Neural-Network-based Programmable State Feedback Controller for Induction Motor Drive

The paper deals with the design of speed and flux state-space controller for induction motor drive. Linear quadratic regulator theory is employed. Optimal controller gain matrix is calculated for a set of operating points. The artificial neural network (ANN) is trained to provide gain matrix for any operating point. Proposed control scheme has been extensively tested in simulation. Results show promising robustness against machine parameters variations. A nonlinear and non-stationary plant control task has been solved with the help of linear time-invariant (LTI) system theory and ANN-based function approximator.