电动汽车永磁同步电机匝间短路检测算法仿真
Simulation of Inter-Turn Short-Circuit Detection Algorithm for Permanent Magnet Synchronous Motor in Electric Vehicles
王良成 1汪源 1张永辉2
作者信息
- 1. 三亚学院理工学院,海南 三亚 572022
- 2. 海南大学信息与通信工程学院,海南 海口 570228
- 折叠
摘要
当电机出现匝间短路故障时,电机内部电阻会大幅度降低,电流突增,造成其余元件功率过大问题.若不能及时监测该故障电机将会导致二次损坏.但是,由于短路故障具有瞬时性,其特征获取难度较大.为此提出电动汽车永磁同步电机匝间短路故障检测方法.构建驱动汽车的永磁同步电机模型,依据当永磁同步电机模型处于匝间短路故障状态时,基波电流与正序电流的制约关系失效的原理,提取电机匝间短路故障特征.基于此利用粒子群算法-最小二乘支持向量机(Particle swarm optimization-Least Squares Support Vector Machine,PSO-LSSVM)获取故障检测结果,实现电动汽车永磁同步电机匝间短路故障的检测.实验结果表明,研究方法在任意时刻检测到的负载力矩均与实际值吻合,且输出的电机残余能量具有较高可靠性,说明了上述方法具有较强的可应用性.
Abstract
When a turn-to-turn short circuit fault occurs in the motor,the internal resistance of the motor will be significantly decrease,and the current will suddenly increase,resulting in excessive power of other components.How-ever,due to the transient nature of short circuit fault,it's difficult to detect the features.Therefore,a method for detec-ting turn-to-turn short circuit of permanent magnet synchronous motor of electric vehicle was put forward.At first,a model of permanent magnet synchronous motor(PMSM)was constructed.According to the principle that the restric-tion relationship between fundamental current and positive sequence current failed when the PMSM model was in the state of turn-to-turn short circuit,the characteristics of turn-to-turn short circuit fault were extracted.On this basis,particle swarm optimization-least squares support vector machine(PSO-LSSVM)was used to obtain the fault detec-tion results.Finally,the detection of turn-to-turn short circuit fault of permanent magnet synchronous motor was a-chieved.Experimental results show that the load torque detected by the research method at any time is consistent with the actual value,and the output residual energy of the motor has high reliability,so the method has strong applicabili-ty.
关键词
永磁同步电机/匝间短路原因/故障特征/改进粒子群算法/支持向量机Key words
Permanent magnet synchronous motor/Cause of turn-to-turn short circuit/Fault characteristics/Im-proved particle swarm algorithm/Support vector machine引用本文复制引用
出版年
2024
NETL
NSTL
万方数据