关键词: Tikhonov支持向量机, 离散小波, 浮吊齿轮箱, 故障诊断

Abstract: Discrete wavelet Tikhonov support vector machines（SVM） is presented to solve the problem of fault diagnosis for 7500 ton floating crane gear box，which combines discrete wavelet transform and Tikhonov support vector machines.Vibration signal is processed by discrete wavelet transform at the input layer and the detail energy parameters are obtained as fault character vectors．Tikhonov support vector machines is trained according to these character vectors.The trained Tikhonov support vector machines diagnoses the fault．This method is successfully used to diagnose the fault of 7500 Ton Floating Crane Gear Box．Experiment result proves that the fault recognition rate of the method is better than the regular BP network.

Key words: tikhonov support vector machines, discrete wavelet transform, floating crane gear box, fault diagnosis