集成模糊推理与定量仿真的故障预测系统研究

为解决传统故障诊断方法在装备维修保障方面的不足，将虚拟样机作为一种新的定量推理机制引入到故障仿真和预测领域，提出了基于虚拟样机的装备故障仿真预测模型。以某型自行火炮为研究对象，在ADAMS环境下建立了火炮虚拟样机。仿真预测过程根据用户输入的初始参量由虚拟样机获取初始状态，由基于动态模糊综合评判的预测方法生成候选故障集，虚拟样机结合知识库确认实际发生的故障并找出相应的故障原因和故障部件。虚拟样机的仿真结果和相关的预测实例验证了模型的有效性，表明基于上述模型的故障仿真预测系统可以获得较高的预测精度，能够满足装备预知维修保障的需求。

Fault Forecast System Based on Integration of Fuzzy Inference and Quantitative Simulation

To solve the deficiencies of traditional fault diagnosis methods in equipment maintenance, virtual prototyping was firstly adopted into fault simulation and prediction fields as a new quantitative reasoning mechanism, The modeling process and designing idea of fault simulation and prediction system based on virtual prototyping was illustrated. Virtual prototyping of a type of Self-Propelled Gun was established in ADAMS. The simulation process obertained the initial state by virtual prototyping from the inputted parameters and provides a possible fault set from the prediction model based on dynamic fuzzy comprehensive evaluation. Then virtual prototyping confirmed the factual faults and found out the corresponding fault components and causations. Simulation results of virtual prototyping and prediction examples prove that the fault simulation and prediction system has adequate accuracy and be able to meet the requirements of predictive maintenance in future tasks.