基于键合图模型的新型故障树故障诊断方法

New Type of Fault Tree Fault Diagnosis Method Based on Bond Graphic Model

针对传统故障树的生成及应用缺陷,研究了基于键合图模型的新型故障树故障诊断方法。该方法根据系统键合图模型推导出时间因果图,进而建立系统故障树,由此产生一个用于反映系统行为变化的假设故障集合;同时,推理预测每个故障的后效特征,并与观测特征相比较,修正假设故障集,最终定位故障源。仿真结果验证了此故障诊断方法应用的可行性和故障诊断的准确性,具有广泛的应用前景。

To against the defects in application of traditional fault tree,the fault diagnosis method of establishing system fault tree based on bond graph mode is researched.In accordance with the bond graphic model of the system,the time causal graph（TCG） is derived;then the fault tree of the system is set up.Thus the hypothetical fault set reflecting the behavior variation of the system is created.In addition,the aftereffect characteristics of each fault is predicted,and compared with the observed characteristics to correct the hypothetical fault tree for locating the fault source.The result of simulation verifies the feasibility and precision of this method,and shows the wider applicable prospect.