考虑不完全维护影响的随机退化设备剩余寿命预测

摘要

针对寿命周期中存在不完全维护影响的随机退化设备剩余寿命难以预测的问题,提出了一种考虑不完全维护影响的退化建模和剩余寿命预测方法.首先,在Wiener过程理论架下,建立了能够表征不完全维护影响的分阶段退化过程模型,然后从阶段时间服从的逆高斯分布出发,利用逆高斯分布的卷积特性,从理论上推导出存在不完全维护下寿命分布的解析解,并将维护效果的随机性和维护次数的影响传递到寿命分布中.进一步通过时间尺度变换,得到了考虑未来存在不完全维护影响下的剩余寿命分布解析解.通过极大似然估计和最小二乘法对模型未知参数进行了估计.最后将本文方法应用到陀螺仪的实际退化过程中,验证了所提方法的有效性.%The current remaining useful life (RUL) prognostic approaches for stochastic degradation device subject to imperfect preventive maintenance suffer from challenge of low accuracy.A new degradation modeling and RUL prognostic approach are proposed considering the effect of imperfect preventive maintenance.First,based on the theory of Wiener process,the stochastic degradation process with the imperfect maintenance is modeled as multi-stage Wiener process.Second,starting with the inverse Gaussian distributed stage time,the lifetime distribution is derived by the convolution property of the inverse Gaussian distribution.Furthermore,the analytical solution of the RUL distribution can be calculated for the stochastic degradation device subject to imperfect preventive maintenance in future.Finally,for verifying the presented approach,a case study for gyros is provided,and the results indicate that the presented approach of this paper can improve the modeling fitting and the accuracy of the estimated RUL.