Investigation of a robust adaptive nonlinear controller for induction motors

In high-performance servo systems, the three-phase induction motor is increasingly replacing the DC motor. Induction motors have simple and rugged construction, with fewer working problems and lower costs. A space field-oriented control approach to controlling an induction motor simplifies torque control as compared to that of a separately excited DC motor. For high-performance servo applications, speed and position control must be not only fast and accurate, but also robust to variations in plant parameters or load disturbances. However, a conventional proportional-integral (PI) controller is insufficient for this purpose. Consequently, an efficient control scheme is required to enhance the performance of the PI controller. This paper investigates such a control scheme, that uses neural network concepts to provide speed control of induction motor that is robust to both the dynamic changes in plant parameters and the introduction of load disturbances.