Speed estimation of induction motor based on hybrid PI and neural networks controller

A novel speed estimation method of an induction motor is discussed in this paper. This speed estimation method is based on a flux observer and a hybrid PI and neural networks controller (HPINNC). The flux observer is used to provide cl-axis rotor flux. The HPINNC is used for the realization of speed estimation instead of conventional PI controller utilizing q-axis rotor flux converging on zero. In HPINNC, the input and the hidden neurons are similar to PI controller, and their weights are constant. Learning of HPINNC is realized by auto-tuning of connective weights between the hidden and the output neurons. The simulation results show that the proposed scheme has good performance at either variable-speed operation or load variation and low sensitivity to load disturbances compared with conventional PI control scheme.