软件导刊Issue(9)：23-25,3.DOI:10.11907/rjdk.143198
D-S证据理论和多模型结合的模拟电路故障诊断
Fault Diagnosis of Analog Circuit Based on a Combination of D-S Evidence Theory and Multi-model
胡秀洁 1孙艳玉 1宋家友1
作者信息
- 1. 郑州大学信息工程学院，河南郑州450001
- 折叠
摘要
Abstract
In order to increase the precision and accuracy of analog circuit fault diagnosis ,the paper uses information fusion method for fault diagnosis .First ,take the output gain under different frequency as characteristic parameters ,three meth-ods ,including ANFIS model ,BP model and RBF model ,are considered as primary diagnosis ,which can obtain independ-ent evidence .Then use the method of D-S evidence theory evidence to decide the fusion fault location ,and join the local di-agnostic accuracy of the acquisition of basic probability assignment .Through its application in a circuit ,it indicates that by multi-level information fusion treatment ,not only significantly increase the credibility of diagnostic conclusions ,but also improve the diagnostic accuracy .关键词
故障诊断/信息融合/BP神经网络/RBF神经网络/ANFIS/证据理论Key words
Fault Diagnosis/Information Fusion/BP Neural Network/RBF Neural Network/ANFIS/Evidence Theory分类
信息技术与安全科学引用本文复制引用
胡秀洁,孙艳玉,宋家友..D-S证据理论和多模型结合的模拟电路故障诊断[J].软件导刊,2014,(9):23-25,3.
