基于神经网络和粒子群算法的永磁同步电机运行与振动性能优化

Optimization of Operation and Vibration Performance of Permanent Magnet Synchronous Motor Based on Neural Network and Particle Swarm Algorithm

为了兼顾新能源驱动电机的运行性能和振动性能,对该型号驱动电机进行综合优化。采用数值计算方法分析8极48槽永磁同步电机的主要结构参数对电机运行性能和振动性能的影响,建立神经网络模型拟合电机结构参数与性能表现之间的映射关系,并基于测试数据集进行验证;使用粒子群算法优化电机的结构参数,从而获得驱动电机振动性能和运行性能之间的非劣最优解。优化后的电机与初始样机相比在驱动性能基本保持不变的基础上,转矩脉动和振动位移方面都有了明显的优化。

In order to take into account the operation and virbation performance of the new energy drive motor,the drive motor of the model was comprehensively optimized.The numerical calculation method was used to analyze the influence of the main structural parameters of the 8-pole 48-slot synchronous motor on the different performance of the motor,and a neural network model was established to fit the mapping relationship between the motor's structural parameters and performance.The particle swarm algorithm was used to optimize the structural parameters of the motor to obtain a non-inferior optimal solution between the vibration performance and the running performance of the drive motor.Compared with the original prototype,the optimized motor had obvious optimization in terms of torque ripple and vibration displacement.