Aeroengine performance prognosis based on feature selection and extreme learning machine

Abstract

Abstract:

For predicting exhaust gas temperature of aeroengine,a modeling method is performed. Original database is derived from real-time monitoring data of engine monitoring system of PW4000 engine. The mean impact value (MIV)is as taken evaluation criteria for feature selection,screening 8 features as input variables. Extreme learning machine (ELM)algorithm for single-hidden layer feedforward neural network is applied for modeling,After the neural network is tested by monitoring data of the same engine,and model's extended performance is tested by data of that same engine after water wash. Result shows that training time of ELMis much less than that of back propagation algorithm,which is efficient for repeated training for reducing error,and the whole modeling method has satisfactory scalability.

Key words: aeroengine, real-time monitoring data, mean impact value, feature selection, extreme learning machine, neural network

CLC Number:

V235.13

XU Jianxin,YUE Minqi. Aeroengine performance prognosis based on feature selection and extreme learning machine[J]. .

References

[1] FREDERICK D K，DECASTRO J A，LIU Y，et al. User’s Guide for the CommercialModularAero-PropulsionSystemSimulation（C-MAPSS）[R]. NASA/TM—2012-217432，2012.

[2] MINK G，BEHBAHANI A. The AFRL ICF Generic Gas Turbine Engine Model[C]//41st AIAA/ASME/SAE/SAEE Joint Propulsion Conference &Exhibit，Tucson，Arizona，July10-13，2005.

[3] ESSAM K，AHMED H，MAHMOUD F，et al. Effect of Bleed Air on Performance of Turbofan Engines[C]//51st AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition，Grapevine（Dallas/Ft. Worth Region），Texas，January 07-10，2013.

[4] 钱德峰. 大涵道比民用涡扇发动机部件级建模技术研究[D]. 南京：南京航空航天大学，2010.

[5] 李国权. 航空发动机滑油系统的现状及未来发展[J]. 航空发动机，2011（6）：49-52.

[6] 郝英，孙健国，白杰. 航空燃气涡轮发动机气路故障诊断现状与展望[J].航空动力学报，2003，18（6）：753-760.

[7] SIMON D L. An Integrated Architecture for On-Board Aircraft Engine Performance Trend Monitoring and Gas Path Fault Diagnostics[R]. 57th Joint Army-Navy-NASA-Air Force （JANNAF）Propulsion Meeting Colorado Springs，Colorado，NASA TM2010-216358，2010.

[8] 耿宏，揭俊.基于QAR数据的飞机巡航段燃油流量回归模型[J].航空发动机，2008（4）：46-50.

[9] 高峰，曲建岭，周玉平，等. 基于飞参数据和支持向量机的航空发动机建模[J]. 测控技术，2013（9）：52-54.

[10] HAYKINSS，HAYKINSS.NeuralNetworksandLearningMachines[M].Upper Saddle River：Pearson Education，2009.

[11] VOLPONI A J，DEPOLD H，GANGULI R，et al. The use of Kalman filter and neural network methodologies in gas turbine performance diagnostics：a comparative study[J]. Journal of Engineering for Gas Turbines and Power，2003，125（4）：917-924.

[12] KOBAYASHI T，SIMON D L. Hybrid neural-network genetic-algorithm technique for aircraft engine performance diagnostics[J]. Journal of Propulsion and Power，2005，21（4）：751-758.

[13] 单晓明，宋云峰，黄金泉，等. 基于神经网络和模糊逻辑的航空发动机状态监视[J].航空动力学报，2009，24（10）：2356-2361.

[14] DASHM，LIU H. Feature selection for classification[J]. Intelligent data analysis，1997，1（3）：131-156.

[15] 刘永建. 基于改进神经网络的民机发动机故障诊断与性能预测研究[D].南京：南京航空航天大学，2012.

[16] HUANG G B，ZHU Q Y，SIEW C K. Extreme Learning Machine：A New Learning Scheme of Feedforward Neural Networks[C]//Proceedings. 2004 IEEE International Joint Conference on. IEEE，2004，2：985-990.

[17] HUANG G，ZHU Q，SIEWC. Extreme learning machine：theory and applications[J]. Neurocomputing，2006，70（1）：489-501.

[18] WANG D，HUANG G B. Protein Sequence Classification Using Extreme Learning Machine[C]//Proceedings. 2005 IEEE International Joint Conference on. IEEE，2005，3：1406-1411.

[19] 周百政，曹惠玲. 基于EHM 软件思路的QAR数据处理[J]. 航空维修与工程，2010（4）：60-62.