Guodong Chen

• Doctor of Philosophy
• PostDoc Position at University of Michigan

View Video Presentation: https://doi.org/10.2514/6.2022-1244.vid This paper presents a machine-learning approach for determining anisotropic initial meshes in the context of an output-based adaptive solution procedure. Artificial neural networks are used to predict the desired element sizing and anisotropy from flow conditions and geometry paramete...

This paper presents a new method to perform output error estimation and mesh adaptation in computational fluid dynamics (CFD) using machine-learning techniques. The error of interest is the functional output error induced by the numerical discretization, including the finite computational mesh and approximation order. Given the data of adaptive flo...

This paper presents a machine-learning approach for determining the optimal anisotropy in a computational mesh, in the context of an output-based adaptive solution procedure. Artificial neural networks are used to predict the desired element aspect ratio from readily accessible features of the primal and adjoint solutions. Whereas the sizing of the...

This paper presents a method to control the discretization error in multipoint aerodynamic shape optimization using output-based adapted meshes. The meshes are adapted via adjoint-based error estimates, taking into account both the objective and constraint output errors. A multi-fidelity optimization framework is then developed by taking advantage...

This paper presents a method for optimizing computational meshes for the prediction of scalar outputs when using hybridized and embedded discontinuous Galerkin (HDG/EDG) discretizations. Hybridization offers memory and computational time advantages compared to the standard discontinuous Galerkin (DG) method through a decoupling of elemental degrees...

The accurate prediction of transition is relevant for aerodynamic analysis and design applications. Extending the laminar flow region over airframes is a potential way to reduce the skin friction drag, which in turn reduces fuel burn and greenhouse gas emissions. This paper introduces a numerical framework that includes the modeling of transition e...

In this paper, we present a method to control the discretization error in constrained aerodynamic shape optimization problems using meshes adapted via adjoint-based error estimates. The optimization constraints may involve outputs that are not directly targeted for optimization, and hence also not for error estimation and mesh adaptation. However,...

2019 AIAA Scitech meeting slides presentation for the paper "High-Reynolds Number Transitional Flow Prediction using a Coupled Discontinuous-Galerkin RANS PSE Framework". The companion paper has the same title.

The accurate prediction of transition is relevant for many aerodynamic analysis and design applications. Extending the laminar flow region over airframes is a potential way to reduce the skin friction drag, which in turn reduces fuel burn and greenhouse gas emissions. This paper introduces a numerical framework that allows for the inclusion of tran...

In this paper, gradient-based aerodynamic shape optimization with output constraints is implemented using adaptive meshes updated via adjoint-based error estimates. All the constraints, including geometry and trim conditions, are handled simultaneously in the optimization. The trim constraints may involve outputs that are not directly targeted for...