基于稀疏自动编码器的发动机机载模型建模方法研究

推进技术 ›› 2017, Vol. 38 ›› Issue (6): 1209-1217.

基于稀疏自动编码器的发动机机载模型建模方法研究

李永进¹,贾爽龙²,张海波¹,张天宏¹

南京航空航天大学能源与动力学院，江苏省航空动力系统重点实验室，江苏南京 210016,中国航空发动机集团公司航空动力控制系统研究所，江苏无锡 214063,南京航空航天大学能源与动力学院，江苏省航空动力系统重点实验室，江苏南京 210016,南京航空航天大学能源与动力学院，江苏省航空动力系统重点实验室，江苏南京 210016

发布日期:2021-08-15
作者简介:李永进，男，博士生，研究领域为航空发动机控制。
基金资助:
国家自然科学基金(51576096)；江苏省“青蓝工程”、“333”人才工程。

Research on Modeling Method of On-Board Engine Model Based on Sparse Auto-Encoder

Nanjing?University?of?Aeronautics?and?Astronautics，Jiangsu?Province?Key?Laboratory?of Aerospace?Power?System，Nanjing?210016，China,Aero Engine Control System Institute，Aero Engine Corporation of China，Wuxi 214063，China,Nanjing?University?of?Aeronautics?and?Astronautics，Jiangsu?Province?Key?Laboratory?of Aerospace?Power?System，Nanjing?210016，China and Nanjing?University?of?Aeronautics?and?Astronautics，Jiangsu?Province?Key?Laboratory?of Aerospace?Power?System，Nanjing?210016，China

Published:2021-08-15

摘要/Abstract

摘要： 为解决分段线性化机载模型精度不足的问题，提出并设计了基于稀疏自动编码器的大包线、具有10输入11输出的发动机机载自适应模型，该模型由稳态、动态两部分组合而成。首先基于一种新的相似准则进行建模所需样本数据的压缩，在保留主要信息的同时，大大降低了数据量及采样时间。用BP算法对简化后的样本数据进行了机载模型稳态部分的建模。针对机载模型动态部分所需样本数据量巨大、BP算法难以训练的问题，建立了基于稀疏自动编码器的动态机载模型。引入准稳态判断逻辑，在动态过程使用稀疏自动编码器的动态机载模型，在稳态过程使用基于BP算法的稳态机载模型。仿真结果表明，所建立的发动机机载模型具有优良的动稳态精度，且实时性好、存储量小，其中动态精度小于1%，稳态精度小于0.6%，一次模型计算时间不大于1ms，模型存储量不大于100kB。

关键词: 机载发动机自适应模型；智能发动机控制；BP算法；稀疏自动编码器；采样数据

Abstract: In order to solve the problem of the low accuracy of the piecewise linear model in the development of the on-board engine model，based on the sparse auto-encoder，an adaptive on-board engine model with 10 inputs 11 outputs for the large envelope is proposed and designed，the model consists of steady and dynamic two parts. In the first place，a new similarity criterion is needed to compress the sample data，which can reduce the amount of data and the sampling time while retaining the main information. Steady on-board engine modeling work is completed by the BP algorithm with the simplified training data. In view of the huge amount of data needed in dynamic modeling，the BP algorithm is difficult to train. Dynamic on-board model is established based on the sparse auto-encoder. By the introduction of quasi steady state judgment logic，in the dynamic process the dynamic on-board model based on the sparse auto-encoder is used，while in the steady state process the steady on-board model based on the BP algorithm is used. Simulation results show that the on-board model obtained has excellent dynamic and steady state accuracy，good real-time performance and a small amount of storage. The dynamic accuracy is within 1%，the steady accuracy is within 0.6%，model computation time is within 1ms once with the storage capacity no more than 100kB.

Key words: Adaptive on-board engine model；Intelligent engine control；BP algorithm；Sparse auto-encoder；Sampled data

李永进,贾爽龙,张海波,张天宏. 基于稀疏自动编码器的发动机机载模型建模方法研究[J]. 推进技术, 2017, 38(6): 1209-1217.

LI Yong-jin¹,JIA Shuang-long²,ZHANG Hai-bo¹,ZHANG Tian-hong¹. Research on Modeling Method of On-Board Engine Model Based on Sparse Auto-Encoder[J]. Journal of Propulsion Technology, 2017, 38(6): 1209-1217.

参考文献

[1] Skira C A, Agnello M. Control System for the Next Century’s Fighter Engines[J]. Journal of Engineering for Gas Turbines and Power, 1992, 114(4): 749-754.
[2] Garg S. NASA Glenn Research in Controls and Diagnostics for Intelligent Aerospace Propulsion Systems[R]. NASA/TM-2005-214036.
[3] Litt J S, Simon D L, Garg S, et al. A Survey of Intelligent Control and Health Management Technologies for Aircraft Propulsion Systems[R]. NASA/TM-2005-213622.
[4] Lietzau K, Kreiner A. Model Based Control Concepts for Jet Engines[C]. Louisiana: Proceedings of ASME Turbo Expo, 2001.
[5] Troha W, Kerner K, Butler G. Development Status of the U.S.Army Small Heavy Fuel Engine (SHFE) VAATE Program [C]. USA: AHS International 63rd Annual Forum Proceedings-American Helicopter Society, 2007.
[6] 方昌德. 航空发动机的发展前景[J]. 航空发动机, 2004, 30(1): 1-5.
[7] Garg S. Introduction to Advanced Engine Control Concepts[C]. Oklahoma City: Fundamentals of Aircraft Engine Control Design Course, 2007.
[8] Garg S. Aircraft Turbine Engine Control Research at NASA Glenn Research Center[R]. NASA/TM 2013-217821.
[9] Brotherton T, Volponi A, Luppold R, et al. eSTORM: Enhanced Self-Tuning on-Board Real-Time Engine Model[C]. Montana: Proceedings of IEEE Aerospace Conference, 2003.
[10] Vloponi A. Enhanced Self-Tuning on-Board Real-Time Model (eSTORM) for Aircraft Engine Performance Health Tracking [R]. NASA/CR 2008-215272.
[11] Simon D L. An Integrated Architecture for Onboard Aircraft Engine Performance Trend Monitoring and Gas Path Fault Diagnostics[R]. NASA TM 2010-216358.
[12] Armstrong J B, Simon D L. Implementation of an Integrated On-Board Aircraft Engine Diagnostic Architecture[R]. AIAA 2011-5859.
[13] Simon D L, Rinehart A W. A Model-Based Anomaly Detection Approach for Analyzing Streaming Aircraft Engine Measurement Data[R]. ASME GT2014-27172.
[14] Rinehart A W, Simon D L. An Integrated Architecture for Aircraft Engine Performance Monitoring and Fault Diagnostics: Engine Test Results [R]. AIAA 2014-3924.
[15] May R D, Csank J, Lavelle T M, et al. A High-Fidelity Simulation of a Generic Commercial Aircraft Engine and Controller [R]. AIAA 2010-6630.
[16] Hunter G W, Lekki J D, Simon D L. Development and Testing of Propulsion Health Management [C]. Bangalore India: Workshop on Integrated Vehicle Health Management and Aviation Safety, 2012.
[17] 鲁峰, 黄金泉, 吕怡秋, 等. 基于非线性自适应滤波的发动机气路部件健康诊断方法[J]. 航空学报, 2013, 34(11): 2529-2538.
[18] 王海泉, 欧阳玲, 黄杰. 涡扇发动机机载自适应模型及其性能蜕化估计[J]. 计算机仿真, 2012, 29(10): 76-80.
[19] WANG Jian-kang, ZHANG Hai-bo, SUN Jian-guo, et al. Application of Hybrid Aero-Engine Model for Integrated Flight/Propulsion Optimal Control[J]. Transactions of Nanjing University of Aeronautics & Astronautics, 2012, 29(1): 16-24.
[20] 任新宇, 杨育武, 樊思齐. 推进系统综合性能寻优控制研究[J]. 推进技术, 2010, 31(1): 61-64. (REN Xin-yu, YANG Yu-wu, FAN Si-qi. Study on Propulsion System Overall Performance Seeking Control Algorithm[J]. Journal of Propulsion Technology, 2010, 31(1): 61-64.)
[21] Hinton G E, Osindero S, The Y. A Fast Learning Algorithm for Deep Belief Nets[J]. Neural Computation, 2006, 18(7).
[22] 李永进, 张海波, 张天宏. 一种考虑非线性余项的机载发动机自适应模型建立及其在寻优控制中的应用[J]. 推进技术, 2016, 37(1): 172-180. (LI Yong-jin, ZHANG Hai-bo, ZHANG Tian-hong. Establishment and Application in Performance Seeking Control of an On-Board Adaptive Aero-Engine Model Considering Nonlinear Remainders[J]. Journal of Propulsion Technology, 2016, 37(1): 172-180.)
[23] Bengio Y, Lamblin P, Popovici D, et al. Greedy Layer-Wise Training of Deep Networks[C]. Whistler: Proceedings of the 20th Annual Conference on Neural Information Processing System, 2006: 153-160.
[24] Hinton G E, Neal R M. The "Wake-Sleep" Algorithm for Unsupervised Neural Networks[J]. Science, 1995, 268(5214): 1158-1161.　　　　　　　　　　　　　*　收稿日期:2016-01-25；修订日期:2016-03-08。基金项目:国家自然科学基金(51576096)；江苏省“青蓝工程”、“333”人才工程。作者简介:李永进,男,博士生,研究领域为航空发动机控制。E-mail: bluedragonz@126.com(编辑:梅瑛)