Parameter estimation of induction motors from manufacturer data for simulation purposes

Abstract

Induction motors form a major component of power system loads and account for a significant share of electrical energy consumption. Their operating characteristics influence power system stability, making simulation studies essential for analyzing performance under transient and steady-state conditions. To carry out such studies, electromagnetic transient simulators such as EMTDC/PSCAD™ and RTDS/RSCAD™ provide well-developed induction motor models that require equivalent circuit parameters as input. However, manufacturers typically provide only key performance characteristics—such as power factor, efficiency, and current—on nameplates and in catalog data. The required equivalent circuit parameters are not readily available to the end user. This thesis presents a methodology for estimating the equivalent circuit parameters of three-phase induction motors using readily available manufacturer data. The estimated parameters are intended for use in electromagnetic transient simulations for power system studies. Two distinct non-linear optimization techniques are employed: the Nelder-Mead Simplex method and a variant of Genetic Algorithms. These methods determine equivalent circuit parameters to ensure that the simulated performance characteristics—such as current, efficiency, and power factor—closely match the manufacturer-published data. The proposed approach is applicable to both single-cage and double-cage induction motor models. The quality of the estimated parameters is assessed by evaluating how closely the simulated characteristics match the published manufacturer data. While the Nelder-Mead Simplex method offers computational efficiency and ease of implementation, the Genetic Algorithm excels in global search capability, reducing the risk of getting trapped in local minima. Overall, the results establish the effectiveness of the proposed methods in estimating reliable equivalent circuit parameters for simulation applications. A key contribution of this work is the integration of the estimation method as an interactive tool in RSCAD, designed with a flexible data input structure that accommodates both minimal nameplate data and additional catalog information. This adaptability ensures compatibility with variations in motor ratings and manufacturer standards, enabling the tool to be used directly in real-time electromagnetic transient simulations.