DEDS故障诊断方法在电梯中的应用

针对电梯这种复杂机电系统难以建模的特点，提出基于随机自动机的离散事件动态系统（DEDS）故障诊断方法．深入研究电梯门系统的故障机理，对各元件的典型故障进行分类，并抽象出各元件的事件集和状态集．将传感器信息和事件发生的可能性信息加入逻辑模型，得到系统的随机自动机模型．把门锁装置从门系统中分离出来，建立电梯门锁装置的随机自动机模型，构建电梯门系统和门锁装置的随机诊断器．通过随机诊断器，对门锁系统故障发生的概率进行在线估计，并对门主体系统的可诊断性进行了分析．此方法为复杂机电系统的故障诊断提供了一条有效途径．

Fault Diagnosis for Elevator Based on DEDS

Elevator, as the complex mechanical system, is hard to modeling, which constrains some fault diagnosis approach, so in this paper DEDS fault diagnosis method is proposed based on stochastic automata. After the failure mechanism of the elevator door system being investigated in detail, the typical faults of components in elevator door system are classified, and the sets of state and event of every component are abstracted. The stochastic automata model is obtained by adding sensor information and the occurrence probability of event to the logical model. The elevator door interlock device is separated from door system, then its stochastic automata model is built. The stochastic diagnosers of the interlock device and the elevator door system are established respectively. With the stochastic diagnosers, the occurrence probability of fault in door interlock device is estimated on line, and the diagnosability of door system is analyzed off-line. The fault diagnosis method based on stochastic automata needn't model the system in detail, which provides an effective fault diagnosis approach for complex mechanical system.