模糊神经网络专家系统在动力锂电池组故障诊断中的应用  

Fuzzy neural network expert system for fault diagnosis in power lithium battery application

作　　者：王一卉[1,2] 姜长泓[1] 

机构地区：[1]长春工业大学电气与电子工程学院,长春130012 [2]长春职业技术学院,长春130000

出　　处：《电测与仪表》2015年第14期118-123,共6页Electrical Measurement & Instrumentation

基　　金：吉林省科技发展计划项目(20140204029sf)

摘　　要：动力锂电池故障的产生原因具有一定的复杂性和不确定性。为此,提出一种基于模糊神经网络的故障诊断专家系统,该方法结合了模糊数学,神经网络以及专家系统的优点。用模糊数学可以将症状模糊化以表征故障的隶属度、神经网络具有良好的自学习能力、专家系统具有推理能力强,三者的相互结合,即提高了系统的准确性和可操作性,又满足了对故障诊断智能化、自动化的要求。试验结果表明该方法可以准确的判断出系统的故障,不仅将故障检测的精度提高到,预测误差在之间,而且检测时间大大缩短。提高了动力锂电池的自适应能力,自主学习能力,为动力锂电池故障诊断提出了一种科学高效的新方法。The cause of power lithium battery failure has a certain complexity and uncertainty. To this end, this paper proposes a fault diagnosis expert system based on fuzzy neural network. This method combines the advantages of fuzzy mathematics, neural network and expert system. Using fuzzy mathematics can be blurred to characterize the member- ship degree of the fault symptoms, the neural network has good self-learning ability, the expert system has strong rea- soning ability, All three together, that is not only to improve the accuracy of the system and operability, but also meet the requirement of the intelligent and automatic diagnosis for faults~ The test results show that the method can accurate- ly judge the fault in the system, it not only to increase the accuracy of fault detection to 0.001, control the prediction error at between 1% and 8 %, but also shorten the testing time. This method improves the self-adaptive ability of the power lithium batteries, the independent learning ability, and puts forward a new scientific and efficient method for power lithium battery fault diagnosis.

关 键 词：模糊 神经网络 动力锂电池 故障诊断 专家系统 

分 类 号：TM711[电气工程—电力系统及自动化]