A coupled framework for power load forecasting with Gaussian implicit spatio temporal block and attention mechanisms network

Computers & Electrical Engineering, Journal Year: 2025, Volume and Issue: 123, P. 110263 - 110263

Published: March 20, 2025

Language: Английский

---

Flo-Sr: Deep Learning-Based Urban Flood Super-Resolution Model

Published: Jan. 1, 2025

Language: Английский

---

Interpretable physics‐informed graph neural networks for flood forecasting

Computer-Aided Civil and Infrastructure Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: April 14, 2025

Abstract Climate change has intensified extreme weather events, with floods causing significant socioeconomic and environmental damage. Accurate flood forecasting is crucial for disaster preparedness risk mitigation, yet traditional hydrodynamic models, while precise, are computationally prohibitive real‐time applications. Machine learning surrogates, such as graph neural networks (GNNs), improve efficiency but often lack physical consistency interpretability. This paper introduces HydroGraphNet, a novel physics‐informed GNN framework that, the first time, integrates Kolmogorov–Arnold Network (KAN) to enhance model interpretability in unstructured mesh‐based forecasting. The embeds mass conservation laws into loss function, ensuring physically consistent predictions. Additionally, it employs an autoregressive encoder–processor–decoder architecture that captures spatiotemporal dynamics mitigating error accumulation over long horizons. Validation on data from White River near Muncie, Indiana, demonstrates 67% reduction prediction error, near‐zero balance 58% improvement critical success index major events compared baseline model. These results highlight potential of proposed advance improved

Language: Английский

---

Internal flow characteristics of a tank with multiple jet inflows and liquid sloshing

Physics of Fluids, Journal Year: 2025, Volume and Issue: 37(4)

Published: April 1, 2025

This paper conducted numerical investigations into the flow field characteristics of a multi-inlet tank, focusing on coupled effects jet inflows and liquid sloshing. Turbulent swirling was numerically investigated using Reynolds Stress Model, combined with Volume Fluid method Adaptive Mesh Refinement technique for accurate free surface capturing. The vorticity structure identified Q-criterion. Numerical simulations were validated against experimental data, confirming reliability accuracy model. A systematic parametric study to investigate inlet pipe diameter, inflow rate, tank immersed depth. Four points fitting curves established describe relationships between maximum velocity these parameters. results indicated that distribution uniformity mostly affected by diameter. Additionally, various filling levels analyzed sloshing induced surge motion. probability index significantly at an depth 0.8 m. hybrid neural network framework integrating Proper Orthogonal Decomposition (POD) Long Short-Term Memory (LSTM) networks developed predict field. POD employed extract dominant modes, while corresponding temporal coefficients fed LSTM prediction. reconstructed fields demonstrated effectiveness POD-LSTM model in accurately predicting evolution field, as confirmed comparisons simulated results.

Language: Английский

---

Location Invariant Flood Prediction using Fourier Neural Operator

Published: Dec. 18, 2024

Language: Английский

---