scholarly journals Integration of Machine Learning and Computational Fluid Dynamics to Develop Turbulence Models for Improved Turbine Wake Mixing Prediction

2021 ◽  
Author(s):  
Harshal D Akolekar ◽  
Yaomin Zhao ◽  
Richard D Sandberg ◽  
Roberto Pacciani
Keyword(s):  
Machine Learning ◽  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Turbulence Models ◽  
Turbine Wake

INTEGRATION OF MACHINE LEARNING AND COMPUTATIONAL FLUID DYNAMICS TO DEVELOP TURBULENCE MODELS FOR IMPROVED LOW-PRESSURE TURBINE WAKE MIXING PREDICTION

2021 ◽  
pp. 1-12
Author(s):  
Harshal D Akolekar ◽  
Yaomin Zhao ◽  
Richard Sandberg ◽  
Roberto Pacciani
Keyword(s):  
Machine Learning ◽  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Model Development ◽  
Learning Approach ◽  
Near Wake ◽  
Low Pressure Turbine ◽  
Machine Learning Approach ◽  
Turbulence Closures ◽  
Turbine Wake

Abstract This paper presents the development of accurate turbulence closures for low-pressure turbine (LPT) wake mixing prediction by integrating a machine-learning approach based on gene expression programming (GEP), with Reynolds Averaged Navier-Stokes (RANS) based computational fluid dynamics (CFD). In order to further improve the performance and robustness of GEP-based data-driven closures, the fitness of models is evaluated by running RANS calculations in an integrated way, instead of an algebraic function. Using a canonical turbine wake with inlet conditions prescribed based on high-fidelity data of the T106A cascade, we demonstrate that the ‘CFD-driven’ machine-learning approach produces physically correct non-linear turbulence closures, i.e., predict the right down-stream wake development and maintain an accurate peak wake loss throughout the domain. We then extend our analysis to full turbine blade cases and show that the model development is sensitive to the training region due to the presence of deterministic unsteadiness in the near-wake. Models developed including the near-wake have artificially large diffusion coefficients to over-compensate for the vortex shedding steady RANS cannot capture. In contrast, excluding the near-wake in the model development produces the correct physical model behavior, but predictive accuracy in the near-wake remains unsatisfactory. This can be remedied by using the physically consistent models in unsteady RANS. Overall, the ‘CFD-driven’ models were found to be robust and capture the correct physical wake mixing behavior across different LPT operating conditions and airfoils such as T106C and PakB.

The Effect of Different Turbulence Models on the Emitter Discharge by Using Computational Fluid Dynamics

2013 ◽  
Vol 662 ◽  
pp. 586-590
Author(s):  
Gang Lu ◽  
Qing Song Yan ◽  
Bai Ping Lu ◽  
Shuai Xu ◽  
Kang Li
Keyword(s):  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Flow Field ◽  
Turbulence Models ◽  
Experimental Results ◽  
Turbulent Model ◽  
Simulation Results ◽  
Rsm Model ◽  
Dynamics Method ◽  
Emitter Discharge

Four types of Super Typhoon drip emitter with trapezoidal channel were selected out for the investigation of the flow field of the channel, and the CFD (Computational Fluid Dynamics) method was applied to simulate the micro-field inside the channel. The simulation results showed that the emitter discharge of different turbulent model is 4%-14% bigger than that of the experimental results, the average discharge deviation of κ-ω and RSM model is 5, 4.5 respectively, but the solving efficiency of the κ-ω model is obviously higher than that of the RSM model.

scholarly journals Validasi Model Turbulensi pada Simulasi Numerik Menggunakan Software Fluent dengan Sayap Onera M6

2019 ◽  
Vol 2 (1) ◽  
Author(s):  
Sulistiya Sulistiya ◽  
Alief Sadlie Kasman
Keyword(s):  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Computational Result ◽  
Computational Simulation ◽  
Turbulence Models ◽  
Wind Tunnels ◽  
Direct Testing ◽  
The World ◽  
Computational Fluid Dynamics Cfd ◽  
Cfd Method

AbstractNumerical simulation using Computational Fluid Dynamics (CFD) method is one way of predicting airflow characteristics on the model. This method is widely used because it is relatively inexpensive and faster in getting desired results compared with performing direct testing. The correctness of a computational simulation output is highly dependent on the input and how it was processed. In this paper, simulation is done on Onera M6 Wing, to investigate the effect of a turbulence model’s application on the accuracy of the computational result. The choice of Onera M6 Wing as a simulation’s model is due to its extensive database of testing results from various wind tunnels in the world. Among Turbulence models used are Spalart-Allmaras, K-Epsilon, K-Omega, and SST.Keywords: CFD, fluent, Model, Turbulence, Onera M6, Spalart-Allmaras, K-Epsilon, K-Omega, SST.AbstraksSimulasi numerik dengan menggunakan metode Computational Fluid Dynamics (CFD) merupakan salah satu cara untuk memprediksi karakteristik suatu aliran udara yang terjadi pada model. Metode ini banyak digunakan karena sifatnya yang relatif murah dan cepat untuk mendapatkan hasil dibandingkan dengan melakukan pengujian langsung. Benar tidak hasil sebuah simulasi komputasi sangat tergantung pada inputan yang diberikan serta cara memproses data inputan tersebut. Pada tulisan ini dilakukan simulasi dengan menggunakan sayap onera M6 dengan tujuan untuk mengetahui pengaruh penggunaan model turbulensi terhadap keakuratan hasil komputasi. Pilihan sayap onera M6 sebagai model simulasi dikarenakan model tersebut sudah memiliki database hasil pengujian yang cukup lengkap dan sudah divalidasi dari berbagai terowongan angin di dunia. Model turbulensi yang digunakan diantaranya Spalart-Allmaras, K-Epsilon, K-Omega dan SST.Kata Kunci : CFD, fluent, Model, Turbulensi, Onera M6, Spalart-Allmaras, K-Epsilon, K-Omega, SST.

scholarly journals Computational fluid dynamics simulation of the turbulence models in the tested section on wind tunnel

2020 ◽  
Vol 11 (4) ◽  
pp. 1201-1209
Author(s):  
Ismail ◽  
Johanis John ◽  
Erlanda A. Pane ◽  
Budhi M. Suyitno ◽  
Gama H.N.N. Rahayu ◽  
...  
Keyword(s):  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Wind Tunnel ◽  
Turbulence Models ◽  
Dynamics Simulation ◽  
Computational Fluid Dynamics Simulation

Computational Fluid Dynamics Based Mixing Prediction for Tilt Pad Journal Bearing TEHD Modeling—Part II: Implementation With Machine Learning

2020 ◽  
Vol 143 (1) ◽  
Author(s):  
Jongin Yang ◽  
Alan Palazzolo
Keyword(s):  
Machine Learning ◽  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Three Dimensional ◽  
Leading Edge ◽  
Cfd Modeling ◽  
Inlet Temperature ◽  
Training Set ◽  
Simplified Approach ◽  
Flow Mixing

Abstract Reynolds based thermo-elasto-hydrodynamic (TEHD) simulations of tilting pad journal bearings (TPJBs) generally provide accurate results; however, the uncertainty of the pad’s leading edge thermal boundary conditions causes uncertainty of the results. The highly complex thermal-flow mixing action between pads (BPs) results from the oil supply nozzle jets and geometric features. The conventional Reynolds approach employs mixing coefficients (MCs), estimated from experience, to approximate a uniform inlet temperature for each pad. Part I utilized complex computational fluid dynamics (CFD) flow modeling to illustrate that temperature distributions at the pad inlets may deviate strongly from being uniform. The present work retains the uniform MC model but obtains the MC from detailed three-dimensional CFD modeling and machine learning, which could be extended to the radially and axially varying MC case. The steps for implementing an artificial neural network (ANN) approach for MC regression are provided as follows: (1) utilize a design of experiment step for obtaining an adaptable training set, (2) conduct CFD simulations on the BP to obtain the outputs of the training set, (3) apply an ANN learning process by Levenverg–Mardquart backpropagation with the Bayesian regularization, and (4) couple the ANN MC results with conventional TEHD Reynolds models. An approximate log fitting method provides a simplified approach for MC regression. The effectiveness of the Reynolds TEHD TPJB model with ANN regression-based MC distributions is confirmed by comparison with CFD based TEHD TPJB model results. The method obtains an accuracy nearly the same as the complete CFD model, but with the computational economy of a Reynolds approach.

Sign in / Sign up

Export Citation Format

Share Document