Airfoil Optimization using a physics-constrained neural network

Duration: 9 mins 50 secs

Share this media item:

Embed this media item:

<iframe width="_width_" height="_height_" src="https://sms.cam.ac.uk/media/3712490/embed" frameborder="0" scrolling="no" allowfullscreen></iframe>

Choose size:

About this item

Available Formats

About this item

Airfoil Optimization using a physics-constrained neural network's image

Description:	Presentation to the American Physical Society at the 2021 meeting of the Division of Fluid Dynamics


Created:	2021-11-25 20:48
Collection:	Matthew Juniper conference presentations
Publisher:	University of Cambridge
Copyright:	Dr Yubiao Sun
Language:	eng (English)
Keywords:	PINN; Neural Networks; Optimization;


Abstract:	Shape optimization, which involves modelling and optimization of a designed geometry to achieve targeted goals, is a prominent but challenging topic. The complexity and high dimensionality of the search space make some existing methods computationally expensive. In this talk, we propose a physics-informed neural networks (PINN) as a solver for the flow around an object and also a provider of gradient information for shape optimization. In this study a PINN is employed to solve the flow around an airfoil and to optimize its shape to maximize lift over drag. The point cloud used for training the PINN is adapted using the gradient of the objective functions so that accurate flow fields can be obtained for geometries closer to the optimal shape.

Abstract:

Shape optimization, which involves modelling and optimization of a designed geometry to achieve targeted goals, is a prominent but challenging topic. The complexity and high dimensionality of the search space make some existing methods computationally expensive. In this talk, we propose a physics-informed neural networks (PINN) as a solver for the flow around an object and also a provider of gradient information for shape optimization. In this study a PINN is employed to solve the flow around an airfoil and to optimize its shape to maximize lift over drag. The point cloud used for training the PINN is adapted using the gradient of the objective functions so that accurate flow fields can be obtained for geometries closer to the optimal shape.

Available Formats

Format	Quality	Bitrate	Size
MPEG-4 Video	640x360	825.13 kbits/sec	59.43 MB	View	Download
WebM	640x360	350.56 kbits/sec	25.29 MB	View	Download
iPod Video	480x360	472.23 kbits/sec	34.01 MB	View	Download
MP3	44100 Hz	249.94 kbits/sec	18.28 MB	Listen	Download
Auto *	(Allows browser to choose a format it supports)