Research of Gear Box Fault Diagnosis Based on PSO-SVM
-
摘要: 针对目前齿轮箱故障诊断存在的检测难度大、主观性强、准确性不高等问题,提出了一种基于粒子群算法和支持向量机的故障诊断方法。运用时域频域分析法对振动信号进行分析获取特征值,利用支持向量机(SVM)技术对齿轮箱特征参数进行模式识别和故障分类,并引入粒子群算法(PSO)用于优化支持向量机参数,建立了齿轮箱典型故障诊断模型。实验结果表明:该方法可以对齿轮箱不同故障类型进行准确的分类,有效的提高了齿轮箱故障诊断的可靠性。Abstract: Aiming at the problems existing in gear box fault diagnosis such that difficult to detecting,strong subjectivity and low accuracy,a fault diagnosis method based on particle swarm optimization(PSO) algorithm and support vector machine(SVM) is proposed. In this method the time domain analysis and frequency domain analysis are conducted to get the characteristic value of vibration signals,and the SVM is used for pattern recognition and fault classification of the characteristic parameters of gearbox,and PSO is introduced in the optimization of SVM parameters. A typical gearbox fault diagnosis model is established and the experimental results show that the method can classify different gear box fault type accurately and the reliability of the gear box fault diagnosis is effectively improved.
-
Key words:
- diagnosis /
- efficiency /
- experiments /
- failure analysis
-
[1] Ma R，Chen Y S. Research on the dynamic mechanism ofthe gear system with local crack and spalling failure [J].Engineering Failure Analysis，2012，26(19):12-20 [2] 李辉，郑海起，杨绍普. 齿轮箱启动过程故障诊断[J]. 振动、 测试与诊断，2009，29(2):167-170Li H，Zheng H Q，Yang S P. During startup of the gearbox fault diagnosis[J]. Vibration，Test and Diagnosis，2009，29 (2):167-170 (in Chinese) [3] 李有荣，曾法力，吕勇，等. 小波包分析在齿轮箱故障诊断中的应用[J]. 振动与冲击，2005，24(5):101-103Li Y R，Zeng F L，Lü Y，et al. Wavelet packetanalysis in gearbox fault diagnosis application [J].Journal of Vibration and Shock，2005，24(5):101-103(in Chinese) [4] 杨通强，郑海起，唐力伟，等. 基于角平均域和倒阶次谱分析的齿轮箱故障诊断[J]. 机械科学与技术，2006，25(4):452-454Yang T Q，Zheng H Q，Tang L W，et al. Based on theaverage angle domain and the order time spectrumanalysis of gearbox fault diagnosis[J]. MechanicalScience and Technology for Aerospace Engineering，2006，25(4):452-454 (in Chinese) [5] Vapnik V N. 统计学习理论的本质[M]. 张学工，译．北京:清华大学出版社，2000Vapnik V. N. The nature of statistical learning theory[M]. Zhang X G，Translator. Beijing: TsinghuaUniversity Press，2000 (in Chinese) [6] 邓乃杨，田英杰. 数据挖掘中的新方法一支持向量机[M]. 北京:科学出版社，N Y，Tian Y J. New method in data mining，support vector machine[M]. Beijing: Science Press，2004 (in Chinese) [7] 张树团，张晓斌. 基于粒子群算法和支持向量机的故障诊断研究[J]. 计算机测量与控制，2008，(11):1573-1575Zhang S T，Zhang X B. Machine fault diagnosisresearch based on particle swarm optimization algorithmand support vector machine fault diagnosis research[J]. Computer Measurement and Control，2008，(11):1573-1575 (in Chinese) [8] 任丽华，王昕，赵毅斌. 基于支持向量机的液压泵故障判断方法研究[J]. 煤矿机械，2009，(3):187-189Ren L H，Wang X，Zhao Y B. Hydraulic pump faultdiagnosis method based on support vector machine(SVM) study [J]. Journal of Coal Mine Machinery，2009，(3):187-189 (in Chinese) [9] 冯志敏，王颖. 智能化故障诊断系统的研究与开发[J].中国航海，2003，(1):13-17Feng Z M，Wang Y. The research and development ofthe intelligent fault diagnosis system [J]. Navigation ofChina，2003，(1) : 13 -17 (in Chinese) [10] 王新峰，邱静，刘冠军. 基于支持向量机的机械故障特征选择方法研究[J]. 机械科学与技术，2005，24(9):1122-1125Wang X F，Qiu J，Liu G J. Mechanical fault featureselection method based on support vector machinestudy [J]. Mechanical Science and Technology forAerospace Engineering，2005，24(9):1122-1125 (in Chinese) -
点击查看大图
计量
- 文章访问数: 153
- HTML全文浏览量: 39
- PDF下载量: 4
- 被引次数: 0
下载:
