Identification of electrical parameters and rotor speed of induction motor using radial basis neural network

A technique for parameters and rotor speed of induction motor using the combination of high-gain observer and radial basis function neuronal predictor is treated in this paper. The algorithms developed here are potentially useful for the design of the drives that can adjust controller parameters automatically. Another possible application is for the detection of failure. In the first scheme, leakage coefficient and rotor time-constant are estimated using the measured rotor speed, stator current and voltage. In the second scheme, rotor resistance is estimated when the rotor speed is available while in the third scheme, both rotor resistance and rotor speed are identified. All the parameters are considered to be time-varying and short-circuit failure is simulated in the rotor resistance. Simulation results illustrate the effectiveness of this technique.