) between anticipated and observed cell voltages. The data reveal that the suggested strategy outperforms existing approaches and cutting-edge optimizers. The two models are used to assess the dependability and performance of the PEMFC. The results are also compared to the non-parametric tests, and it is found that the suggested method outperforms the other algorithms in both suggested models. • Enhancing the Tunicate Swarm Optimizer (ETSO) to better find unknown settings (parameters) within PEMFC fuel cell stacks used in real-life situations. • Validating the effectiveness of the ETSO on real-world PEMFC units by applying it to two specific models. • To demonstrate the superiority of the ETSO, the proposed algorithm is compared with well know meta-heuristic algorithms. • Ten Benchmark test function is also tested to verify the algorithm.