Gear Transmission Fault Tree Analysis Based on Bayesian Network

As a kind of important mechanical drive forms, the reliability level of gear transmission directly haves an effect on the performance of mechanical products. At present, the analysis method of main gear transmission failure is the fault tree analysis, which has several limits in describing multi-state events. Bayesian network which is very suitable to express multi-state events and uncertain logical relationships has been successfully applied to fault diagnosis field. Therefore, Bayesian network is researched in this paper in order to improve gear transmission fault tree, solve limits of the fault tree out and make the failure analysis results more objective and accurate.