Neural network based speed identification for speed-sensorless induction motor drives

Controlled induction motor drives without mechanical speed sensors at the motor shaft have the attractions of low cost and high reliability. For these speed sensorless AC drive system, it is key to realize speed estimation accurately. Because the induction motors have some inherent characteristics such as multivariate, parameter indeterminacy, strong coupling and non-linearity, speed sensorless observer scheme based on model usually have some limitations. A speed identification scheme based on fuzzy theory and neural network was presented in this paper. Simulation results show that the fuzzy inference based neural network speed identification has not only the advantage of accurate identification, but also the virtue of quick learning convergence speed.