论文部分内容阅读
针对如何有效利用设备故障监测过程中获取的有限的直接状态信息和大量的间接状态信息,进行故障预测的问题。首先建立了Gamma退化过程状态空间模型;进而在EM中嵌入PF技术,实现了参数求解,并通过仿真方法验证了估计精度;最后根据UH60直升机主减速器行星架的振动特征和裂纹长度数据,将预测模型应用于行星架裂纹增长过程,对任意时刻的裂纹长度进行了预测。实例表明该预测模型可以以较高的精度估计行星架实际裂纹长度。
Aiming at how to effectively use the limited direct state information and a great deal of indirect state information obtained during the equipment fault monitoring to predict the fault. Firstly, the state space model of Gamma degenerative process was established. Then the PF technology was embedded in the EM, and the parameters were solved. The simulation accuracy was verified by simulation. Finally, according to the vibration characteristics and crack length data of the planetary gearbox of UH60 helicopter, The prediction model is applied to the crack growth of the planet carrier, and the crack length at any moment is predicted. The example shows that the prediction model can estimate the actual crack length of the planet carrier with higher accuracy.