基于感知模糊Petri网的自动装弹机故障分析方法

One Approach for Failure of Auto-Loader Base onPerceptron Algorithm and Fuzzy Petri Net

摘要:

坦克自动装弹机故障发生率高,原因复杂,为了缩短自动装弹机故障诊断时间,提高诊断准确度,在分析了自动装弹机内部原理及获取相关专家经验后,提出了一种基于感知模糊Petri网的故障诊断方法。结合自动装弹机的具体构造建立对应的PFPN故障模型,通过模糊Petri网表示故障传播过程,利用感知机误差反向传递方法学习有限的专家经验,确定Petri网络中触发事件的权值。通过正向推理,实现了对自动装弹机故障的准确判断。逆向推理结合最小割级方法,缩小排查范围,提高推理效率。以旋转故障机为例建立对应的 PFPN故障模型,与故障树模型诊断结果和历史诊断数据对比,验证该故障分析方法的合理性和有效性,实现了对自动装弹机的快速准确的故障诊断。

Abstract:

the failure rate of the tank auto-loader is high and the causes of failure are complicated. In order to shorten the fault diagnosis time of the auto-loader and improve the diagnostic accuracy, after analyzing the internal principle of the automatic loading machine and obtaining relevant expert experience, a perception-based Fault diagnosis method for fuzzy Petri nets. Combined with the specific construction of the automatic loader, the corresponding PFPN fault model is established. The fuzzy Petri net is used to represent the fault propagation process. The perceptron error back-transfer method is used to learn the limited expert experience and determine the weight of the trigger event in the Petri network. Through forward reasoning, accurate judgment of the failure of the automatic loader is realized. Reverse reasoning combined with the minimum cut level method reduces the scope of investigation and improves the efficiency of reasoning. By taking the Rotary ammunition loader as an example, the corresponding PFPN fault model is established. By comparing with the fault tree model diagnosis results and historical diagnostic data, the rationality and effectiveness of the fault analysis method are verified.The fault diagnosis of automatic loader is realized quickly and accurately.