ライブラリ登録: Guest

ISSN Online: 2688-7231

ISBN Online: 978-1-56700-524-0

Proceedings of the 26thNational and 4th International ISHMT-ASTFE Heat and Mass Transfer Conference December 17-20, 2021, IIT Madras, Chennai-600036, Tamil Nadu, India
December, 17-20, 2021, IIT Madras, Chennai, India

Machine Learning Technique to assist Computational Fluid Dynamics

Get access (open in a dialog) DOI: 10.1615/IHMTC-2021.2810
pages 1859-1865

要約

The key to better design of an industrial scale wind turbine is to understand the influence of blade geometry and its dynamics on the complicated flow-structures. An industrial-scale wind turbine can be numerically represented using various approaches (from simpler 2D steady flow to complex 3D with moving mesh) that can alter the results substantially. Therefore, in this work the NREL 5MW turbine is used for understanding the associated property-complexities due to various geometric approximation. Numerical Analysis carried out on turbine blades and Machine learning is applied to predict the performance of the turbine. Ansys fluent was used to simulate with different inlet conditions and obtain datasets. These data were provided to the machine learning model which then predicts suitable equation that depicts the output character. The main objective is to apply machine learning model which reduces the computational time of the prediction without any reduction in the accuracy.