Flux-Oriented Control for Asynchronous Three-Phase Motors Using Internal Model

This paper introduces the stator flux vector-oriented control method for the three-phase asynchronous motor using internal model. In which, the three-layers feed-forward neural network with on-line learning algorithm is applied to design the reverse model for the independent torque and flux controllers, while the forward model is trained to identify the current in the d-q coordinate system of the motor. Finally, a 3-level cascaded inverter system with common-mode voltage reduction algorithm is applied to increase the stability of the controller. The simulation and experiment process are carried out based on the Matlab/Simulink software for 1-hp, 150-rad/s squirrel cage rotor motor. The simulation and experimental results show that the tracking performance of rotor speed is guaranteed at the angular frequency varying from the lowest 3-rad/s up to highest 150-rad/s. In addition, the motor control system remains stable when the stator and rotor resistance values are increased by 2 times in the presence of unknown disturbances.