基于D-S融合的电子电路故障诊断研究

Study on electronic circuit fault diagnosis based on D-S evidential theory

根据电子电路故障元件诊断的不确定性问题,给出了多传感器D-S信息融合实现电路故障诊断的方法。通过检测电路工作时电子元件的温度和关键点电压两方面数据信息,得出两传感器对各待诊断元件的信度函数分配。再利用D-S联合规则结合模糊逻辑理论,得到融合后的信度函数分配,从而确定故障元件。实验结果说明:多传感器融合算法具有较高的准确率。

According to uncertain problem of circuit fault diagnosis, a method of using Dempster-Shafer （D-S） evidential theory and multi-sensor information fusion theory to realize circuit fault diagnosis is presented. The belief function assignment of two sensors to circuit is obtained through the measurement results of the temperature and voltage of circuit component. The fusion belief function assignment is gotten by using D-S rule and fuzzy logic theory ,and fault component is found. Experimential results show that multi-sensor information fusion is proved more accurate.