no code implementations • 18 Apr 2024 • Jose Florido, He Wang, Amirul Khan, Peter K. Jimack
Physics-informed neural networks (PINNs) provide a means of obtaining approximate solutions of partial differential equations and systems through the minimisation of an objective function which includes the evaluation of a residual function at a set of collocation points within the domain.
no code implementations • 8 Jul 2023 • Maria Luisa Taccari, Oded Ovadia, He Wang, Adar Kahana, Xiaohui Chen, Peter K. Jimack
This paper presents a comprehensive comparison of various machine learning models, namely U-Net, U-Net integrated with Vision Transformers (ViT), and Fourier Neural Operator (FNO), for time-dependent forward modelling in groundwater systems.
no code implementations • 9 Apr 2022 • Maria Luisa Taccari, Jonathan Nuttall, Xiaohui Chen, He Wang, Bennie Minnema, Peter K. Jimack
This manuscript presents an Attention U-Net model that attempts to capture the fundamental input-output relations of the groundwater system and generates solutions of hydraulic head in the whole domain given a set of physical parameters and boundary conditions.
no code implementations • 15 Apr 2020 • Zheyan Zhang, Yongxing Wang, Peter K. Jimack, He Wang
The framework that we have developed is based around training an artificial neural network (ANN) to guide standard mesh generation software, based upon a prediction of the required local mesh density throughout the domain.
