基于KPCA 和DBN 的航空发动机排气温度基线模型

为了给航空发动机整体性能的实时监控与健康管理提供技术手段,提出1种基于核主成分分析和深度置信网络相结合的航空发动机排气温度基线模型构建方法。以配装CFM56-7B发动机的飞机在运行过程中各系统产生的快速存取数据作为原始的数据样本,利用核主成分分析进行降维处理,选用高斯函数作为核函数,将降维后的数据作为深度置信网络的输入,建立航空发动机EGT基线模型,通过大量QAR数据验证了模型的有效性和正确性。与传统神经网络建模方法相比,所提出的建模方法不但降低了网络结构的复杂度,同时也提高了模型的精度。

Exhaust Gas Temperature Baseline Model of Aeroengine Based on Kernel Principal Component Analysis and Deep Belief Network

In order to provide technology means for real-time monitoring and health management of the overall performance of aeroengine,a method based on Kernel Principal Component Analysis(KPCA)and Deep Belief Network(DBN)was proposed to construct the baseline model of aeroengine exhaust gas temperature. The Quick Access Recorder (QAR)data generated by the aircraft equipped with CFM56-7B engine in the course of operation was taken as the original data sample. The KPCA was used to reduce the dimension,the Gaussian function was selected as the kernel function,and the reduced data was used as the input of the DBN. The EGT baseline model of aeroengine is established,and the validity and correctness of the model are verified by a large number of QAR data. Compared with the traditional neural network modeling method,the proposed modeling method not only reduces the complexity of the network structure,but also improves the accuracy of the model.