考虑脆弱性的航空发动机剩余寿命预测

比例危险模型采用协变量的方式综合多个参数实现对系统的剩余寿命预测,已在航空发动机中得到了应用。然而由于传统比例危险模型是以假设多个研究样本独立同分布,即基本失效率相同为基础的,但对于航空发动机而言,由于运行环境和使用的不同,基本失效率存在差异。研究表明,忽略样本之间的差异会导致估计结果存在严重偏差。为此,本研究将引入用于表达样本之间非独立性的脆弱性概念,针对航空发动机,建立更具普遍性的脆弱性比例危险模型,首次应用于航空发动机的剩余寿命预测。通过实例表明,基于脆弱性比例危险模型得到的预测结果平均误差为4.6%,而基于传统比例危险模型的预测结果平均误差为6.9%,验证了脆弱性比例危险模型在航空发动机剩余寿命预测中的有效性。

Remaining Useful Life Prediction of Aeroengine Considering Frailty

The proportional hazard model uses multiple variables to synthesize multiple parameters to predict the remaining useful life of the system. However, because the traditional proportional hazard model is based on the assumption that multiple research samples are independent and identically distributed, that is, the basic failure rate is the same, but for aero engines, there are differences in basic failure rates due to different operating environments and uses. Studies have shown that ignoring differences between samples can lead to serious bias in the estimates. To this end, this study will introduce the concept of frailty that is used to express the non-independence between samples, and establish a more general vulnerability proportional hazard model for aircraft engines, which will be used for the first time to predict the remaining useful life of aircraft engines. The example shows that the average error of the prediction results based on the proportional hazard model with frailty is 4.6%, while the average error of the prediction results based on the traditional proportional hazard model is 6.9%, which verifies the effectiveness of the proportional hazard model with frailty in the prediction of the remaining useful life of aero engine.