Nonlinear Dynamic Global-Local Retaining Projection-Based Anomaly Detection of Mach Number in Wind Tunnel Flow Field
Keywords:
Wind tunnel, Mach number, Anomaly detection, Nonlinear dynamics, Global and local features, Feature extractionAbstract
Mach number anomaly detection is crucial in wind tunnel experiments. Moreover, there is a need for an anomaly detection model that can detect new samples based solely on historical normal data. This paper proposes an anomaly detection model for wind tunnel flow field Mach numbers using the nonlinear dynamic global-local retaining projection method (NDGLPP). This paper first analyzes the key process variables that affect the Mach number. Then, an iterative GLPP (IGLPP) combined with implicit polynomial expansion is employed to process the selected process variables, thereby extracting both global and local features along with their nonlinear relationships. Next, the corresponding statistical metrics, squared prediction error (SPE) and T², are calculated, and the control limits are determined through the cumulative distribution function (CDF). Finally, the constructed model is applied for anomaly detection. To validate the effectiveness of the model, a comparative analysis is conducted using principal component analysis (PCA), GLPP, and other methods. The experimental results indicate that the NDGLPP-based anomaly detection model for wind tunnel flow field Mach numbers not only achieves higher accuracy in detecting abnormal values but also effectively balances the capture of both global and local features, further confirming its effectiveness and superiority.
References
Ayesha S, Hanif MK, Talib R (2020) Overview and comparative study of dimensionality reduction techniques for highdimensional data. Inform Fusion 59:44-58. https://doi.org/10.1016/j.inffus.2020.01.005
Chen C, Thiele K (2023) Wind tunnel tests on the unsteady galloping of a bridge deck with open cross section in turbulent flow. J Wind Eng Ind Aerod 233. https://doi.org/10.1016/j.jweia.2022.105293
De Gregorio F (2012) Flow field characterization and interactional aerodynamics analysis of a complete helicopter. Aerosp Sci Technol 19(1):19-36. https://doi.org/10.1016/j.ast.2011.11.002
Fu Y, Zhao ZK, Ma XL (2022) Simulation calculation of aerodynamic performance of Mira model based on virtual wind tunnel. Proc Spie 12259. https://doi.org/10.1117/12.2638771
Gunasekaran H, Thangaraj T, Jana T, Kaushik M (2020) Effects of wall ventilation on the shock wave/viscous-layer interactions in a Mach 2.2 intake. Processes 8(2). https://doi.org/10.3390/pr8020208
Guo J, Zhang R, Cui XC, Ma WT, Zhao LP (2023) Multimode wind tunnel flow field system monitoring based on KPLS. Processes 11(1). https://doi.org/10.3390/pr11010178
Issa L, Lakkis I (2014) Reduced-order modeling of low Mach number unsteady microchannel flows. J Fluid Eng-T Asme 136(5). https://doi.org/10.1115/1.4026199
Jiang QC, Yan XF (2015) Nonlinear plant-wide process monitoring using MI-spectral clustering and Bayesian inferencebased multiblock KPCA. J Process Contr 32:38-50. https://doi.org/10.1016/j.jprocont.2015.04.014
Marinho LPB, Avelar AC, Fisch G, Roballo ST, Souza LF, Gielow R, Girardi RM (2009) Studies using wind tunnel to simulate the atmospheric boundary layer at the Alcântara Space Center. J Aerosp Technol Manag 1(1):91-98. Available in https://jatm.com.br/jatm/article/view/11
Lee JM, Yoo CK, Choi SW, Vanrolleghem PA, Lee IB (2004) Nonlinear process monitoring using kernel principal component analysis. Chem Eng Sci 59(1):223-234. https://doi.org/10.1016/j.ces.2003.09.012
Li JJ, Jing MM, Lu K, Zhu L, Shen HT (2019) Locality preserving joint transfer for domain adaptation. IEEE T Image Process 28(12):6103-6115. https://doi.org/10.1109/Tip.2019.2924174
Li SN, Caracoglia L (2020) Wind tunnel experimental variability of aerodynamic loads for wind turbine blades. J Phys Conf Ser 1452. https://doi.org/10.1088/1742-6596/1452/1/012054
Li Z, Tan C, Zhang SM, Dong F (2024) Global-local structure analysis-based status monitoring for oil-gas water flow. IEEE Sens J 24(24):41610-41622. https://doi.org/10.1109/Jsen.2024.3487176
Luo LJ, Sao SY, Mao JF, Tang D (2016) Nonlinear process monitoring based on kernel global-local preserving projections. J Process Contr 38:11-21. https://doi.org/10.1016/j.jprocont.2015.12.005
Qiu HT, Shi MT, Zhu YD, Lee C (2024) Boundary layer transition of hypersonic flow over a delta wing. J Fluid Mech 980. https://doi.org/10.1017/jfm.2024.42
Richter K, Rosemann H (2013) Numerical simulation of wind tunnel wall effects on the transonic flow around an airfoil model. Notes Numer Fluid Me 121:525-532. https://doi.org/10.1007/978-3-642-35680-3_62
Sabnis K, Babinsky H (2019) Nozzle geometry effects on corner boundary layers in supersonic wind tunnels. AIAA J 57(8):3620-3623. https://doi.org/10.2514/1.J058310
Silva MG, Gamarra VO, Koldaev V (2009) Control of Reynolds number in a high speed wind tunnel. J Aerosp Technol Manag 1(1):69-77. Available in https://jatm.com.br/jatm/article/view/9/148
Tong CD, Yan XF (2014) Statistical process monitoring based on a multi-manifold projection algorithm. Chemometr Intell Lab 130:20-28. https://doi.org/10.1016/j.chemolab.2013.09.006
Wang B, Mao ZZ (2019) Integrating Mach number prediction with outlier detection for wind tunnel systems. J Aerospace Eng 32(5). https://doi.org/10.1061/(ASCE)AS.1943-5525.0001041
Wang YG, Jian ST, Wang ZG, Fan R, Han P, Lyu Y (2023) An intelligent algorithm for aerodynamic parameters calibration of wind tunnel experiment at a high angle of attack. Int J Aerospace Eng. https://doi.org/10.1155/2023/3093526
Yao T, Zhou SD, Wang M, Liao QX, Zhu JP (2023) Theoretical calibration and experimental test of hemispherical seven-hole probe under 3-D flow field in wind tunnel. IEEE T Instrum Meas 72. https://doi.org/10.1109/TIM.2023.3301890
Yuan P, Zhao LP (2022) KPLS-Based Mach number prediction for multi-mode wind tunnel flow system. Processes 10(9). https://doi.org/10.3390/pr10091718
Zhan C, Li S, Yang Y (2017) Enhanced fault detection based on ensemble global-local preserving projections with quantitative global-local structure analysis. Ind Eng Chem Res 56(38):10743-10755. https://doi.org/10.1021/acs.iecr.7b01642
Zhang HL, Qi YS, Wang L, Gao XJ, Wang XC (2017) Fault detection and diagnosis of chemical process using enhanced KECA. Chemometr Intell Lab 161:61-69. https://doi.org/10.1016/j.chemolab.2016.12.013
Zhang MG, Ge ZQ, Song ZH, Fu RW (2011) Global-local structure analysis model and its application for fault detection and identification. Ind Eng Chem Res 50(11):6837-6848. https://doi.org/10.1021/ie102564d
Zhao HT, Lai ZH, Chen YD (2019) Global-and-local-structure-based neural network for fault detection. Neural Networks 118:43-53. https://doi.org/10.1016/j.neunet.2019.05.022
Zhao HY, Yuan P, Mao ZZ, Wang BA (2021) Improving identification of wind tunnel systems using clustering-based ensemble outlier detection model. J Aerospace Eng 34(3). https://doi.org/10.1061/(ASCE)AS.1943-5525.0001235
Zhao HY, Yu D, Wang Y, Wang B (2020) Enhancing the prediction of Mach number in wind tunnel with a regression-based outlier detection framework. IEEE Access 9:27668-27677. https://doi.org/10.1109/Access.2021.3058096
Zhao Z, Zhao CC, Pei ZC, Chen WH (2021) Design and implementation of a low-speed wind tunnel system. Paper presented 2021 16th Conference on Industrial Electronics and Applications. IEEE Industrial Electronics Chapter of Singapore, Southwest Jiaotong University, and IEEE Singapore Section; Chengdu, China. https://doi.org/10.1109/Iciea51954.2021.9516145
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Luping Zhao, Shihan Gao
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons — Attribution 4.0 International — CC BY 4.0. Authors are free to Share (copy and redistribute the material in any medium or format) and Adapt (remix, transform, and build upon the material for any purpose, even commercially). JATM allow the authors to retain publishing rights without restrictions.
