Deep Learning-Based Prediction of Unsteady Reynolds-Averaged Navier-Stokes Solutions for Vertical-Axis Turbines
The following study investigates the effectiveness of a deep learning-based method for predicting the flow field and flow-driven rotation of a vertical-axis hydrokinetic turbine operating in previously unseen free-stream velocities. A Convolutional Neural Network (CNN) is trained and tested using th...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/16/3/1130 |
| _version_ | 1797624775080673280 |
|---|---|
| author | Chloë Dorge Eric Louis Bibeau |
| author_facet | Chloë Dorge Eric Louis Bibeau |
| author_sort | Chloë Dorge |
| collection | DOAJ |
| description | The following study investigates the effectiveness of a deep learning-based method for predicting the flow field and flow-driven rotation of a vertical-axis hydrokinetic turbine operating in previously unseen free-stream velocities. A Convolutional Neural Network (CNN) is trained and tested using the solutions of five two-dimensional (2-D), foil-resolved Unsteady Reynolds-Averaged Navier-Stokes (URANS) simulations, with free-stream velocities of 1.0, 1.5, 2.0, 2.5, and 3.0 m/s. Based on the boundary conditions of free-stream velocity and rotor position, the flow fields of x-velocity, y-velocity, pressure, and turbulent viscosity are inferred, in addition to the angular velocity of the rotor. Three trained CNN models are developed to evaluate the effects of (1) the dimensions of the training data, and (2) the number of simulations used as training cases. Reducing data dimensions was found to diminish mean relative error in predictions of velocity and turbulent viscosity, while increasing it in predictions of pressure and angular velocity. Increasing the number of training cases from two to three was found to reduce relative error for all predicted unknowns. With the best achieved CNN model, the variables of x-velocity, y-velocity, pressure, turbulent viscosity, and angular velocity were inferred with mean relative errors of 6.93%, 9.82%, 10.7%, 7.48%, and 0.817%, respectively. |
| first_indexed | 2024-03-11T09:47:15Z |
| format | Article |
| id | doaj.art-c58dd6397d9c4705a35d64fc9d4d32ee |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-03-11T09:47:15Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-c58dd6397d9c4705a35d64fc9d4d32ee2023-11-16T16:33:12ZengMDPI AGEnergies1996-10732023-01-01163113010.3390/en16031130Deep Learning-Based Prediction of Unsteady Reynolds-Averaged Navier-Stokes Solutions for Vertical-Axis TurbinesChloë Dorge0Eric Louis Bibeau1Department of Mechanical Engineering, University of Manitoba, 75 Chancellors Cir, Winnipeg, MB R3T 5V6, CanadaDepartment of Mechanical Engineering, University of Manitoba, 75 Chancellors Cir, Winnipeg, MB R3T 5V6, CanadaThe following study investigates the effectiveness of a deep learning-based method for predicting the flow field and flow-driven rotation of a vertical-axis hydrokinetic turbine operating in previously unseen free-stream velocities. A Convolutional Neural Network (CNN) is trained and tested using the solutions of five two-dimensional (2-D), foil-resolved Unsteady Reynolds-Averaged Navier-Stokes (URANS) simulations, with free-stream velocities of 1.0, 1.5, 2.0, 2.5, and 3.0 m/s. Based on the boundary conditions of free-stream velocity and rotor position, the flow fields of x-velocity, y-velocity, pressure, and turbulent viscosity are inferred, in addition to the angular velocity of the rotor. Three trained CNN models are developed to evaluate the effects of (1) the dimensions of the training data, and (2) the number of simulations used as training cases. Reducing data dimensions was found to diminish mean relative error in predictions of velocity and turbulent viscosity, while increasing it in predictions of pressure and angular velocity. Increasing the number of training cases from two to three was found to reduce relative error for all predicted unknowns. With the best achieved CNN model, the variables of x-velocity, y-velocity, pressure, turbulent viscosity, and angular velocity were inferred with mean relative errors of 6.93%, 9.82%, 10.7%, 7.48%, and 0.817%, respectively.https://www.mdpi.com/1996-1073/16/3/1130deep learningvertical-axis turbineturbine interactionarray optimizationURANSCFD |
| spellingShingle | Chloë Dorge Eric Louis Bibeau Deep Learning-Based Prediction of Unsteady Reynolds-Averaged Navier-Stokes Solutions for Vertical-Axis Turbines Energies deep learning vertical-axis turbine turbine interaction array optimization URANS CFD |
| title | Deep Learning-Based Prediction of Unsteady Reynolds-Averaged Navier-Stokes Solutions for Vertical-Axis Turbines |
| title_full | Deep Learning-Based Prediction of Unsteady Reynolds-Averaged Navier-Stokes Solutions for Vertical-Axis Turbines |
| title_fullStr | Deep Learning-Based Prediction of Unsteady Reynolds-Averaged Navier-Stokes Solutions for Vertical-Axis Turbines |
| title_full_unstemmed | Deep Learning-Based Prediction of Unsteady Reynolds-Averaged Navier-Stokes Solutions for Vertical-Axis Turbines |
| title_short | Deep Learning-Based Prediction of Unsteady Reynolds-Averaged Navier-Stokes Solutions for Vertical-Axis Turbines |
| title_sort | deep learning based prediction of unsteady reynolds averaged navier stokes solutions for vertical axis turbines |
| topic | deep learning vertical-axis turbine turbine interaction array optimization URANS CFD |
| url | https://www.mdpi.com/1996-1073/16/3/1130 |
| work_keys_str_mv | AT chloedorge deeplearningbasedpredictionofunsteadyreynoldsaveragednavierstokessolutionsforverticalaxisturbines AT ericlouisbibeau deeplearningbasedpredictionofunsteadyreynoldsaveragednavierstokessolutionsforverticalaxisturbines |