随机滤波模型在变速箱剩余寿命预测中的应用研究

摘要：基于变速箱加速寿命实验的振动信号，提取信号的多种特征参量．通过分析各特征参量的敏感性和稳定性，选择幅域裕度指标、峰值指标、波形指标和RMS（RootMeanSquare）作为特征参量，建立一种基于随机滤波的变速箱剩余寿命预测模型，给出剩余寿命后验概率密度的递推公式，设计了极大似然估计方法求解模型参数，并经实例验证模型的有效性和实用性．

Research of Stochastic Filtering Model for Gear-Box Residual Useful Life Prediction

Based on the vibration signal from gear-box Accelerated Life Testing, many characteris- tics are picked up. By analyzing the sensitiveness and stability of these characteristics, the Clear- ance Factor,Crest Factor,Shape Factor and Root Mean Square （RMS） are selected and used,and then a gear-box residual useful life prediction model based on stochastic filtering is established. The model provides a recursion formula of the probability density function of residual useful life, and designes a maximum likelihood estimation method to estimate the unknown parameters. Fi- nally, the validity and practicability of the model are validated by an example.