Characteristic model–based adaptive fault-tolerant control for four-PMSM synchronization system

Abstract

In four-motor servo systems, actuator failures influence control performance seriously through huge inertia ratio changes and unknown disturbances. To solve this problem, an adaptive fault-tolerant control scheme based on characteristic modeling and extended state observer is proposed. First, an adaptive sliding mode observer is designed as fault detection part and offers motor information for controller. Second, to simplify complex dynamic model, this servo system is described by a second-order difference equation. This model integrates uncertainties into three time–varying characteristic parameters to reflect system status. Third, a discrete-time extended state observer is applied to restrain system error caused by actuator failure. Then, a fault-tolerant controller is designed based on characteristic model, and its stability is guaranteed in the sense of Lyapunov stability theorem. These four parts make up the adaptive control scheme and its effectiveness in system control, and fault tolerant is evaluated by both simulation and experiment results.