基于可传递置信模型的模拟电路融合诊断方法研究  

Fault location of analog circuits using transerable belief model

作　　者：杨易旻[1] 彭敏放[1] 王佩丽[1] 

机构地区：[1]湖南大学电气与信息工程学院,长沙410082

出　　处：《微计算机信息》2010年第7期135-137,共3页Control & Automation

基　　金：基金申请人:彭敏放;项目名称:基于多元信息智能融合的模拟电路在线故障诊断研究;基金颁发部门:国家自然科学基金委(60673084);基金申请人:彭敏放;项目名称:基于多层融合模式的模拟电路智能诊断研究方法;基金颁发部门:湖南省科技厅(06JJ4075)

摘　　要：本文提出了一种有效的基于可转移置信模型的模拟电路故障融合诊断方法,解决模拟电路诊断中故障特征信息缺乏和决策融合中异质信息不相容等问题。该法首先采用所提出的电路温度故障信息的提取规则,将获得的温度故障信息输入隶属度函数进行故障预识别,再结合基于可测点电压神经网络故障预识别结果,利用可传递置信模型进行电路融合诊断。通过具体电路实验仿真,表明了该方法的有效性,提高了模拟电路融合诊断准确率。A new analog circuit diagnosis method based on transferable belief model is proposed so as to solve the problems of hard to extract sufficient information for fault diagnosis and obtain reliable fusion results with inconsistent data.The characteristic temperature information is extracted by proposed classifier and then primary fault identification is performed based on membership function from this information.Meanwhile,another separate primary fault diagnosis is dealt with by a neural network based on accessible node voltages.By using transferable belief model,the synthetic diagnosis is eventually completed.The experiment results show that the feasibility of this proposed method is verified and gains satisfactory accuracy.

关 键 词：故障定位 信息融合 不一致性 可传递置信模型 模拟电路 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TN431.1[自动化与计算机技术—控制科学与工程]