摘要
在电路故障诊断中,可通过直流分析、交流分析和灵敏度分析等方法,对电路的故障进行诊断。但由于不同的诊断方法对不同的故障敏感度不同,使得每种方法都带有局限性。为此,本文提出了采用数据融合进行电路故障诊断的新方法,介绍了D-S证据理论算法在电路故障诊断中的应用,给出了具体算法和仿真实例。理论分析和仿真结果表明,将数据融合技术用于电路的故障诊断是可行的。不同的诊断方法提供的信息经多次融合、反复抽取有用信息后,大大降低了判断的盲目性,提高了电路故障诊断的准确性。
Circuit faults can be diagnosed through DC analysis, AC analysis, sensitivity analysis and so on. Each has its own advantages and disadvantages. Sometimes, the diagnosis result is not reliable when only one method is used. For this reason, a new idea using data fusion technology in the circuit fault diagnosis is put forward in this paper. The application of the D-S algorithm to the circuit fault diagnosis is presented. The processes and results of data fusion simulation are described. According to the theory analysis and the simulation result, it is feasible that the data fusion is used to the circuit fault diagnosis. After the information provided by different diagnosis method is fused several times, the blindness of decision is reduced obviously. The method raises the accuracy of circuit fault diagnosis.
出处
《电路与系统学报》
CSCD
北大核心
2008年第3期30-34,共5页
Journal of Circuits and Systems
关键词
电路故障诊断
数据融合
证据理论
circuit fault diagnosis
data fusion
evidence theory
作者简介
许军（1961-），女，教授，博士，主要研究方向为信号检测与估计，信息融合，电路故障诊断等。
