Intelligent gear fault detection based on relevance vector machine with variance radial basis function kernel

Detecting machine faults at an early stage is very important. In this study, an intelligent fault detection method based on relevance vector machine (RVM) is proposed for incipient fault detection of gear. First, by combining wavelet packet transform with Fisher criterion, it is able to adaptively find the optimal decomposition level and select the global optimal features from all node energies of full wavelet packet tree. Then, RVM is adopted to train the fault detection model. Improved from Gaussian radial basis function (RBF), a new kernel function denoted variance radial basis function (VRBF) is proposed and used for RVM. Experimental results validate the effectiveness of the proposed method and demonstrate that VRBF_RVM can significantly improve generalization performance over RBF_RVM.