摘要
针对风电机组低速齿轮箱故障的故障特点,提出了一种应用小波包分析(WPA)和最小二乘支持向量机(LS-SVM)相结合的故障诊断方法。将低速齿轮箱不同故障状态下的振动信号经小波包分解后获得各频带能量,经过归一化处理后作为特征向量构成训练样本和测试样本。通过训练样本训练LS-SVM故障诊断模型,用测试样本检验LS-SVM故障诊断模型的正确率。实验结果表明,WPA和LS-SVM相结合的故障诊断方法具有良好的诊断效果。
A method for the fault characteristics of low- speed gearbox fault diagnosis of wind turbine is pro- posed by means of the wavelet packet analysis （WPA） and least square - support vector machine （LS - SVM）. The energy of frequency bands generated by wavelet packet decomposition of the low - speed gearbox vibration signals in different fault states is normalized as eigenvectors, thus forming training and testing samples of LS - SVM fault classifi- er. The LS- SVM fault diagnosis model is trained through the training samples and the accuracy is tested with the test- ing sample. The result shows that, the fault diagnosis method based on the WPA and KS- SVM has good diagnostics effect.
出处
《机械传动》
CSCD
北大核心
2013年第10期129-133,共5页
Journal of Mechanical Transmission
基金
中央高校基本科研业务费专项资金(13MS102)
关键词
小波包
最小二乘支持向量机
低速齿轮箱
故障诊断
Wavelet packet Least square- support vector machine Low- speed gearbox Fault diagnosis
