期刊文献+

基于径向基神经网络的质子交换膜燃料电池智能参数辨识 被引量:1

Intelligent Parameter Identification of Proton Exchange Membrane Fuel Cells Based on Radial Basis Function Neural Network
原文传递
导出
摘要 在质子交换膜燃料电池(Proton Exchange Membrane Fuel Cell, PEMFC)中,准确地辨识未知参数对于建立可靠而精确的模型至关重要.然而,PEMFC参数辨识难以被常规的数值分析方法解决,这是一个涉及多个变量且有强耦合的非线性问题.此外,噪声对数据的影响、收集数据的不足以及电池记录数据的丢失都会增加获取精确参数的难度.针对以上问题,本文提出一种基于径向基函数(Radial Basis Function, RBF)神经网络联合启发式算法的参数识别策略.先对RBF进行训练,并利用RBF对数据进行降噪与预测处理,以解决噪声对数据的影响、收集数据的不足以及电池数据丢失的情况;再利用启发式算法对PEMFC模型参数进行辨识.结果表明,经过RBF处理后可以显著降低异常情况对参数辨识的影响,极大程度提高启发式算法参数辨识的准确性,其中V-I拟合精度达到99.56%. In Proton Exchange Membrane Fuel Cells(PEMFC),accurately identifying unknown parameters is crucial for establishing reliable and precise models.However,the identification of PEMFC parameters is a challenge that cannot be readily solved by conventional numerical analysis methods,as it involves a multi-variable and strongly coupled nonlinear problem.Moreover,the impact of noise on data,insufficient data collection,and the loss of battery record data all increase the difficulty of obtaining accurate parameters.To address these issues,this paper proposes a parameter identification strategy based on Radial Basis Function(RBF)neural networks combined with heuristic algorithms.Initially,the RBF is trained and utilized to denoise and predict data,addressing the impact of noise,insufficient data collection,and loss of battery data.Subsequently,a heuristic algorithm is used for the identification of PEMFC model parameters.The results indicate that after RBF processing,the impact of anomalies on parameter identification is significantly reduced,greatly improving the accuracy of the heuristic algorithm for parameter identification,with a V-I fitting accuracy reaching 99.56%.
作者 刘明群 孟贤 何廷一 和鹏 许珂玮 杨博 LIU Mingqun;MENG Xian;HE Tingyi;HE Peng;XU Kewei;YANG Bo(Yunnan Power grid limited liability power Science Research Institute,Kunming 650000,China;Faculty of Electric Power Engineering,Kunming University of Science and Technology,Kunming 650500,China)
出处 《昆明理工大学学报(自然科学版)》 北大核心 2024年第2期81-90,共10页 Journal of Kunming University of Science and Technology(Natural Science)
基金 国家自然科学基金项目(62263014) 云南省自然科学基金项目(202201AT070857)。
关键词 质子交换膜燃料电池 智能参数辨识 径向基神经网络 启发式算法 降噪处理 预测处理 proton exchange membrane fuel cell intelligent parameter identification radial basis function neural network meta-heuristic algorithm noise reduction prediction process
  • 相关文献

参考文献12

二级参考文献153

共引文献67

同被引文献10

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部