Parameter Estimation for Three-Phase Induction Motor Based on Performance Curves and Differential Evolution

The estimation of electrical parameters in three-phase induction motors is necessary for tasks such as performance analysis, control design, and predictive maintenance. Nonetheless, manufacturers typically provide limited electrical data, and existing methods often require invasive testing or additional sensors, increasing system complexity and cost. Existing approaches predominantly focus on online estimation, which is based on continuous data acquisition and may be impractical in industrial environments with limited hardware resources. Moreover, many offline techniques still require experimental measurements or specific instrumentation, limiting their applicability. This paper introduces a novel and easy-to-implement offline parameter estimation method that eliminates the need for additional instrumentation by leveraging manufacturer performance curves and differential evolution optimization algorithm. Four equivalent circuit models are analyzed to evaluate the impact of core loss resistance and supplementary losses on estimation accuracy. The methodology is experimentally validated on a 1 HP induction motor under varying voltage, frequency, and load conditions, with estimated parameters compared with experimental measurements and IEEE Std 112-2017 test results. Furthermore, the study includes a comparative analysis that includes seven evolutionary algorithms, focusing on solution distribution, computational complexity, and convergence characteristics. The results demonstrate that the proposed method provides accurate parameter estimation while reducing computational effort, establishing it as a low-cost and practical alternative for industrial applications requiring reliable motor modeling and control optimization.