基于D-S证据理论信息融合的转辙机故障诊断方法研究

Fault Diagnosis Method for Switch Machine Based on D-S Evidence Theory Information Fusion

在列车提速后S700K型电动转辙机被普遍安装在正线道岔的背景下,本文针对单一故障诊断方法的诊断精度偏低问题,提出了基于信息融合故障诊断模型和故障诊断方法.该方法分别用BP神经网络和模糊综合评判对转辙机进行故障诊断,利用神经网络输出和模糊综合评判输出来构造D-S证据理论中的概率分配,然后利用D-S证据理论将BP神经网络和模糊综合评判对转辙机的故障诊断结果在决策级进行融合,诊断转辙机是否有故障并判断故障的模式.诊断结果表明,该诊断方法具有较高的故障诊断精度,诊断结论的可信度有明显提高.

Since the train speed was increased, S700K Electric Switch Machine has been widely installed in the line switch. In addition, the single fault diagnosis method shows low precision. Under such background, the paper proposes the fault diagnosis method based on the module of information fusion. This method works in a way as follows： the fault is diagnosed by using the BP neural network and fuzzy comprehensive evaluation, the basic probability assignment function of D-S evidence theory is structured by using the output of the neural network and the fuzzy comprehensive evaluation; according to the D-S evidence theory, the diagnosis result is then fused on the decision-level to determine whether the Electric Switch Machine has fault and the type of the fault. The result indicated that this diagnostic method has a high accuracy of fault diagnosis and high credibility.