article · Energy Reports
Proton exchange membrane fuel cells serve as clean energy generators, converting chemical reactions into electricity. Designing and understanding these systems relies on accurate operational models that closely reflect physical polarisation curves. To resolve unknown model parameters, a comprehensive collection of modern metaheuristic optimisation algorithms was evaluated, including variants of Differential Evolution, the Whale Optimization Algorithm, and the Gray Wolf Optimizer. The algorithms were tested against three practical fuel cell stacks: a BCS 500-W, a 500-W SR-12PEM, and a 250-W unit, across diverse operating environments. Parameter precision was quantified using the sum of squared errors against experimental data. The resulting polarisation curves aligned closely with manufacturer data, showing stronger accuracy and convergence characteristics than comparative approaches reported in earlier literature.
Proton exchange membrane fuel cells are important components of clean energy transitions. To design and manage them effectively, engineers require simulation tools that mirror physical performance. Using robust optimisation algorithms to determine operational parameters allows developers to predict fuel cell behaviour accurately under varied conditions, reducing the need for costly physical trials and improving system control.
This work represents an applied software methodology validated against existing commercial fuel cell hardware, specifically 250 W and 500 W stacks. It could directly assist fuel cell manufacturers and power system developers seeking to improve simulation software, cell monitoring, and diagnostic tools. Operating at an applied testing stage, the mathematical models can be incorporated into engineering simulation suites to refine product design and accelerate performance tuning.
AI-generated from the published abstract. Always read the original work before citing.
The electro-chemical proton exchange membrane fuel cell (PEMFC) is an inventing electrical generator from chemical reaction process as a green energy source. An accurate PEMFC model with its precise parameters should be used to carefully fitting of polarization curve to best study and design of its characteristics and performance. This paper introduces an accurate PEMFC model based on recent metaheuristics algorithms to evaluate precisely the unknown parameters of PEMFC. Algorithms of; Whale Optimization Algorithm (WOA) , Weighted Differential Evolution Algorithm (WDE) , Differential evolution algorithm with strategy adaptation (SADE) , Moth- Flame Optimization Algorithm (MFO) , adaptive differential evolution with optional external archive (JADE), Improved mine blast algorithm (IMBA), Gray Wolf Optimizer (GWO), Dragonfly algorithm (DA), Differential EVOLUTION ALGORITHM (DE) , Cumulative Population Distribution Information in Differential Evolution (CPIJDE) , Differential evolution based on covariance matrix learning (COBIDE) , Covariance Matrix Adaptation Evolution Strategy (CMA-ES) , Bernstain-search differential evolution algorithm (BSD), Backtracking Search Optimization Algorithm (BSA) , Bezier Search Differential Evolution Algorithm (BESD ) , DIFFERENTIAL SEARCH ALGORITHM (DSA) and Bijective DSA (B-DSA) , Biogeography-based optimization (BBO); have been applied to estimate model of PEMFC. The verification of the suggested optimizing algorithms is applied on three practical PEMFC stacks of BCS 500-W PEM, 500 W SR-12PEM and 250 W stacks, for different operating conditions. The accuracies of the PEMFC extracted parameters are measured in sum of square errors (SSE) between the results obtained by the optimizing parameters and the test results of the fuel cell stacks in the objective function. Also, the applied methods have been validated as compared results with different research works that were listed in literatures. Moreover, the polarization curves of the applied methods are clear and coinciding with manufacturing polarization curves for all the case study results. So, the suggested PEMFC optimizing model has superiority on the comparative models with respect to the system accuracy and convergence process.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1016/j.egyr.2021.09.145
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
