A MACHINE LEARNING APPROACH FOR THE PREDICTION OF TIME-AVERAGED UNSTEADY FLOWS IN TURBOMACHINERY

被引:0
|
作者
Blechschmidt, Dominik [1 ,2 ]
Mimic, Dajan [1 ,2 ]
机构
[1] Leibniz Univ Hannover, Inst Turbomachinery & Fluid Dynam, Hannover, Germany
[2] TU Braunschweig, Cluster Excellence Sustainable & Energy Efficient, Braunschweig, Germany
来源
PROCEEDINGS OF ASME TURBO EXPO 2023: TURBOMACHINERY TECHNICAL CONFERENCE AND EXPOSITION, GT2023, VOL 13D | 2023年
关键词
Unsteady Flow; Axial Compressors; Deep Learning; Machine Learning; NETWORKS;
D O I
暂无
中图分类号
V [航空、航天];
学科分类号
08 ; 0825 ;
摘要
Recent advances in deep learning have led to its increased application in the field of fluid dynamics. By using a data-driven approach instead of a more conventional numerical approach, it is possible to reduce the computational cost of fluid simulations significantly. Especially unsteady computational fluid dynamics (CFD) are known to require a considerable amount of time and resources. Hence, it is common practice in turbomachinery to model the flow as steady by averaging the flow between the rotor and stator in a so-called mixing plane. While this approach is numerically efficient, the full interactions between the rotor and the stator can no longer be predicted accurately due to the averaged upstream flow field. Contrary to this, the time-average of an actual unsteady flow does not contain such a modeling error while still resulting in a flow field that is decoupled from its temporal fluctuations. In this work, we introduce a graph neural network (GNN), which predicts the time-averaged flow field of a rotor stage in an axial compressor based on its steady solution. Because GNNs are able to operate directly on the high-fidelity CFD mesh, we are able to retain the spatial resolution necessary to depict more complex flow behaviour. Consequently, the fidelity of our predictions can compete with conventional high-accuracy flow simulations. Our model is capable of predicting the velocity, pressure, density, and temperature field of a single rotor stage from a 4 1/2 -stage axial compressor test case and shows significant improvements compared to the steady state solution, while also being substantially faster than conventional unsteady CFD simulations.
引用
收藏
页数:18
相关论文
共 50 条
  • [41] COMPARISON OF NUMERICAL METHODS FOR THE PREDICTION OF TIME-AVERAGED FLOW QUANTITIES IN A COOLED MULTISTAGE TURBINE
    Marpu, Ritu P.
    Custer, Chad H.
    Subramanian, Venkataramanan
    Weiss, Jonathan M.
    Hall, Kenneth C.
    ASME TURBO EXPO: TURBINE TECHNICAL CONFERENCE AND EXPOSITION, 2015, VOL 2C, 2015,
  • [42] Time-marching solution of unsteady turbomachinery flows - Calculation of rotating stall and partial admission
    He, L
    LATEST ADVANCES IN THE AERODYNAMICS OF TURBOMACHINERY WITH SPECIAL EMPHASIS UPON UNSTEADY FLOWS, 1996, 1996 (21): : 47 - 65
  • [43] Variational assimilation of sparse time-averaged data for efficient adjoint-based optimization of unsteady RANS simulations
    Plogmann, Justin
    Brenner, Oliver
    Jenny, Patrick
    COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 2024, 427
  • [44] Prediction of force chains for dense granular flows using machine learning approach
    Cheng, Ching-Hung
    Lin, Cheng-Chuan
    PHYSICS OF FLUIDS, 2024, 36 (08)
  • [45] Ensemble Kalman inversion for sparse learning of dynamical systems from time-averaged data
    Schneider, Tapio
    Stuart, Andrew M.
    Wu, Jin-Long
    JOURNAL OF COMPUTATIONAL PHYSICS, 2022, 470
  • [46] A simple approach for estimating short-term peak concentrations with time-averaged models
    Peterson, H
    O'Neill, S
    Lamb, B
    12TH SYMPOSIUM ON BOUNDARY LAYERS AND TURBULENCE, 1997, : 112 - 113
  • [47] A Machine Learning Approach for Prediction of On-time Performance of Flights
    Thiagarajan, Balasubramanian
    Srinivasan, Lakshminarasimhan
    Sharma, Aditya Vikram
    Sreekanthan, Dinesh
    Vijayaraghavan, Vineeth
    2017 IEEE/AIAA 36TH DIGITAL AVIONICS SYSTEMS CONFERENCE (DASC), 2017,
  • [48] A Machine Learning Approach to Waiting Time Prediction in Queueing Scenarios
    Kyritsis, Athanasios, I
    Deriaz, Michel
    2019 SECOND INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE FOR INDUSTRIES (AI4I 2019), 2019, : 17 - 21
  • [49] Time-Averaged Aerodynamics of a High-Speed Low-Pressure Turbine Cascade With Cavity Purge and Unsteady Wakes
    Lopes, G.
    Simonassi, L.
    Lavagnoli, S.
    JOURNAL OF TURBOMACHINERY-TRANSACTIONS OF THE ASME, 2024, 146 (02):
  • [50] Time-averaged velocity and scalar fields of the flow over and around a group of cylinders: a model experiment for canopy flows
    Wangsawijaya, D. D.
    Nicolai, C.
    Ganapathisubramani, B.
    FLOW, 2022, 2