基于模糊贝叶斯网络的门座式起重机起升机构电气故障诊断的研究

Research on Electrical Fault Diagnosis of Hoisting Mechanism of Portal Crane Based on Fuzzy Bayesian Network

为了明确门座式起重机起升机构电气故障发生的关键因素,按照相关资料文献,利用贝叶斯网络构建了拓扑结构,并结合模糊理论的相关知识,利用三角模糊数和条件概率表(CPT)得出事件的条件概率,依据贝叶斯网络推理得出故障诊断的后验概率。利用Ge NIe软件的诊断推理和灵敏度分析,可以准确找出导致门座式起重机起升机构发生故障的最关键因素。将计算得到的结果与实际经验对比表明该方法对于诊断门座式起重机起升机构故障是切实可行的。

In order to clarify the key factors of electrical failure of hoisting mechanism of gantry crane, according to the relevant documents, the topology structure is constructed by using Bayesian network, and the relevant knowledge of fuzzy theory is combined. The conditional probability of event is obtained by using triangular fuzzy number and conditional probability table, and the posterior probability of fault diagnosis is obtained by Bayesian network reasoning. By using the diagnostic reasoning and sensitivity analysis of GeNIe software, the most critical factors that cause the hoisting mechanism of gantry crane can be found out accurately. The compar-ison between the calculated results and the actual experience shows that this method is feasible for diagnosing the hoisting mecha-nism faults of gantry cranes.