{"id":"arxiv/2601.04408","url":"https://arxiv.org/abs/2601.04408","tags":["math.DS"],"score":0,"title":"G-KdVNet: ANN-ADM Surrogate for Geophysical KdV Equation","domain":"math.DS","source":"arxiv","authors":["Mrutyunjaya Sahoo","Arup Kumar Sahoo","Snehashish Chakraverty"],"reviews":0,"subject":"math.DS","summary":"This research examines the influence of the Coriolis parameter on the behaviour of the geophysical Korteweg-de Vries (KdV) equation. To efficiently approximate its solution, a novel surrogate framework, termed G-KdVNet, is proposed by integrating artificial neural networks with the Adomian decomposition method (ADM). In the proposed approach, ADM is first employed to generate reliable semi-analytical solution data, which are subsequently used to train the neural network model. The developed model demonstrates strong predictive capability in capturing the nonlinear dynamics of the KdV system.…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-01-07","reactions":{}}