高空台特种调节阀建模方法对比研究

doi:10.13675/j.cnki. tjjs. 180546

推进技术 ›› 2019, Vol. 40 ›› Issue (8): 1895-1901.DOI: 10.13675/j.cnki. tjjs. 180546

高空台特种调节阀建模方法对比研究

1.北京航空航天大学能源与动力工程学院,北京 100191;2.先进航空发动机协同创新中心,北京 100191;3.中国航发四川燃气涡轮研究院高空模拟技术重点实验室，四川绵阳;621703

发布日期:2021-08-15

Comparative Study on Modeling Methods of Special Control Valves in Altitude Simulation Test Facility

1.School of Energy and Power Engineering,Beihang University,Beijing 100191,China;2.Collaborative Innovation Center for Advanced Aero-Engine,Beijing 100191,China;3.Science and Technology on Altitude Simulation Laboratory,AECC Sichuan Gas Turbine Establishment, Mianyang 621703,China

Published:2021-08-15

摘要/Abstract

摘要： 为了获得准确的轮盘式特种调节阀流量特性模型，提高高空舱进口流量预测精度，提出了基于BP神经网络和NARX网络的建模方法。在对调节阀与传感器测点位置分析的基础上，将调节阀和阀后容腔作为整体进行建模。对比研究了流量系数、静态BP神经网络以及基于Gamma Test的动态NARX网络建模方法，并给出了工程中选取建模方法的建议。以试验流量数据为基准，仿真对比了不同阀门开度变化时，各模型输出流量的稳态误差和动态误差。结果表明，BP神经网络方法和NARX网络方法建模精度要优于流量系数法。同时，BP神经网络模型最大稳态误差为0.52kg/s，优于NARX网络模型和流量系数模型。NARX网络模型的最大动态误差为2.04kg/s，相比于BP神经网络模型和流量系数模型，能够更准确地反映流量的动态特性。

关键词: 高空台;特种调节阀;流量特性;NARX网络;动态模型

Abstract: In order to achieve accurate flow characteristics model of the disc type control valve and improve the accuracy of inlet flow prediction of altitude cabin, modeling method based on BP neural network and NARX network is proposed. Based on the analysis of the position of the control valve and sensor point, the control valve and the pipe behind the valve are modeled as a whole. The modeling methods of the flow coefficient, static BP neural network, and dynamic NARX network based on Gamma Test (GT), are compared, and suggestions on selecting in engineering are given. Finally, based on the test flow data, the steady-state error and dynamic error of the flow output of each model are simulated and compared when the valve changes differently. The results show that the modeling precision of the BP neural network method and the NARX network method is better than the flow coefficient method. Meanwhile, the maximum steady-state error of the BP neural network model is 0.52kg/s, which is preferable to the models of NARX network and flow coefficient. Compared with the models of BP neural network and flow coefficient, the NARX network model whose maximum dynamic error is 2.04kg/s, can show the dynamic flow characteristics accurately.

Key words: Altitude simulation test facility;Special control valves;Flow characteristics;NARX network;Dynamic model

. 高空台特种调节阀建模方法对比研究[J]. 推进技术, 2019, 40(8): 1895-1901.

参考文献

[1] 侯敏杰, 刘冬根. 航空发动机高空台的发展与展望[J]. 航空科学技术， 2012， (3): 1-4.
[2] 侯敏杰，陈建民，徐国，等. 先进航空动力崛起的基石——高空台[C]. 深圳: 大型飞机关键技术高层论坛暨中国航空学会2007年学术年会， 2007.
[3] Mcamis R， Miller J， Burdette R， et al. Modeling Fluid Flow Networks[C]. Florida: 32nd Joint Propulsion Conference and Exhibit， 1996.
[4] Garrard D， Vaughn D， Milhoan A， et al. Checkout Testing of the New Basic Process Control System at the Aerodynamic and Propulsion Test Unit[C]. France: International Space Planes and Hypersonic Systems and Technologies Conferences， 2012.
[5] Montgomery P A， Burdette R， Krupp B. A Real-Time Turbine Engine Facility Model and Simulation for Test Operations Modernization and Integration[R]. ASME2000-GT-0576.
[6] Schmidt K， Merten R， Menrath M， et al. Adaptation of the Stuttgart University Altitude Test Facility for BR700 Core Demonstrator Engine Tests[R]. ASME 98-GT-556.
[7] Bolk S. Multivariable Control for the Altitude Testing Facility at the University of Stuttgart[D]. Stuttgart: University of Stuttgart， 2011.
[8] K?cke S. Simulation of an Altitude Test Facility to Analyse the Compressor Surge Preventing Control[D]. Stuttgart: University of Stuttgart， 2010.
[9] 朱美印，裴希同，张松，等. 一种轮盘式特种调节阀流量特性的修正算法[J]. 燃气涡轮试验与研究， 2016， 29(5): 40-45.
[10] 朱美印，张松，但志宏，等. 一种大口径蝶阀流量特性的坐标定位回归算法[J]. 燃气涡轮试验与研究， 2017， 30(4): 39-44.
[11] 裴希同，朱美印，张松，等. 一种特种阀流量特性计算的经验公式迭代方法[J]. 燃气涡轮试验与研究， 2016， 29(5): 35-39.
[12] 缪柯强，朱美印，王曦，等. 基于神经网络和试验数据的高空台特种调节阀特性修正方法研究[C]. 绵阳:航空发动机试验与测试工程技术论坛， 2018.
[13] 裴希同. 某特种流量调节阀的流量特性及控制技术研究[D]. 北京:北京航空航天大学， 2014.
[14] 马卡洛夫. 节流装置计算[M]. 姚文华，译. 北京:冶金工业出版社， 1957.
[15] Asgari H. Gas Turbines Modeling， Simulation， and Control Using Artifcial Neural Networks[M]. London: CRC Press， 2016.
[16] Chang F J， Chen P A， Lu Y R， et al. Real-Time Multi-Step-Ahead Water Level Forecasting by Recurrent Neural Networks for Urban Flood Control[J]. Journal of Hydrology， 2014， 517(1): 836-846.
[17] Stefánsson A， Kon?ar N， Jones A J. A Note on the Gamma Test[J]. Neural Computing & Applications， 1997， 5(3): 131-133.
[18] Evans D， Jones A J. A Proof of the Gamma Test[J]. Proceedings Mathematical Physical & Engineering Sciences， 2002， 458(2027): 2759-2799.
[19] Noori R， Karbassi A R， Sabahi M S. Evaluation of PCA and Gamma Test Techniques on ANN Operation for Weekly Solid Waste Prediction[J]. Journal of Environmental Management， 2010， 91(3): 767-771.
[20] Chang F J， Chen P A， Liu C W， et al. Regional Estimation of Groundwater Arsenic Concentrations Through Systematical Dynamic-Neural Modeling[J]. Journal of Hydrology， 2013， 499(9): 265-274.
[21] Goodfellow I， Bengio Y， Courville A. Deep Learning[M]. Massachusetts: MIT Press， 2016.

高空台特种调节阀建模方法对比研究

Comparative Study on Modeling Methods of Special Control Valves in Altitude Simulation Test Facility

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价