Controlling the Remaining Useful Lifetime using Self- Optimization

Abstract

Self-optimizing mechatronic systems offer possibilities well beyond those of traditional mechatronic systems. Among these is the adaptation of the system behavior to the current situation. To do so, they are able to choose from different working points, which are pre-calculated using multiobjective optimization and are thus Pareto-optimal with regard to the chosen objective functions. In this contribution, a method is presented that allows to continuously control the system degradation by adapting the behavior of a self-optimizing system throughout its complete lifetime. The current remaining useful lifetime is estimated and then related to the spent lifetime and the desired useful lifetime. Using this information, a reliability-related objective is prioritized using a closed-loop control, which in turn is used to determine the working point of the self-optimizing system. This way, the desired useful lifetime can be achieved. To exemplify the setup of the controller structure and to demonstrate the adaptation of the system behavior, a dynamic model of a clutch system is used. It can be seen that the closed loop controller is able to correct for external perturbations, such as changed requirements, as well as changed system parameters. This way, the modeled system is able to achieve the desired lifetime reliably.