On the spatial prediction of the turbulent flow behind an array of cylinders via echo state networks

被引:0
|
作者
Ghazijahani, M. Sharifi [1 ]
Cierpka, C. [1 ]
机构
[1] Tech Univ Ilmenau, Inst Thermodynam & Fluid Mech, D-98684 Ilmenau, Germany
关键词
Echo state networks; Spatial prediction; Turbulence; Wake flows; LAMINAR-FLOW; RECONSTRUCTION;
D O I
10.1016/j.engappai.2025.110079
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This study investigates the capabilities of the spatial prediction of the turbulent flow behind an array of seven cylinders via Echo State Networks (ESN). The flow is dominated by five distinct unsteady vortex streets, each with its own characteristics, that interact with each other. The goal is to reconstruct the entire section of the flow field by training the ESN to predict the values based on the available low resolution experimental data from the rest of the flow field. Four scenarios - forward, backward, vertical, and central - of data availability to the ESN have been explored. The ESN is capable of reconstructing realistic flow fields in terms of the shape of the packs of up and downward vortices alongside their correct arrangements behind each cylinder. Furthermore, the variable phase synchronization of the neighboring cylinders has been well preserved. Nevertheless, there are preferable regions that result in better predictions of the rest of the flow when they are available to the network. Prediction of the downstream region is more challenging due to the increased entropy of the flow as the vortices move downstream. Furthermore, the most compact area of the information seems to be the central vortex street, as it produces better predictions with the lowest amount of input grid points. The spatial predictions remain stable over time, unaffected by divergence, and are largely insensitive to variations in network hyperparameters, except incases of very small leaking rates.
引用
收藏
页数:11
相关论文
共 50 条
  • [21] Design of polynomial echo state networks for time series prediction
    Yang, Cuili
    Qiao, Junfei
    Han, Honggui
    Wang, Lei
    NEUROCOMPUTING, 2018, 290 : 148 - 160
  • [22] Time Series Prediction with Evolved, Composite Echo State Networks
    Webb, Russell Y.
    SIMULATED EVOLUTION AND LEARNING, PROCEEDINGS, 2008, 5361 : 555 - 564
  • [23] MODULAR ECHO STATE NEURAL NETWORKS IN TIME SERIES PREDICTION
    Babinec, Stefan
    Pospichal, Jiri
    COMPUTING AND INFORMATICS, 2011, 30 (02) : 321 - 334
  • [24] Time series prediction using deep echo state networks
    Taehwan Kim
    Brian R. King
    Neural Computing and Applications, 2020, 32 : 17769 - 17787
  • [25] Analyzing the state space property of echo state networks for chaotic system prediction
    Xi, JH
    Shi, ZW
    Han, M
    Proceedings of the International Joint Conference on Neural Networks (IJCNN), Vols 1-5, 2005, : 1412 - 1417
  • [26] Spatial resolution effects on unsteady turbulent flow computations for the rectangular cylinders byκ-ε turbulence models
    Sangsan Lee
    KSME International Journal, 1997, 11 : 339 - 347
  • [27] Optimal Training of Echo State Networks via Scenario Optimization
    Armenio, Luca Bugliari
    Fagiano, Lorenzo
    Terzi, Enrico
    Farina, Marcello
    Scattolini, Riccardo
    IFAC PAPERSONLINE, 2020, 53 (02): : 5183 - 5188
  • [28] Modeling reward functions for incomplete state representations via echo state networks
    Bush, K
    Anderson, C
    Proceedings of the International Joint Conference on Neural Networks (IJCNN), Vols 1-5, 2005, : 2995 - 3000
  • [29] CONTROL OF JET IMPINGEMENT HEAT-TRANSFER BY A WAKE FLOW BEHIND AN ARRAY OF CIRCULAR-CYLINDERS
    KATAOKA, K
    HAMANO, S
    ONISHI, A
    KAWASAKI, H
    LI, GY
    JOURNAL OF CHEMICAL ENGINEERING OF JAPAN, 1992, 25 (01) : 39 - 43
  • [30] Short-term traffic flow prediction based on secondary hybrid decomposition and deep echo state networks
    Hu, Guojing
    Whalin, Robert W.
    Kwembe, Tor A.
    Lu, Weike
    PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS, 2023, 632