Global Sensitivity Analysis of Long-Stator Linear Synchronous Motor Based on the Sobol’ Method

Parameter uncertainties in long-stator linear synchronous motors (LSLSMs) can significantly degrade the performance of high-speed maglev systems. This study presents a comprehensive analysis through the following methodology: First, we develop a three-dimensional finite-element model (FEM) based on the LSLSMs configuration used in the TR-08 maglev train. Subsequently, we construct an efficient surrogate model employing Polynomial Chaos Expansion (PCE), enabling computationally tractable sensitivity analysis. Using the Sobol’ method on this surrogate model, we conduct a systematic parameter sensitivity analysis of the motor's levitation force. Our results quantify the relative importance of various design parameters, providing valuable theoretical guidance for the optimization of maglev vehicle propulsion systems. The findings offer practical insights for robust motor design that accounts for parameter uncertainties in operational conditions.