Practical sensorless induction motor drive employing an artificial neural network for online parameter adaptation

Abstract

This paper describes the design and operation of a high performance sensorless induction motor drive. The drive derives its speed measurement from a model reference adaptive system, and hence high dynamic speed and torque response is obtained. The performance of the drive deteriorates as the motor heats up and motor parameters change. A specially trained artificial neural network is used to monitor these changes online. The advantage of this control system is that commissioning of the drive can be fully automated - nonlinear relationships are automatically characterized.