Yang, Bo, Liang, Boxiao, Zhou, Shuai, Qian, Yucun, Zheng, Ruyi, Shu, Hongchun, He, Peng, Wang, Jingbo 
ORCID: 0000-0002-6316-2678, Jiang, Lin ORCID: 0000-0001-6531-2791, Sang, Yiyan ORCID: 0009-0005-7990-5615 et al (show 1 more authors)
(2026)
A novel bio-inspired caterpillar fungus (Ophiocordyceps sinensis) optimizer for SOFC parameter identification via GRNN.
Renewable Energy, 256.
p. 123995.
ISSN 0960-1481, 1879-0682
|
Text
manuscript_clear.pdf - Author Accepted Manuscript Available under License Creative Commons Attribution. Download (8MB) | Preview |
Abstract
Accurate parameter identification is crucial for the optimal control and performance assessment of solid oxide fuel cells (SOFCs) due to the high non-linearity in its modeling. To solve this, this study develops a novel caterpillar fungus optimizer (CFO) for SOFC parameter identification, coupled with generalized regression neural network (GRNN) for data preprocessing. The proposed CFO is characterized by powerful searching capabilities and strategic operators designed to overcome the challenges of local optimums. For a comprehensive validation, twenty-three standard benchmark functions are applied for analysis, demonstrating the effectiveness of CFO in finding the optimal solution and proficiency in escaping local optimums. Regarding the implementation for SOFC parameter identification, initially, GRNN is employed to filter out noise from the experimental data. The refined data are then transferred to CFO alongside four other competitive algorithms to identify unknown SOFC parameters. In this work, two widely studied SOFC models, i.e., electrochemical model (ECM) and simple electrochemical model (SECM) are adopted for validation under MATLAB and SimuNPS. The simulation results demonstrate that CFO, after data preprocessing, can identify the optimal parameters with robustness, speed, and accuracy. For instance, it achieves a maximum improvement in identification accuracy of 94.41 % and 94.10 % for ECM and SECM, respectively.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | 40 Engineering |
| Divisions: | Faculty of Science and Engineering Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 11 Aug 2025 07:34 |
| Last Modified: | 11 Aug 2025 07:34 |
| DOI: | 10.1016/j.renene.2025.123995 |
| Related Websites: | |
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3194003 |
Altmetric
Altmetric