H∞ Control Using Linear Parameter Varying Approach for Motion Control Systems Under Communication Delays: Application to PMSM

Abstract

Over the past decades, various control methodologies have been studied to improve the performance of motion control systems. However, such methods have no general guidelines for controller gain tuning or performance optimization in the presence of nonlinearities and/or uncertainties. Linear parameter-varying (LPV) control methods have recently been studied to improve the performance of motion control systems by treating the nonlinearities as parameter variations. By contrast, with motion control systems, the plant, microprocessors, controllers, actuators, and sensors are interconnected using a communication network, where the network inherently induces communication delays while sending and receiving information. Hence, when designing an LPV control system, it is necessary to consider both communication delays and nonlinearities. In this study, we address an LPV-approach based H∞ control problem for motion control systems under communication delays between the actuator and plant, and between the sensor and plant. We obtained a linear matrix inequality (LMI) condition under which a closed-loop system using the proposed controller satisfies the H∞ control performance and stability when subjected to a communication delay. The main technique applied to obtain the corresponding LMI condition for an LPV system is interpolating the scheduling variables of the nonlinear parameter variations at each vertex. Finally, we applied the proposed control method to the driver set of a permanent magnet synchronous motor and verified the H∞ control performance under a communication delay.