Qizhi He

TL;DR: It is demonstrated that the physics-informed deep neural networks used for estimating space-dependent hydraulic conductivity, hydraulic head, and concentration fields from sparse measurements are significantly more accurate than standard data-driven DNNs when the training set consists of sparse data.

TL;DR: This method is able to improve the boundary description quality of the optimal result with much less design variables as compared with the case of global refinement, and therefore can greatly reduce the computational burden involved in the sensitivity analysis and optimization process.

TL;DR: In this article, an adaptive method for topology optimization of structures, by using independent error control for the separated displacement and material density fields, is proposed, which can achieve high quality and high-accuracy optimal solutions comparable to those obtained with fixed globally fine analysis meshes and fine distributed density points, but with much less computational cost.

TL;DR: In this paper, a new data-driven simulation approach coupled with a locally convex reconstruction is proposed to enhance accuracy and robustness against noise and outliers in data sets in the data driven computing.

TL;DR: Based on the element-free Galerkin (EFG) method, an analysis-independent density variable approach is proposed for topology optimization of geometrically nonlinear structures as mentioned in this paper.