摘要
可靠性低和耐久性差是目前制约质子交换膜燃料电池发展的主要瓶颈,故障诊断技术作为解决这些问题的重要途径之一受到广泛关注。总结了近年来基于典型机器学习算法的质子交换膜燃料电池故障诊断方法,分别对神经网络、模糊聚类、神经模糊、贝叶斯网络、支持向量机、随机森林等方法进行归纳分析,并结合当前算法存在的问题,对机器学习在燃料电池故障诊断应用领域的发展方向进行了探讨。
The problems of low reliability and poor durability are the main bottleneck restricting the development of proton exchange membrane fuel cells.As one of the important ways to solve these problems,fault diagnosis technology has received extensive attention.The state art of fault diagnosis techniques of proton exchange membrane fuel cells based on typical machine learning algorithms was summarized.The methods widely used in fault diagnosis were analyzed,such as neural network,fuzzy clustering,neuro-fuzzy,Bayesian network,support vector machine and random forest.Combined with the advantages and disadvantages of the current algorithms,the outlook of the application of machine learning technology in fuel cell fault diagnosis was discussed.
作者
张杰
谌祺
韩小涛
ZHANG Jie;CHEN Qi;HAN Xiaotao(Wuhan National High Magnetic Field Center,Huazhong University of Science and Technology,Wuhan Hubei 430074,China;State Key Laboratory of Advanced Electromagnetic Engineering and Technology,Huazhong University of Science and Technology,Wuhan Hubei 430074,China)
出处
《电源技术》
CAS
北大核心
2022年第7期710-715,共6页
Chinese Journal of Power Sources
基金
国家自然科学基金项目(52077092)。
关键词
质子交换膜燃料电池
故障诊断
机器学习
proton exchange membrane fuel cell
fault diagnosis
machine learning
