考虑随机时变环境影响的设备剩余贮存寿命预测及预测维护研究

Residual storage life (RSL) is an important performance index of the long-term storage equipment. The influences of operation state and external environment have to be considered for accurately predicting the RSL of such equipment. The switching of operation state and external environment often takes on stochastic and time-varying characteristic. For RSL prediction and predictive maintenance of the equipment affected by the stochastic and time-varying environment, the existing researches cannot solve these problems well. Considering the fact that the degradation process of equipment is influenced directly by the stochastic and time-varying environment, the method based on degradation modeling is utilized for researching the above problems. In theory, (1) the modeling of degradation process and the RSL prediction method considering the influence of stochastic and time-varying environment are studied. (2) The modeling of hidden degradation process and the RSL prediction method under stochastic and time-varying environment are investigated. (3) The model of multi-criteria decision function based on the RSL information will be constructed and the optimal maintenance time will be determined. In engineering, Under the background of the gyro in the missile, the abovementioned theory researches are verified by applying to the gyro. This project aims to investigate some crucial problems of RSL prediction and predictive maintenance, and provide decision basis to lifetime extension and optimal maintenance of the weapon.

剩余贮存寿命是衡量长贮存设备的重要性能指标。为准确预测此类设备的剩余贮存寿命并据此制定合理的预测维护策略，设备运行状态与外界环境的影响是必须要考虑的关键问题。运行状态与外界环境的切换往往具有随机时变特性，现有研究无法很好地解决随机时变环境影响下的剩余贮存寿命预测和预测维护问题。考虑到设备的退化过程会受到随机时变环境的直接影响，本项目拟采用退化过程建模的方法，对此类问题进行研究。在理论上，（1）研究考虑随机时变环境影响的退化过程建模及剩余贮存寿命预测方法；（2）研究随机时变环境下的隐含退化过程建模及剩余贮存寿命预测方法；（3）构建基于剩余贮存寿命信息的多标准预测维护模型，确定最优维护时机。在工程上，以导弹武器系统的惯性部件陀螺仪为研究对象，结合实验室测试数据和部队实际监测数据验证所提方法的效果。项目旨在研究剩余贮存寿命预测和预测维护中的一些关键问题，为武器装备的延寿和最优维护提供决策依据。

剩余贮存寿命是衡量导弹战术技术性能的重要指标。准确预测导弹的剩余贮存寿命对于提高导弹的运行可靠性、保证各项任务的圆满完成具有非常重要的作用。为准确预测此类设备的剩余贮存寿命，制定合理的维护检测策略，设备运行状态与外界环境的影响是必须要考虑的关键问题。本项目对随机时变环境影响下基于退化过程建模的剩余贮存寿命预测方法进行了研究，首先构建了基于半随机滤波的剩余贮存寿命预测模型，通过仿真实验验证了该模型的效果。进一步分析了贮存期退化和环境因素对退化过程的影响，提出了一种考虑不确定性影响的剩余贮存寿命预测方法，推导出剩余贮存寿命分布的解析式。初步构建了随机时变环境下基于隐含非线性退化过程的剩余贮存寿命预测模型并进行了相关理论分析，得到了剩余贮存寿命的概率分布。针对导弹日常维护检测的实际情况，构建了基于重要度分析的部件检测方式决策模型。研究工作对导弹武器装备的延寿和最优维护具有一定的借鉴意义。