基于PCA-VNWOA-LSSVM的感应电机轴承故障诊断

Fault diagnosis of induction motor bearing based on PCA-VNWOA-LSSVM

轴承作为感应电机的关键部件,其运行状态直接影响船舶电力拖动系统安全。为解决船舶感应电机轴承故障诊断难题,本文提出一种基于PCA-VNWOA-LSSVM的故障诊断模型。选用美国凯斯西储大学轴承振动数据,利用离散小波分解(discrete wavelet transformation, DWT)从振动信号中提取内圈、外圈和滚动体故障特征,按不同故障类型和直径进行分组、主成分分析(principal component analysis,PCA)降维,结合改进的鲸鱼优化算法(von neumann whale optimization algorithm, VNWOA)对最小二乘支持向量机(least squares support vector machine, LSSVM)初始参数δ^(2)和γ寻优,搭建其故障识别模型,最后将遗传算法(genetic algorithm, GA)和粒子群算法(particle swarm optimization, PSO)的寻优诊断结果与之对比。结果表明:基于PCA-VNWOA-LSSVM的模型故障诊断精度高,且具有良好的稳定性及诊断速度。

As a key component of induction motor,the running state of bearing has a direct impact on the safety of ship electric drive system.In order to solve the problem of bearing fault diagnosis of marine induction motor,we propose a fault diagnosis model based on PCA-VNWOA-LSSVM in this paper.Based on the bearing vibration data of Case Western Reserve University,discrete wavelet decomposition(DWT)is used to extract the fault features of inner ring,outer ring and rolling element from the vibration signals,which are grouped according to different fault types and diameters.And further,principal component analysis(PCA)dimensionality reduction is carried out.Combined with the improved whale optimization algorithm(VNWOA),the initial parameters of the least squares support vector machine(LSSVM)-δ2 andγare optimized.Finally,the optimization diagnosis results of genetic algorithm(GA)and particle swarm optimization(PSO)are compared with them.The results show that the fault diagnosis model based on PCA-VNWOA-LSSVM has high accuracy,good stability and diagnosis speed.