基于小波包和SVM的风机齿轮箱故障诊断方法  

Wind Turbine Gearbox Fault Diagnosis Method Based on Wavelet Packet and SVM

作　　者：欧淇源 姚为星 周求湛[1] 程文阁 吴艳茹[2] 

机构地区：[1]吉林大学通信工程学院,长春 [2]天津海运职业学院,天津

出　　处：《声学与振动》2013年第4期37-43,共7页Open Journal of Acoustics and Vibration

摘　　要：为了对风力发电机组中故障高发的核心部件齿轮箱进行实时监控及故障分析,提出了一种根据齿轮箱在工作时不同部位所产生的振动信号及齿轮箱常见故障事件的分析方法。首先设计了一套风机工作信号感知系统,采用高精度传感器获取齿轮箱工作信号;其次,根据齿轮箱工作时的振动信号特性,通过小波包变换方法对工作信号进行特征提取;将这些特征值送到支持向量机(SVM)中进行训练和分类,可以实现故障的智能诊断;最后得出分析结果,通过在实验室现有的齿轮箱实验台进行验证,在小样本情况下能达到了97.5%以上的分类精度。In order tomonitor gearbox real-timely, which is the core component of wind turbine, amethod is put forward. This method is based on some vibration signals that arecaused by different parts of gearbox at work and common faults of gearbox.Firstly, a gearbox working signal acquiring system is created. It uses highprecision sensors to acquire signals when the wind turbine gearbox is working.Secondly, according to features of vibration signals of gearbox at work, usingwavelet packet transform method can extract characteristics from workingsignals. Sending those data to Support Vector Machine (SVM), the system canimplement intelligent fault diagnosis. At last, through experiments underlaboratory condition, this method can reach more than 97.5% classificationaccuracy in the case of small sample.

关 键 词：风力发电机 齿轮箱 小波包 SVM 故障诊断 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]