A sensorless neural model reference adaptive control for induction motor drives

Abstract

In this paper, a high performance speed control approach using artificial neural networks (ANNs) and Fuzzy Logic for the Field Oriented induction motor (IM) is proposed. This control method is developed using model reference adaptive control (MRAC) to improve the performance of the IM speed. By using an adaptive neural network controller (ANNC) in the MRAC method, the speed of an IM can be controlled to follow an arbitrarily selected speed trajectory. In particular, an accurate tracking of the speed can still be obtained when uncertainties in the motor and its load exists. The Fuzzy logic is used for the neural controller adaptation to improve the robustness of the generated command. In the proposed control scheme, the Model Reference Adaptive System (MRAS) is used to estimate the rotor speed and the rotor time constant parameter variation. The algorithm is verified by a series of computer simulations and the obtained results show that the designed control system can achieve satisfactory performances in the tracking to the speed trajectory and to the adaptability of the rotor time constant variation.