Scholarlink

<p>The growth of industrial and commercial fuel cell applications as a clean energy source is one of the focal points for energy sector researchers, leading to a constant search for cost-effective and accurate modeling techniques. This study meets this need by proposing a systematic method of determining solid oxide fuel cell (SOFC) stack models by appropriately choosing unknown parameters, where the main objective is to minimize the sum of squared errors between model output voltage and experimental data. The proposed walrus optimization (WO) algorithm is used to obtain an improved efficiency and better results. The reliability of the system is thoroughly tested in two cases, in which the temperature and pressure are varied (from 1073 to 1273&#xa0;K and from 1 to 9&#xa0;atm). A detailed comparative analysis with nine other metaheuristic algorithms proves the superior effectiveness of the proposed technique. The results show that the WO algorithm has significantly better performance and reaches the lowest mean squared error (MSE) values of 2.57E−07. Statistical analysis of 100 independent runs further confirms its outstanding stability as indicated by the smallest standard deviations (0.191 at 1073&#xa0;K) and the shortest computational times, in the range of 0.20 to 0.41&#xa0;s. The robustness of the algorithm is clearly proven by its top rank in the Friedman ranking test with a score between 1.125 and 1.5625 in all the test cases. These statistical results, along with the convergence and boxplot analyses, clearly demonstrate the excellent efficiency, precision, and reliability of the proposed WO-based technique for SOFC parameter identification.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Walrus optimization algorithm for enhanced solid oxide fuel cell (SOFC) model parameter identification

  • 期刊论文

  • 发表日期：

    2025年10月27日

查看全文
  • Manish Kumar Singla,
  • Manpreet Singh,
  • Ramesh Kumar,
  • Pradeep Jangir,
  • Arpita,
  • Reena Jangid

摘要

The growth of industrial and commercial fuel cell applications as a clean energy source is one of the focal points for energy sector researchers, leading to a constant search for cost-effective and accurate modeling techniques. This study meets this need by proposing a systematic method of determining solid oxide fuel cell (SOFC) stack models by appropriately choosing unknown parameters, where the main objective is to minimize the sum of squared errors between model output voltage and experimental data. The proposed walrus optimization (WO) algorithm is used to obtain an improved efficiency and better results. The reliability of the system is thoroughly tested in two cases, in which the temperature and pressure are varied (from 1073 to 1273 K and from 1 to 9 atm). A detailed comparative analysis with nine other metaheuristic algorithms proves the superior effectiveness of the proposed technique. The results show that the WO algorithm has significantly better performance and reaches the lowest mean squared error (MSE) values of 2.57E−07. Statistical analysis of 100 independent runs further confirms its outstanding stability as indicated by the smallest standard deviations (0.191 at 1073 K) and the shortest computational times, in the range of 0.20 to 0.41 s. The robustness of the algorithm is clearly proven by its top rank in the Friedman ranking test with a score between 1.125 and 1.5625 in all the test cases. These statistical results, along with the convergence and boxplot analyses, clearly demonstrate the excellent efficiency, precision, and reliability of the proposed WO-based technique for SOFC parameter identification.