On-line Tuning of Torque Controllers for Induction Machines

Anton Hofer^*, Bojan Grcar, Gorazd Stumberger and Peter Cafuta

In any torque control concept for induction machines the capability to produce a desired torque with the minimum stator current (maximum torque per ampere ratio) becomes a demanding requirement. In this contribution it is shown how a recently published torque control strategy which provides the maximum torque per ampere ratio feature can be augmented by some self-tuning capabilities. Since the behaviour of the proposed controller is mainly influenced by the rotor time constant which is subjected to variations due to thermal influences on the rotor resistance a method for an on-line tuning of this parameter is developed. It is based on the extremum seeking approach, which is a well established on-line optimization method especially in the field of adaptive control. A discrete time version of the extremum seeking method combined with a time optimal flux controller is implemented. The performance of the proposed self-tunig concept is demonstrated by experimental results.

Mathematics Subject Classification: 49J21

Keywords: Extremum seeking; Torque control; Efficiency maximization

Minisymposion: Modelling and Control of Mechatronic Systems