航天复杂系统的加权模糊Petri网故障诊断建模

Fault diagnosis method for complex aerospace systems based on weighted fuzzy Petri net

针对模糊Petri网模型难以解决航天复杂系统故障诊断中存在的状态组合空间爆炸、故障关系不清晰、故障源对故障影响程度等问题,提出了基于加权模糊Petri网的故障诊断方法,同时定义了加权模糊Petri网转化成分层着色Petri网的规则。该方法将加权思想与模糊Petri网相结合,建立加权模糊Petri网模型;将加权模糊Petri网转换为分层着色Petri网模型;通过图形反向推理诊断思想实现加权模糊Petri网的知识表示与推理规则,诊断故障发生的原因。最后,对某航天测试系统的一级稳定系统中的放大器系统故障诊断实例进行了仿真,结果表明:该方法对本实例的应用效果较好,故障传播路径清晰,对于规则变化具有良好的适应性,为其他复杂系统的故障诊断提供了一定的参考。

To solve the problem that the fuzzy Petri net model is difficult to solve the state space explosion, fault relation ambiguity and impact of fault source on fault between aerospace complex system in fault diagnosis, the Weighted Fuzzy Petri Net (WFPN) fault diagnosis method was proposed. In addition, the rule for transferring WFPN to Hierarchical Colored Petri Net (HCPN) was defined. This method combined weighted idea with fuzzy Petri net and transfered WFPN to HCPN model to diagnose the cause of fault by graphical reverse reasoning diagnosis idea. Through the fault diagnosis simulation for the amplifier system in primary stabilization system of aerospace test system, the effectiveness of the proposed method was proved. It had clear fault propagation path and good adaptability to rule-changing, and offered a reference to fault diagnosis of complex systems.