【目的】针对永磁同步电机(PMSM)运行中产生的电磁振动噪声，注入谐波电流是一种有效的主动抑振方法。然而，传统方法在确定谐波电流的参数上存在困难，且忽略了径向电磁力与转矩性能的耦合。【方法】首先，本文以10极60槽PMSM为研究对象。考虑开槽作用，通过二维快速傅里叶变换分析出主导电磁力分量为0阶12倍频，并分析其来源。其次，分析了谐波电流与径向电磁力的关联机制。针对谐波电流与电磁性能间的强耦合特性，构建以抑振性能与转矩性能兼顾的多目标优化模型，采用多目标遗传算法(MOGA)优化电流谐波的幅值和相位。最后，建立电磁-结构耦合的多物理场仿真模型，对优化前后的电机振动响应进行对比验证。【结果】结果表明，电枢反应槽谐波和永磁体磁场的作用对主导径向电磁力贡献最大。此外，注入由MOGA优化的幅值和相位谐波电流后，电机振动加速度可削弱12.75%，转矩脉动降低2.61%，验证了优化算法的有效性和注入谐波电流抑振策略的可行性。【结论】本文所提基于MOGA的谐波电流参数优化方法，在抑制振动与提高转矩性能之间实现了有效平衡，为电磁力成因分析及其主动抑制提供了理论和实践参考。

[Objective] Harmonic current injection is an effective approach for suppressing electromagnetic vibration and noise in permanent magnet synchronous motor (PMSM). However, traditional methods face difficulties in determining the parameters of harmonic currents and neglect the coupling between the radial electromagnetic force and torque. [Methods] Firstly, a 10-pole 60-slot PMSM was taken as the research object in this paper. Considering the slotting effect, the dominant electromagnetic force component was identified as the 0-order 12-time frequency through two-dimensional fast Fourier transform, and its sources were analyzed. Subsequently, the coupling mechanism between harmonic currents and radial electromagnetic forces was analyzed. To address the strong interaction between harmonic currents and electromagnetic performance, a multi-objective optimization model was formulated that considers vibration suppression and torque performance. The amplitudes and phases of the harmonic currents were optimized using multi-objective genetic algorithm (MOGA). Finally, a multi-physics simulation model incorporating electromagnetic-structural coupling was developed to validate the vibration responses of the motor. [Results] The results demonstrated that the armature reaction slot harmonics and permanent magnet field interactions were the dominant contributors to radial electromagnetic forces. Further experimental verification showed that injecting harmonic currents with MOGA-optimized amplitudes and phases achieved a 12.75% reduction in motor vibration acceleration and a 2.61% decrease in torque ripple. These improvements validated both the effectiveness of the optimization algorithm and the feasibility of the harmonic current injection strategy for vibration suppression. [Conclusion] The MOGA-based harmonic current parameter optimization method proposed in this paper achieves an effective trade-off between vibration suppression and torque performance enhancement, providing both theoretical insights and practical references for electromagnetic force mechanism analysis and active suppression strategies.

基金项目: 国家重点研发计划项目(2022YFF1400202)