Relevance Vector Machine Based Gear Fault Detection

Recently, condition monitoring of machinery has become global due to the potential advantages to be gained from reduced maintenance costs, improved productivity and increased machine availability. In this paper, a novel fault detection method based on relevance vector machine (RVM) is proposed for gear condition monitoring. Empirical results demonstrated that, using similar training time, the RVM model has shown comparable generalization performance to the popular and state-of-the-art support vector machine (SVM), while the RVM requires dramatically fewer kernel functions and needs much less testing time. The results lead us to believe that the RVM is a more powerful tool for on-line fault detection than the SVM.