执行器隐含退化下系统寿命预测与延寿方法

Research on remaining useful life prediction and life extension of control system with implicit degradation of actuators

针对执行器退化监测数据存在测量误差的闭环控制系统寿命预测与延寿控制问题,首先,以带测量误差的非线性Wiener模型描述了不确定测量值与执行器隐含退化状态之间的关系,并基于卡尔曼滤波技术估计出实际退化状态;其次,通过定义因执行器隐含退化而驱使控制系统退化的三个阶段,结合动态与稳态性能约束集确定系统退化失效阈值,并据此及执行器退化状态的估计值得到控制系统剩余寿命预测分布的解析解;然后,将执行器退化状态纳入线性二次调节器(LQR)控制器设计中,提出了基于执行器实时退化状态的自适应延寿控制策略,通过折衷平衡系统性能与控制作用来延缓执行器的退化,从而实现延长控制系统生命周期的目的;最后,通过双容水箱液位系统验证了所提出方法的有效性.

Aimed at the problems of life prediction and extended life control of closed-loop control systems with measurement errors in actuator degradation monitoring data,firstly,the relationship between the uncertain measurement data and the implied degradation state of the actuator was described by the nonlinear Wiener model with measurement error,and the actual degradation state was estimated by Kalman filtering technique.Secondly,combined the dynamic and steady-state performance constraint sets,the system’s degradation failure threshold was determined by the three degradation stages that drive the control system to be degraded due to the actuator’s implicit degradation.Through the estimation of system degradation threshold and actuator degradation state,the analytical solution of residual life prediction distribution of control system was obtained.Then,the actuator degradation state was incorporated into the design of the linear quadratic regulator(LQR) controller,and an adaptive life extension control strategy based on the real-time degradation state of the actuator was proposed.By the trade-offs between control system performance and control action to slow degradation the actuator,thus the purpose of extending the life cycle of the control system was achieved.Finally,the validity of the method in the paper was verified by the double-tank water tank level system.