Research on gear-box residual useful life prediction based on stochastic filtering model

Many characteristics were distilled based on the vibration signal from gear-box Accelerated Life Testing. After analyzing the stability and sensitiveness of these characteristics, the factors, such as Clearance Factor, Crest Factor, Shape Factor and Root Mean Square, were selected and used. Then a residual useful life prediction model of gear-box based on stochastic filtering was established, which provided a recursion formula of the probability density function of residual useful life. A maximum likelihood estimation method was designed, which can estimate the unknown parameters. Finally, the validity and practicability of the model are validated by an example.