进气畸变模拟器动态总压畸变指数神经网络预测

张韬; 蔡文祥; 赵伟; 陈宝延; 张扬军

doi:10.13224/j.cnki.jasp.20230042

进气畸变模拟器动态总压畸变指数神经网络预测

doi: 10.13224/j.cnki.jasp.20230042

张韬^{1, 3,},
蔡文祥^2,*, ,,
赵伟¹,
陈宝延¹,
张扬军³

1.
北京动力机械研究所，北京 100074
2.
南京理工大学机械工程学院特种动力技术教育部重点实验室，南京 210094
3.
清华大学车辆与运载学院，北京 100084

详细信息

作者简介:
张韬（1986-），男，高级工程师，博士，研究领域为涡扇发动机气动稳定性。E-mail：383561870@qq.com

通讯作者:
蔡文祥（1978-），男，副教授，博士，研究领域为发动机燃烧。E-mail：caiwx_2005@njust.edu.cn

中图分类号: V231.2
计量
- 文章访问数: 547
- HTML浏览量: 386
- PDF量: 33
- 被引次数: 0
出版历程
- 收稿日期: 2023-01-30
- 网络出版日期: 2025-06-15

Prediction of dynamic total pressure distortion index in the distortion generator based on back-propagation artificial neural network

ZHANG Tao^{1, 3
,},
CAI Wenxiang^{2,*
, ,},
ZHAO Wei¹,
CHEN Baoyan¹,
ZHANG Yangjun³

1.
Beijing Power Machinery Institute，Beijing 100074，China
2.
Key Laboratory of Special Engine Technology，Ministry of Education，School of Mechanical Engineering，Nanjing University of Science and Technology，Nanjing 210094，China
3.
School of Vehicle and Mobility，Tsinghua University，Beijing 100084，China

摘要

摘要:
对进气道总压畸变模拟器稳态流场进行数值分析，所得总压畸变图谱和稳态周向畸变指数与试验结果符合良好，证明了数学模型及方法的可靠性。以数值分析所获得进气道/发动机气动交界面的流场参数为基础，与试验所得紊流度相结合，基于改进的紊流关联模型方程，利用反向传播人工神经网络建立了动态畸变指数的预测方法。经过验证，该动态总压畸变指数预测方法的最大误差为4.19%，可用于指导进气总压畸变模拟试验的相关研究。最后，应用所搭建的神经网络紊流度预测模型对扇形板畸变模拟器的动态畸变图谱进行了预测，为其工程应用提供了重要指导。
- 动态总压畸变 /
- 湍流度 /
- 反向传播人工神经网路 /
- 预测 /
- 压力畸变器
Abstract:
The steady state flow field of the total pressure distortion generator was numerically analyzed. The total pressure distortion pattern and steady circumferential distortion index obtained from CFD were in good agreement with the experiment results, which proved the reliability of the mathematical model and method. Then, on the basis of flow field parameters of the inlet/engine aerodynamic interface obtained by numerical analysis, the prediction method of dynamic distortion index was established by using backpropagation artificial neural network based on the improved turbulence correlation model equation, by combining with the turbulence obtained by experiments. After verification, the maximum error of the predicted dynamic total pressure distortion index was 4.19%. The proposed prediction method can be used to guide the related research of the inlet total pressure distortion simulation experiment. Finally, the established turbulence prediction model was used to predict the dynamic distortion pattern of the distortion generator, thus providing an important guidance for its engineering application.
- dynamic total pressure distortion /
- turbulence correlation /
- back-propagation artificial neural network /
- prediction /
- pressure distortion generator

HTML

图 1 畸变板结构状态

Figure 1. States of sector flashboard

下载: 全尺寸图片幻灯片

图 2 计算对象及网格结构

Figure 2. Computational model and details of grids

下载: 全尺寸图片幻灯片

图 3 总压云图

Figure 3. Total pressure contours

下载: 全尺寸图片幻灯片

图 4 紊流动能云图

Figure 4. Turbulence kinetic energy contours

下载: 全尺寸图片幻灯片

图 5 紊流涡旋频率云图

Figure 5. Turbulence eddy frequency contours

下载: 全尺寸图片幻灯片

图 6 50°总压恢复系数试验测量云图

Figure 6. Experiment total pressure recovery coefficient pattern of 50°

下载: 全尺寸图片幻灯片

图 7 50°总压恢复系数数值模拟云图

Figure 7. Computational total pressure recovery coefficient pattern of 50°

下载: 全尺寸图片幻灯片

图 8 紊流度BP-ANN预测网络

Figure 8. Network of BP-ANN for turbulence prediction

下载: 全尺寸图片幻灯片

图 9 训练集 Case 1～ Case 3紊流度

Figure 9. Turbulence of training Cases 1—Case 3

下载: 全尺寸图片幻灯片

图 10 训练集 Case 1～Case 3动态畸变指数

Figure 10. Dynamic distortion index of training Cases 1—Case 3

下载: 全尺寸图片幻灯片

图 11 紊流度预测验证（Case 4）

Figure 11. Validation of predicted turbulence （Case 4）

下载: 全尺寸图片幻灯片

图 12 动态畸变指数预测验证

Figure 12. Validation of predicted dynamic distortion index

下载: 全尺寸图片幻灯片

图 13 AIP界面Case 1～Case 4紊流度预测图谱

Figure 13. Prediction of AIP turbulence pattern of Cases 1—Case 4

下载: 全尺寸图片幻灯片

图 14 AIP界面紊流度预测图谱随Ma变化

Figure 14. Prediction of AIP turbulence pattern under different Ma

下载: 全尺寸图片幻灯片

图 15 AIP界面紊流度预测图谱随轴向变化

Figure 15. Prediction of AIP turbulence pattern under different axis distance

下载: 全尺寸图片幻灯片

图 16 Case 4动态畸变指数随轴向位置变化

Figure 16. Dynamic distortion index of Case 4 under different axis distance

下载: 全尺寸图片幻灯片

表 1 网格无关性分析

Table 1. Analysis of grid independence

网格数量/万出口平均马赫数

60 0.3875

300 0.3997

600 0.3993

下载: 导出CSV

表 2 畸变器试验条件

Table 2. Experiment conditions of distortion generator

Case 扇形板相位角θ/（°） p_t0/kPa

1 30 115～180

2 40 115～180

3 50 115～180

4 60 115～180

下载: 导出CSV

参考文献(20)

[1]	刘大响, 叶培梁, 胡骏, 等. 航空燃气涡轮发动机稳定性设计与评定技术[M]. 北京: 航空工业出版社, 2004. LIU Daxiang, YE Peiliang, HU Jun, et al. Stability design and evaluation technology of aviation gas turbine engine[M]. Beijing: Aviation Industry Press, 2004. (in Chinese LIU Daxiang, YE Peiliang, HU Jun, et al. Stability design and evaluation technology of aviation gas turbine engine[M]. Beijing: Aviation Industry Press, 2004. (in Chinese)
[2]	李亮, 胡骏, 王志强, 等. 多种形式插板的压气机进气总压畸变实验[J]. 航空动力学报, 2009, 24(4): 925-930. LI Liang, HU Jun, WANG Zhiqiang, et al. Experimental study of inlet total-pressure distortion on four kinds of flat baffles[J]. Journal of Aerospace Power, 2009, 24(4): 925-930. (in Chinese LI Liang, HU Jun, WANG Zhiqiang, et al. Experimental study of inlet total-pressure distortion on four kinds of flat baffles[J]. Journal of Aerospace Power, 2009, 24(4): 925-930. (in Chinese)
[3]	XIA Aiguo, HUANG Xudong, TUO Wei, et al. Experimental study of a controlled variable double-baffle distortion generator engine test rig[J]. Chinese Journal of Aeronautics, 2018, 31(9): 1880-1893. doi: 10.1016/j.cja.2018.06.015
[4]	张韬, 赵伟, 秦德胜, 等. 扇形板压力畸变模拟器特性研究[J]. 推进技术, 2019, 40(9): 2113-2120. ZHANG Tao, ZHAO Wei, QIN Desheng, et al. Research of triangle total pressure distortion generator[J]. Journal of Propulsion Technology, 2019, 40(9): 2113-2120. (in Chinese ZHANG Tao, ZHAO Wei, QIN Desheng, et al. Research of triangle total pressure distortion generator[J]. Journal of Propulsion Technology, 2019, 40(9): 2113-2120. (in Chinese)
[5]	甘甜, 王如根, 李少伟, 等. 插板式静态进气畸变数值模拟与试验验证[J]. 空军工程大学学报(自然科学版), 2013, 14(5): 9-12. GAN Tian, WANG Rugen, LI Shaowei, et al. The numerical simulation and experimental verification of the steady insert board inlet distortion[J]. Journal of Air Force Engineering University (Natural Science Edition), 2013, 14(5): 9-12. (in Chinese GAN Tian, WANG Rugen, LI Shaowei, et al. The numerical simulation and experimental verification of the steady insert board inlet distortion[J]. Journal of Air Force Engineering University (Natural Science Edition), 2013, 14(5): 9-12. (in Chinese)
[6]	王如根, 江勇, 张启发. 对进气道插板试验数据的一点探讨[J]. 航空动力学报, 2002, 17(2): 170-172. WANG Rugen, JIANG Yong, ZHANG Qifa. Analysis of distortion test data by inlet flat baffle[J]. Journal of Aerospace Power, 2002, 17(2): 170-172. (in Chinese doi: 10.3969/j.issn.1000-8055.2002.02.006 WANG Rugen, JIANG Yong, ZHANG Qifa. Analysis of distortion test data by inlet flat baffle[J]. Journal of Aerospace Power, 2002, 17(2): 170-172. (in Chinese) doi: 10.3969/j.issn.1000-8055.2002.02.006
[7]	王勤. 在动态畸变流场中的脉动压力统计特性初探[J]. 航空发动机, 2000, 26(2): 14-20. WANG Qin. Preliminary study on statistical characteristics of fluctuating pressure in dynamic distorted flow field[J]. Aeroengine, 2000, 26(2): 14-20. (in Chinese WANG Qin. Preliminary study on statistical characteristics of fluctuating pressure in dynamic distorted flow field[J]. Aeroengine, 2000, 26(2): 14-20. (in Chinese)
[8]	江勇, 张百灵, 陈世国, 等. 基于实验的发动机插板式进气畸变压力谐振分析[J]. 航空动力学报, 2009, 24(9): 2057-2062. JIANG Yong, ZHANG Bailing, CHEN Shiguo, et al. Test-based syntonic analysis of the aero-engine inserted-board inlet dynamic distortion pressure[J]. Journal of Aerospace Power, 2009, 24(9): 2057-2062. (in Chinese JIANG Yong, ZHANG Bailing, CHEN Shiguo, et al. Test-based syntonic analysis of the aero-engine inserted-board inlet dynamic distortion pressure[J]. Journal of Aerospace Power, 2009, 24(9): 2057-2062. (in Chinese)
[9]	钟亚飞, 马宏伟, 李金原, 等. 航空发动机进气总压畸变地面试验测试技术进展[J]. 航空发动机, 2020, 46(6): 62-77. ZHONG Yafei, MA Hongwei, LI Jinyuan, et al. Technological progress of ground test measurement of aeroengine inlet total pressure distortion[J]. Aeroengine, 2020, 46(6): 62-77. (in Chinese ZHONG Yafei, MA Hongwei, LI Jinyuan, et al. Technological progress of ground test measurement of aeroengine inlet total pressure distortion[J]. Aeroengine, 2020, 46(6): 62-77. (in Chinese)
[10]	钟亚飞, 马宏伟, 郭君德, 等. 航空发动机进气总压畸变地面试验数据处理方法综述[J]. 航空发动机, 2021, 47(1): 72-85. ZHONG Yafei, MA Hongwei, GUO Junde, et al. Review of ground test data processing method of aeroengine inlet total pressure distortion[J]. Aeroengine, 2021, 47(1): 72-85. (in Chinese ZHONG Yafei, MA Hongwei, GUO Junde, et al. Review of ground test data processing method of aeroengine inlet total pressure distortion[J]. Aeroengine, 2021, 47(1): 72-85. (in Chinese)
[11]	孙鹏, 高海洋, 钟兢军, 等. 插板式畸变发生器后非均匀流场结构数值模拟[J]. 推进技术, 2013, 34(2): 173-180. SUN Peng, GAO Haiyang, ZHONG Jingjun, et al. Numerical simulation of non-uniform flow field structure behind classic flat baffle[J]. Journal of Propulsion Technology, 2013, 34(2): 173-180. (in Chinese SUN Peng, GAO Haiyang, ZHONG Jingjun, et al. Numerical simulation of non-uniform flow field structure behind classic flat baffle[J]. Journal of Propulsion Technology, 2013, 34(2): 173-180. (in Chinese)
[12]	甘甜, 王如根, 张杰, 等. 不同湍流模型对插板式进气畸变的数值模拟[J]. 推进技术, 2014, 35(7): 891-896. GAN Tian, WANG Rugen, ZHANG Jie, et al. Numerical simulation of inlet distortion with interceptor with different turbulence models[J]. Journal of Propulsion Technology, 2014, 35(7): 891-896. (in Chinese GAN Tian, WANG Rugen, ZHANG Jie, et al. Numerical simulation of inlet distortion with interceptor with different turbulence models[J]. Journal of Propulsion Technology, 2014, 35(7): 891-896. (in Chinese)
[13]	周游天, 李军, 彭生红, 等. 插板进气畸变与压气机的耦合数值模拟[J]. 航空动力学报, 2017, 32(3): 568-576. ZHOU Youtian, LI Jun, PENG Shenghong, et al. Numerical simulation of flat baffle inlet distortion coupled with compressor[J]. Journal of Aerospace Power, 2017, 32(3): 568-576. (in Chinese ZHOU Youtian, LI Jun, PENG Shenghong, et al. Numerical simulation of flat baffle inlet distortion coupled with compressor[J]. Journal of Aerospace Power, 2017, 32(3): 568-576. (in Chinese)
[14]	LADD J, NORBY W. Dynamic inlet distortion predictions using a CFD/distortion synthesis approach: AIAA 1998-2735[R]. Albuquerque: American Institute of Aeronautics and Astronautics, 1998.
[15]	NORBY W P, LADD J A, YUHAS A J. Dynamic inlet distortion prediction with a combined computational fluid dynamics and distortion synthesis approach[R]. NASA Contract NAS 3-26617, 1996.
[16]	ZHANG Lifen, LIU Zhenxia, QU Jiyun, et al. An improved CFD-synthesis method for predicting inlet dynamic distortion[J]. Chinese Journal of Aeronautics, 2009, 22(5): 467-473. doi: 10.1016/S1000-9361(08)60127-2
[17]	戴敏, 谢椿. 基于模糊加权有色网和BP神经网络的飞机发动机故障诊断[J]. 科学技术与工程, 2012, 12(35): 9552-9555, 9561. DAI Min, XIE Chun. The fault diagnosis for aircraft generator based on fuzzy colored Petri net and BP neural network[J]. Science Technology and Engineering, 2012, 12(35): 9552-9555, 9561. (in Chinese DAI Min, XIE Chun. The fault diagnosis for aircraft generator based on fuzzy colored Petri net and BP neural network[J]. Science Technology and Engineering, 2012, 12(35): 9552-9555, 9561. (in Chinese)
[18]	WANG Xiangmin, WANG Jun, PRIVAULT M. Artificial intelligent fault diagnosis system of complex electronic equipment[J]. Journal of Intelligent & Fuzzy Systems, 35(4): 4141-4151.
[19]	张霞妹, 夏树丹, 解梦涛, 等. 基于BP神经网络的平板叶片阻尼反演方法[J]. 科学技术与工程, 2018, 18(16): 284-288. ZHANG Xiamei, XIA Shudan, XIE Mengtao, et al. Inversion for damping ratio of flat blade based on BP neural network[J]. Science Technology and Engineering, 2018, 18(16): 284-288. (in Chinese ZHANG Xiamei, XIA Shudan, XIE Mengtao, et al. Inversion for damping ratio of flat blade based on BP neural network[J]. Science Technology and Engineering, 2018, 18(16): 284-288. (in Chinese)
[20]	黄鸿鑫, 张会锁, 张帆, 等. 基于BP人工神经网络的聚能射流速度预测方法[J]. 火炮发射与控制学报, 2020, 41(1): 61-65. HUANG Hongxin, ZHANG Huisuo, ZHANG Fan, et al. Prediction method of shaped energy jet velocity based on BP artificial neural network[J]. Journal of Gun Launch & Control, 2020, 41(1): 61-65. (in Chinese HUANG Hongxin, ZHANG Huisuo, ZHANG Fan, et al. Prediction method of shaped energy jet velocity based on BP artificial neural network[J]. Journal of Gun Launch & Control, 2020, 41(1): 61-65. (in Chinese)