City Research Online

Abstract

There is relatively limited research work on sloshing in cylinder geometries compared to the extensive studies conducted on upright cylinders, spherical containers and cuboids. As the mesh-free approach, the incompressible Smoothed Particle Hydrodynamics (ISPH) method is emerging as a potential tool for simulating the breaking waves and wave-structure interactions. The pressure in the conventional ISPH method is obtained by solving the pressure Poisson’s equation (PPE), which is the most time-consuming part. Recently, the machine learning (ML) techniques has been widely used in the fluid dynamics. In this paper, a graph neural network (GNN) supported ISPH method, in which the GNN model is combined with ISPH and used to predict the fluid pressure instead of solving the PPE directly, is applied to simulate sloshing in a circular tank. The ISPH_GNN with one trained GNN model based on training data generating from relatively simple cases will be extended to simulate sloshing in a circular tank under forced harmonic horizontal and vertical excitation. For comparison, results from the conventional ISPH method will also be presented. In addition, the computational efficiency of ISPH_GNN will also be investigated.

Publication Type:	Conference or Workshop Item (Paper)
Publisher Keywords:	machine learning, simulation, computational efficiency, artificial intelligence, circular tank, isph method, conventional isph method, training data, neural network, gnn
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) T Technology > TC Hydraulic engineering. Ocean engineering
Departments:	School of Science & Technology School of Science & Technology > Department of Engineering
SWORD Depositor:	Symplectic Administrator

Export

Downloads