A feasibility study of two adaptive-control methods for vector-controlled induction motor drives

Abstract

The authors evaluate two adaptive control methods for vector-controlled AC drives under parameter variations, namely a minimum variance controller (MVC) and a model reference exact-matching (MREM) method. They compare them to a nonadaptive model reference complementary controller (MRCC) and address their potential impact on designing PWM (pulse width modulation) inverters. Digital simulation results indicate that although all the methods considered had similar responses for nominal conditions, the MVC and MREM exhibited a higher performance than the MRCC under variations of the rotor time constant. However, the MVC PWM signal placed a more severe burden on the PWM stage. It is concluded that a comprehensive solution for the problem of parameter variations should include a combination of an identification method for the rotor parameters and adaptive control method to meet the desired specifications (rise time, bandwidth, speed range, etc.).<>