Abstract:The precise model for Polymer Electrolyte Membrane Fuel Cells (PEMFCs) is vital for simulation, control, and performance analysis of PEMFCs. It is crucial to accurately estimate the model parameters. Over the past decade, the extraction of unknown parameters of PEMFCs model is formulated as optimization problem and several meta-heuristic techniques have emerged to solve this problem. Despite the development of these techniques to tackle the problem, the slow convergence rate and susceptibility to being trapped in local minima are regarded as weaknesses of these methods. Additionally, because the PEMFC model is a nonlinear and complex model, not all optimization algorithms are suitable for solving it. This article introduces a novel approach that utilizes the Autonomous Groups Particle Swarm Optimization (AGPSO) algorithm for extracting precise values for uncertain parameters inherent in PEMFC model of 250Wstack and BCS-500W stack utilized significantly in the literature. The optimization problem's fitness function is formulated as the total squared errors (TSEs) between the voltage datasets measured and estimated. Three different versions of AGPSO algorithm are presented. Statistical analysis is performed on these three versions to assess their robustness, and the most robust version is identified. Furthermore, the performance of the most robust version is extensively tested and analyzed through a complete comparison with recent optimization algorithms findings from the updated state-of-the-art literature such as BO, QOBO, TGA, HHO and ASO. Further, a statistical analysis was carried out that ensured the reliability and robustness of the introduced AGPSO. The results highlight the efficacy and feasibility of AGPSO-based approach across all compared algorithms, demonstrating an enhancement in the accuracy of the PEMFC model. A two-phase sensitivity analysis, conducted through Monte Carlo Simulation (MCS), was employed to assess how variations in optimized parameters affect the objective function (TSEs). Furthermore, the uncertainty in the maximum power reference current is identified.

Parameter identification of PEMFC via feedforward neural network-pelican optimization algorithm

A novel P systems based optimization algorithm for parameter estimation of proton exchange membrane fuel cell model

Parameter identification of PEMFC model based on hybrid adaptive differential evolution algorithm

Cuckoo Search Algorithm with Onlooker Bee Search for Modeling PEMFCs Using T2FNN

Generalized Regression Neural Network Based Meta-Heuristic Algorithms for Parameter Identification of Proton Exchange Membrane Fuel Cell

Parameter identification of proton exchange membrane fuel cell via Levenberg-Marquardt backpropagation algorithm

Parameter identification of PEMFC steady-state model based on p-dimensional extremum seeking via simplex tuning optimization method

Parameter extraction of PEMFC via Bayesian regularization neural network based meta-heuristic algorithms

Optimal parameters identification for PEMFC using autonomous groups particle swarm optimization algorithm

Parameter Identification of Solid Oxide Fuel Cell Using Elman Neural Network and Dynamic Fitness Distance Balance-Manta Ray Foraging Optimization Algorithm

Attack-defense strategy assisted osprey optimization algorithm for PEMFC parameters identification

Nonlinear Modeling of PEMFC Based on Neural Networks Identification

Nonlinear simulating of the Proton Exchange Membrane Fuel Cells utilizing Ridgelet Neural Network optimized using a hybrid form of Northern Goshawk Optimizer

Parameter Identification of a Quasi-3D PEM Fuel Cell Model by Numerical Optimization

Optimized Parameter Estimation Of A Pemfc Model Based On Improved Grass Fibrous Root Optimization Algorithm

Bald Eagle Search Algorithm for Parameter Identification of Proton Exchange Membrane Fuel Cell

Parameter Identification for PEM Fuel-Cell Mechanism Model Based on Effective Informed Adaptive Particle Swarm Optimization

Modeling of A Proton Exchange Membrane Fuel Cell Based on the Hybrid Particle Swarm Optimization with Levenberg-Marquardt Neural Network

Improved grey wolf optimizer based on neighborhood trust model for parameter identification of PEMFC

Model parameter estimation of the PEMFCs using improved Barnacles Mating Optimization algorithm

Bi-subgroup optimization algorithm for parameter estimation of a PEMFC model