基于PCA-BPNN的航空发动机使用经济性模型

摘要

从航空发动机的使用经济性角度出发,对其主要使用经济性指标耗油率,提出了基于主成分分析与BP神经网络的航空发动机使用经济性建模方法.利用主成分分析从耗油率相关因素中提取主特征分量,消除样本问的相关性,降低BP网络的规模.接着,利用BP网络高度非线性映射能力构建模型.最后通过实例,验证了该模型的有效性,从而为涡扇发动机的方案设计提供了一种新的选择方法.%Based on the aero-engine operational economy, aiming at its main indictor fuel consumption rate, the modeling approach for aero-engine operational economy based on principal component analysis and BP neural network was given. This approach extracted several the principal features of fuel consumption rate indices by PCA which wiped off the correlation of the data and descended the scale of BP neural network. Then a model was built with the high-effective nonlinear mapping capability of BP network. In the end, the validity was proven by using an instance and it provided a kind of new selection method for the aero-engine scheme design.