Abstract

PI controllers have been widely used in industrial systems application, because they have a simple structure and offer a satisfactory performance for a wide range operation. However, for varieties of plant parameters and the nonlinear operating condition, fixed gain PI controllers cannot provide the desired control performance. In this paper, advanced PI controller was utilized to control the speed of the induction motor in Direct Torque Control (DTC) for electric propulsion application. The proposed method is developed from conventional PI controller combined with feed-forward neural networks (FFNN). The FFNN is used to tune the gain of PI controller. The effectiveness of the complete proposed control scheme is clarified with a variation of speed reference and load torque applied to the motor. Load torque of induction motor depends on the speed rotation and pitch of propeller. Simulation results show the FFNN tuning technique provides better speed control performance.

Original languageEnglish
Pages (from-to)6135-6145
Number of pages11
JournalInternational Journal of Innovative Computing, Information and Control
Volume7
Issue number11
Publication statusPublished - Nov 2011

Keywords

  • Direct torque control
  • Electric propulsion
  • Neural network
  • PI controller

Fingerprint

Dive into the research topics of 'Feed-forward neural network for direct torque control of induction motor'. Together they form a unique fingerprint.

Cite this