考虑不确定测量的惯性平台剩余寿命自适应预测

针对现有剩余寿命预测方法未考虑测量误差引起的不确定性,且存在预测不确定性大的问题,提出了一种基于Wiener过程的退化过程建模方法,并将测量的不确定性考虑到剩余寿命预测中,推导出了剩余寿命的概率分布。为了实时更新模型参数,利用Kalman滤波算法实时估计Wiener过程中的漂移系数,并利用期望最大化算法实时估计其它相关参数。以某型号惯性平台的退化测量数据为例,进行了实验验证,结果表明,相比其它算法,文中算法能够降低剩余寿命预测的不确定性,提高预测精度。

Adaptive remaining useful life prediction for inertial platform with uncertain measurements

Current methods for the estimation of remaining useful life（RUL）ignore the random measurement errors,and thus lead to the difficulty in reducing the prediction uncertainty.Here we present a Wiener-based degradation process modeling to derive the RUL distribution, while considering the measurement errors.In order to update the model parameters in real-time,Kalman filter is used to predict the drift coefficient in the Wiener process and the Expectation Maximization（EM）algorithm to the other related parameters.The method is applied into a practical inertial platform and the results show that it can improve the estimation accuracy of the RUL comparing to other methods.