Improved Estimation Procedure of Cage-Induction-Motor-Equivalent Circuit Parameters Based on Two-Stage PSO Algorithm

This paper analyzes errors in the estimation of induction-motor-equivalent circuit parameters using an improved combined two-stage Particle Swarm Optimization (PSO) method. The proposed method accounts for variations in rotor parameters based on both linear and square root speed approximations, as well as two different approaches for the stator and rotor leakage reactance ratios. The first approach assumes that the starting rotor leakage reactance is equal to the stator leakage reactance, while the second considers them as distinct. Improvement of the algorithm consists of increasing the accuracy of the approximations of parameter changes on the rotor. Thanks to more accurate determination of the initial rotor parameters, both approximations provide better results in parameter estimation. The analysis involved sixteen induction motors with four different power ratings and four different pole numbers. The analysis aimed to assess the impact of these approximations and assumptions on equivalent circuit parameter estimation errors. The estimated torque-speed characteristics closely matched the manufacturer’s reference data, including starting, maximum, and full-load torques. The deviation of the estimated torque-speed characteristics from the reference characteristics, within a defined speed range, is defined as the mean absolute percentage error. Based on the obtained results, the mean absolute percentage error is complex and depends on rotor parameter speed approximations, stator and rotor leakage reactance ratios, and the full power of the induction motor.