Utilise Statistical Analysis to Develop an Efficient Parameter Estimation Retuning App for the Direct Torque Control of AC Induction Motors

Induction motors are the most widely adopted alternating current machines in industry. Requirements for precise and efficient control of such motors has led to developments in high performance drives, namely direct torque control. The induction motors parameters must be ascertained to accurately tune the drive to its optimum performance, failing to do so would result in an inadequate response from the drive, leading to sluggish dynamics and premature deterioration of the motor and drive. The process of optimizing the drive parameters is laborious, costly and time consuming. MathWorks provides software to model and modify the parameters of a direct torque control drive, however the trial and error retuning methods are tedious. This paper presents the development of an AC induction motor and direct torque control drive parameter retuning app which can accurately predict the motor and drive parameters and then simulate its performance using solely the motor nominal power rating. The app utilises the AC4 block and pre-set induction motor models from within the Math Works Specialised Power Systems library. The effectiveness of the app was first tested with known data points and then at random query points and operating conditions, all of which produced satisfactory results in significantly reduced time.