Accelerating HEC-RAS: A Recurrent Neural Operator for Rapid River Forecasting

2025-07-21

AI Summary

Traditional high-fidelity hydraulic models (e.g., HEC-RAS) incur prohibitive computational costs for real-time flood forecasting and are ill-suited for ensemble prediction. To address this, we propose a hybrid autoregressive surrogate model that integrates a Gated Recurrent Unit (GRU) with a geometry-aware Fourier Neural Operator (Geo-FNO). Innovatively, HEC-RAS serves solely as a high-fidelity data generator; the surrogate learns spatiotemporally coupled hydrodynamic dynamics implicitly from only eight lightweight input features. Evaluated across 67 river reaches, the model achieves a median absolute water level error of just 0.31 ft. Crucially, full-ensemble forecasting time drops from 139 minutes to 40 minutes—a 3.5× speedup—while preserving physical consistency. This advancement substantially enhances the practical feasibility of high-accuracy hydraulic modeling in real-time operational decision-making and large-scale ensemble forecasting.

Technology Category

• Machine Learning: Ensemble Methods
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Physics-based solvers like HEC-RAS provide high-fidelity river forecasts but are too computationally intensive for on-the-fly decision-making during flood events. The central challenge is to accelerate these simulations without sacrificing accuracy. This paper introduces a deep learning surrogate that treats HEC-RAS not as a solver but as a data-generation engine. We propose a hybrid, auto-regressive architecture that combines a Gated Recurrent Unit (GRU) to capture short-term temporal dynamics with a Geometry-Aware Fourier Neural Operator (Geo-FNO) to model long-range spatial dependencies along a river reach. The model learns underlying physics implicitly from a minimal eight-channel feature vector encoding dynamic state, static geometry, and boundary forcings extracted directly from native HEC-RAS files. Trained on 67 reaches of the Mississippi River Basin, the surrogate was evaluated on a year-long, unseen hold-out simulation. Results show the model achieves a strong predictive accuracy, with a median absolute stage error of 0.31 feet. Critically, for a full 67-reach ensemble forecast, our surrogate reduces the required wall-clock time from 139 minutes to 40 minutes, a speedup of nearly 3.5 times over the traditional solver. The success of this data-driven approach demonstrates that robust feature engineering can produce a viable, high-speed replacement for conventional hydraulic models, improving the computational feasibility of large-scale ensemble flood forecasting.

Problem

• Accelerate HEC-RAS simulations for rapid river forecasting
• Maintain accuracy while reducing computational time
• Replace traditional hydraulic models with deep learning surrogates

Innovation

• Hybrid GRU and Geo-FNO for river forecasting
• Learns physics from minimal feature vectors
• Reduces simulation time by 3.5 times