Stable Long-Horizon Spatiotemporal Prediction on Meshes Using Latent Multiscale Recurrent Graph Neural Networks

• URL: http://arxiv.org/abs/2602.18146v1
• Date: Fri, 20 Feb 2026 11:22:47 GMT
• Title: Stable Long-Horizon Spatiotemporal Prediction on Meshes Using Latent Multiscale Recurrent Graph Neural Networks
• Authors: Lionel Salesses, Larbi Arbaoui, Tariq Benamara, Arnaud Francois, Caroline Sainvitu,
• Abstract summary: We propose a deep learning framework for predicting full temperature histories directly on meshes.<n>The framework maintains over thousands of time steps and generalizing across heterogeneous geometries.<n>Experiments on simulated powder bed fusion data demonstrate accurate and temporally stable long-horizon predictions.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Accurate long-horizon prediction of spatiotemporal fields on complex geometries is a fundamental challenge in scientific machine learning, with applications such as additive manufacturing where temperature histories govern defect formation and mechanical properties. High-fidelity simulations are accurate but computationally costly, and despite recent advances, machine learning methods remain challenged by long-horizon temperature and gradient prediction. We propose a deep learning framework for predicting full temperature histories directly on meshes, conditioned on geometry and process parameters, while maintaining stability over thousands of time steps and generalizing across heterogeneous geometries. The framework adopts a temporal multiscale architecture composed of two coupled models operating at complementary time scales. Both models rely on a latent recurrent graph neural network to capture spatiotemporal dynamics on meshes, while a variational graph autoencoder provides a compact latent representation that reduces memory usage and improves training stability. Experiments on simulated powder bed fusion data demonstrate accurate and temporally stable long-horizon predictions across diverse geometries, outperforming existing baseline. Although evaluated in two dimensions, the framework is general and extensible to physics-driven systems with multiscale dynamics and to three-dimensional geometries.

Related papers

• Online learning of subgrid-scale models for quasi-geostrophic turbulence in planetary interiors [41.99844472131922]
  We study quasi-geostrophic turbulent flow in an axisymmetric bounded domain.<n>Flow is driven by a prescribed analytical forcing.<n>We show that an SGS model trained on data spanning only one turnover time remains stable and accurate over integrations at least a hundred times longer than the training period.
  arXiv  Detail & Related papers  (2025-11-18T15:21:38Z)
• Breaking the Discretization Barrier of Continuous Physics Simulation Learning [16.740327071700268]
  We propose a purely data-driven method to model continuous physics simulation from partial observations.<n>Specifically, we employ multiplicative filter network to fuse and encode spatial information with the corresponding observations.<n>We customize geometric grids and use message-passing mechanism to map features from original spatial domain to the customized grids.
  arXiv  Detail & Related papers  (2025-09-22T16:10:58Z)
• Transformer with Koopman-Enhanced Graph Convolutional Network for Spatiotemporal Dynamics Forecasting [12.301897782320967]
  TK-GCN is a two-stage framework that integrates geometry-aware spatial encoding with long-range temporal modeling.<n>We show that TK-GCN consistently delivers superior predictive accuracy across a range of forecast horizons.
  arXiv  Detail & Related papers  (2025-07-05T01:26:03Z)
• Geometry-aware Active Learning of Spatiotemporal Dynamic Systems [4.251030047034566]
  This paper proposes a geometry-aware active learning framework for modeling dynamic systems.<n>We develop an adaptive active learning strategy to strategically identify spatial locations for data collection and further maximize the prediction accuracy.
  arXiv  Detail & Related papers  (2025-04-26T19:56:38Z)
• Conservation-informed Graph Learning for Spatiotemporal Dynamics Prediction [84.26340606752763]
  In this paper, we introduce the conservation-informed GNN (CiGNN), an end-to-end explainable learning framework.<n>The network is designed to conform to the general symmetry conservation law via symmetry where conservative and non-conservative information passes over a multiscale space by a latent temporal marching strategy.<n>Results demonstrate that CiGNN exhibits remarkable baseline accuracy and generalizability, and is readily applicable to learning for prediction of varioustemporal dynamics.
  arXiv  Detail & Related papers  (2024-12-30T13:55:59Z)
• Generalizing Weather Forecast to Fine-grained Temporal Scales via Physics-AI Hybrid Modeling [55.13352174687475]
  This paper proposes a physics-AI hybrid model (i.e., WeatherGFT) which generalizes weather forecasts to finer-grained temporal scales beyond training dataset.<n>Specifically, we employ a carefully designed PDE kernel to simulate physical evolution on a small time scale.<n>We also introduce a lead time-aware training framework to promote the generalization of the model at different lead times.
  arXiv  Detail & Related papers  (2024-05-22T16:21:02Z)
• Towards Fast Simulation of Environmental Fluid Mechanics with Multi-Scale Graph Neural Networks [0.0]
  We introduce MultiScaleGNN, a novel multi-scale graph neural network model for learning to infer unsteady continuum mechanics. We demonstrate this method on advection problems and incompressible fluid dynamics, both fundamental phenomena in oceanic and atmospheric processes. Simulations obtained with MultiScaleGNN are between two and four orders of magnitude faster than those on which it was trained.
  arXiv  Detail & Related papers  (2022-05-05T13:33:03Z)
• Multivariate Time Series Forecasting with Dynamic Graph Neural ODEs [65.18780403244178]
  We propose a continuous model to forecast Multivariate Time series with dynamic Graph neural Ordinary Differential Equations (MTGODE) Specifically, we first abstract multivariate time series into dynamic graphs with time-evolving node features and unknown graph structures. Then, we design and solve a neural ODE to complement missing graph topologies and unify both spatial and temporal message passing.
  arXiv  Detail & Related papers  (2022-02-17T02:17:31Z)
• Leveraging the structure of dynamical systems for data-driven modeling [111.45324708884813]
  We consider the impact of the training set and its structure on the quality of the long-term prediction. We show how an informed design of the training set, based on invariants of the system and the structure of the underlying attractor, significantly improves the resulting models.
  arXiv  Detail & Related papers  (2021-12-15T20:09:20Z)
• Stochastically forced ensemble dynamic mode decomposition for forecasting and analysis of near-periodic systems [65.44033635330604]
  We introduce a novel load forecasting method in which observed dynamics are modeled as a forced linear system. We show that its use of intrinsic linear dynamics offers a number of desirable properties in terms of interpretability and parsimony. Results are presented for a test case using load data from an electrical grid.
  arXiv  Detail & Related papers  (2020-10-08T20:25:52Z)
• Learning to Simulate Complex Physics with Graph Networks [68.43901833812448]
  We present a machine learning framework and model implementation that can learn to simulate a wide variety of challenging physical domains. Our framework---which we term "Graph Network-based Simulators" (GNS)--represents the state of a physical system with particles, expressed as nodes in a graph, and computes dynamics via learned message-passing. Our results show that our model can generalize from single-timestep predictions with thousands of particles during training, to different initial conditions, thousands of timesteps, and at least an order of magnitude more particles at test time.
  arXiv  Detail & Related papers  (2020-02-21T16:44:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.