基于气路参数融合的涡扇发动机性能退化预测

针对单参数驱动的涡扇发动机性能退化预测精度不高的问题，提出了一种基于气路参数融合的涡扇发动机性能退化预测的方法。通过监测发动机性能退化过程中多源参数，采用专家经验和核主成分分析相结合的方法，进行发动机性能参数的选择和融合，从而构建健康参数。基于非线性Wiener过程构建涡扇发动机退化模型，采用极大似然方法求得发动机退化模型的离线参数估计值；由于不同发动机性能退化的差异性，基于贝叶斯更新理念对随机参数进行实时更新，可以实现对单台发动机的性能退化实时预测。通过实例验证，采用此方法在预测末端方均根误差为0.028 3，整体预测精度提升了54.5%，可以辅助指导维修决策。 

Turbofan engine performance degradation prediction based on gas path parameter fusion

In view of the problem of low accuracy in predicting performance degradation of turbofan engine driven by a single parameter，a turbofan engine performance degradation prediction method based on gas path parameter fusion was proposed.By monitoring multi-source parameters in the process of engine performance degradation，a combination of expert experience and nuclear principal component analysis was used to select and integrate engine performance indicators to construct health parameters.The turbofan engine degradation model was constructed based on the nonlinear Wiener process，and the maximum likelihood method was used to obtain the offline parameter estimates of the engine degradation model secondly，due to the difference of performance degradation of different engines，real-time updating of random parameters based on Bayesian updating concept can realize real-time prediction of performance degradation of single engine.Finally，through the verification of examples，the root mean square error at the end of prediction using this method was 0.028 3，and the overall prediction accuracy was improved by 54.5%，which can assist in guiding maintenance decision-making.