基于参数优化VMD和多尺度模糊熵的星-三角接法FS-PMM匝间短路故障诊断方法

星-三角（Y-Δ）接法多层绕组结构可以抑制分数槽永磁电机电枢磁场的低次谐波磁动势，提高电机运行效率。但其结构较常规Y接法双层绕组结构更为复杂，发生匝间短路的风险更高。首先，建立Y-Δ接法分数槽永磁电机(FS-PMM)发生匝间短路的电路模型，推导出故障发生后电机三相电流，分析出三相故障电流会受匝间短路位置的影响。其次，利用自适应权重和柯西-高斯变异策略对鲸鱼捕食优化算法进行改进，用于优化变分模态分解参数，提出一种参数优化的变分模态分解算法对电流信号进行处理。最后，通过计算电流和电磁转矩的多尺度模糊熵，提出Y-Δ接法FS-PMM匝间短路故障诊断方法。该方法故障诊断准确率达到95%以上，并可以对Y接法部分和Δ接法部分的匝间短路故障进行有效区分。

Fault Diagnosis of Inter-Turn Short Circuit Fault in Star-Delta Connection FS-PMM Based on Parameter Optimization VMD and Multi-Scale Fuzzy Entropy

The fractional slot permanent magnet motor winding adopts a star-delta (Y-Δ) connection multi-layer winding structure, which can restrain the low order harmonic of armature magnetic field and improve the operation efficiency. However, this structure is more complex than the conventional Y connection double-layer winding structure, and the risk of inter-turn short circuit is higher. Firstly, a circuit model of inter-turn short circuit in Y-Δ connection fractional-slot permanent magnet motor (FS-PMM) is established. The fault three-phase current is derived, and fault current will be affected by the position of the inter-turn short circuit. Secondly, the whale optimization algorithm is improved by adaptive weights and Cauchy Gaussian mutation strategy. The improved whale optimization algorithm is adopted for optimizing the parameters of variational mode decomposition, and a parameter optimized variational mode decomposition algorithm for processing current signal is proposed. Finally, by calculating the multi-scale fuzzy entropy of current and torque, a fault diagnosis method for Y-Δ connection FS-PMM inter-turn short circuit is proposed. The accuracy of this fault diagnosis method can reach over 95%, and it can effectively distinguish the inter-turn short circuit fault between the Y connection part and the Δ connection part.