Line-starting permanent magnet synchronous motors (LSPMSM) are different from other permanent magnet motors due to their double-sided slotted characteristics of the stator and rotor, making it difficult to analyze the cogging torque. The overall optimization of motor cogging torque and torque ripple rate is also challenging. This paper optimizes the rotor design of the LSPMSM based on multi-objective particle swarm optimization algorithm, considering the saturation performance of the motor.

Firstly, this paper establishes the relationship between harmonic magnetomotive force and cogging torque to predict the cogging torque of the LSPMSM. The cogging torque is analyzed as a dynamic function relationship of various motor parameters, providing corresponding optimization parameters for the subsequent optimization of the prototype. At the same time, under the premise that the maximum magnetic energy product of the permanent magnet remains unchanged, the relationship between the saturation degree of the motor and the parameters of the permanent magnet is determined, and the selection range of the motor's permanent magnet parameters is determined based on this relationship. Define three working states of the motor: unsaturated, peak saturation, and after saturation, to make the analysis and optimization results of the motor more accurate.

Secondly, combined with response surface methodology (RSM) and multi-objective particle swarm optimization algorithm (MOPSO), this paper proposes a comprehensive optimization strategy to optimize key objectives such as motor cogging torque, torque ripple rate, and efficiency. The optimal design solutions in different regions can be obtained by using the response surface algorithm to classify numerical groups and shorten the computation time of the particle swarm optimization algorithm. The optimal solutions under six different conditions are determined after considering whether the stator is skewed and the three working states of the motor.

Finally, simulation analysis verifies the relationship between cogging torque and harmonics under different saturation states, the trend of no-load back electromotive force, and the main harmonic order changing with the number of slots per pole of the rotor. An experimental platform is built to verify the accuracy of theoretical and simulation analysis. This paper provides optimization solutions for the design of motors of the same type.