Identification of induction motor parameters from transient stator currentmeasurements

This paper describes three methods for estimating the lumped model paramete rs of an induction motor using startup transient data. A three-phase balanc ed induction motor is assumed. Measurements of the stator currents and volt ages are required for the identification procedure, but no measurements fro m the motor shaft are needed. The first method presented applies simple mod els with limited temporal domains of validity and obtains parameter estimat es by extrapolating the model error bias to zero, This method does not mini mize any specific error criterion and is presented as a means of finding a good initial guess for a conventional iterative maximum-likelihood or least -squares estimator. The second method presented minimizes equation errors i n the induction motor model in the least-square sense using a Levenburg-Mar quardt iteration, The third identification method is a continuation of the Levenburg-Marquardt method, motivated by observed properties of some pathol ogical loss functions. The third method minimizes errors in the observation s in the least-squared sense and is, therefore, a maximum-likelihood estima tor under appropriate conditions of normality. The performance of the ident ification schemes is demonstrated with both simulated and measured data, an d parameters obtained using the methods are compared with parameters obtain ed from standard tests.