Scholarlink

<p>Accurate estimation of fuel cell parameters is critical for enhancing system performance, control precision, and reliability under varying thermal and pressure conditions. However, the nonlinear, multivariable dynamics of fuel cells present significant challenges for conventional optimization methods, which often suffer from premature convergence and limited generalization. To address this, a novel metaheuristic optimization framework is proposed that builds upon the Rime-Ice algorithm by integrating three key enhancements: a chaos-based exploration mechanism, adaptive weighting for balancing global and local search, and Gaussian mutation for refining solution diversity and accuracy. The proposed method was validated across ten test cases involving five distinct operating temperatures (1073 to 1273&#xa0;<i>K</i>) and five pressure levels (1 to 9&#xa0;<i>atm</i>). Comparative analysis was conducted against nine state-of-the-art optimization techniques. The proposed optimizer consistently outperformed all competitors, achieving a minimum mean squared error as low as 4.96 × 10<sup>−5</sup>, an average improvement of 99.93% over the worst-performing algorithm, and a top Friedman ranking score of 1.00 across all scenarios. The error curves remained within ± 0.005&#xa0;V for all cases, confirming excellent predictive fidelity. These results demonstrate that the proposed strategy delivers robust, accurate, and efficient parameter identification for fuel cell models under diverse conditions. This capability has direct implications for real-time modeling, adaptive control, and digital twin deployment in electrochemical energy systems. Future work will explore hardware implementation and extension to hybrid energy storage systems.</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

An enhanced RIME based metaheuristic with chaos and Gaussian mutation for accurate solid oxide fuel cell parameter identification

  • 期刊论文

  • 发表日期：

    2025年8月12日

查看全文
  • Toshika R. Agrawal,
  • Mohammad Aljaidi,
  • Sanjeev Maheshwari,
  • Gadug Sudhamsu,
  • Bh V. S. Narayana,
  • Monika Singla,
  • Sunilkumar P. Agrawal,
  • Arpita,
  • Pradeep Jangir,
  • Reena Jangid

摘要

Accurate estimation of fuel cell parameters is critical for enhancing system performance, control precision, and reliability under varying thermal and pressure conditions. However, the nonlinear, multivariable dynamics of fuel cells present significant challenges for conventional optimization methods, which often suffer from premature convergence and limited generalization. To address this, a novel metaheuristic optimization framework is proposed that builds upon the Rime-Ice algorithm by integrating three key enhancements: a chaos-based exploration mechanism, adaptive weighting for balancing global and local search, and Gaussian mutation for refining solution diversity and accuracy. The proposed method was validated across ten test cases involving five distinct operating temperatures (1073 to 1273 K) and five pressure levels (1 to 9 atm). Comparative analysis was conducted against nine state-of-the-art optimization techniques. The proposed optimizer consistently outperformed all competitors, achieving a minimum mean squared error as low as 4.96 × 10−5, an average improvement of 99.93% over the worst-performing algorithm, and a top Friedman ranking score of 1.00 across all scenarios. The error curves remained within ± 0.005 V for all cases, confirming excellent predictive fidelity. These results demonstrate that the proposed strategy delivers robust, accurate, and efficient parameter identification for fuel cell models under diverse conditions. This capability has direct implications for real-time modeling, adaptive control, and digital twin deployment in electrochemical energy systems. Future work will explore hardware implementation and extension to hybrid energy storage systems.