论文部分内容阅读
剩余寿命预测为管理者制定预防性维修策略以保证设备不发生非正常停机提供重要信息。针对设备状态呈现非线性变化以及工程实际中的实时性寿命预测要求,提出一种基于无迹卡尔曼滤波的在线剩余寿命预测方法。该方法首先建立状态空间模型描述设备的退化过程,然后利用监测信号使用无迹卡尔曼滤波估计模型的状态,并通过期望最大化算法估计模型的参数,进而利用当前时刻模型的状态和模型的参数递推求解设备的剩余寿命,最后将该方法应用于直升机主减速器的剩余安全寿命预测。结果表明:该方法能够较准确的在线预测出直升机主减速器的寿命。
Remaining life expectancy provides important information for managers to develop preventive maintenance strategies to ensure that equipment does not shut down abnormally. Aiming at the non-linear change of equipment status and the requirement of real-time life prediction in engineering practice, an online residual life prediction method based on unscented Kalman filter is proposed. Firstly, the state space model is established to describe the degradation process of the equipment. Then the state of the model is estimated by using the unscented Kalman filter and the parameters of the model are estimated by the expectation maximization algorithm. Then the state of the current model and the model parameters Recursively solve the remaining life of the equipment, and finally apply the method to predict the remaining safety life of the helicopter final drive. The results show that this method can predict the life of helicopter main reducer more accurately.