Design Optimization of Parameters for Resolver Software Decoding Based on Surrogate Model Management

In order to improve the adaptability of the resolver software decoding system to the electric drive of commercial vehicles and to suppress torque and speed fluctuations during motor operation, this paper proposes a design optimization method for resolver software decoding parameters based on update management using a surrogate model. Firstly, an evaluation index for precision in position angle outputs from resolver software decoding was established, based on the principle of third-order angle observers, and parameter optimization was clarified for the resolver software decoding system. Next, a surrogate model was constructed for the software decoding system, using the Latin hypercube design (LHD) method and based on a radial basis function. Finally, a genetic algorithm (GA) was utilized to optimize parameters in the resolver software decoding system through a solving process, and a partial encryption method was introduced to iteratively update the surrogate model, yielding a partially optimal solution that meets the error requirements. The study results show that the proposed parameter optimization method is effective in quickly identifying optimal matching parameters in resolver software decoding applied to the electric drive system of commercial vehicles, thus offering an effective approach to improve the operational stability of permanent magnet synchronous motors.