Learned Response-Field Inertia Operator for HEC-RAS 2D Water-Surface Elevation Prediction

2026-06-04

AI Summary

This study addresses the accuracy and computational efficiency limitations of the HEC-RAS 2D model on non-uniform grids, which arise from raster remapping-induced errors in water surface elevation prediction. To overcome these challenges, the authors propose an incremental surrogate model operating directly on native computational cells. The approach introduces a learnable response field inertia operator (LRFIO) that functions without external forcing, decoupling static inputs, current hydraulic states, and future targets to enable adaptive selection among persistence-based, global, or piecewise response structures. Integrated with residual correction and a validation-driven complexity control strategy, the model achieves a maximum validation regret of 4.30% across four benchmark cases, with single-step rolling inference times ranging from 0.003 to 0.242 seconds—yielding a speedup of up to 2.75 × 10⁴ compared to HEC-RAS while maintaining high fidelity.

Abstract

This article presents a cross-dataset evaluation of learned native-cell surrogate models for solver-consistent water-surface elevation (WSE) prediction in HEC-RAS 2D. To avoid raster remapping error and information-access confounding, surrogates are evaluated directly on the original nonuniform computational cells under an explicit policy that separates static project inputs, current hydraulic state, project-input forcing, calibration-derived quantities, and future solver-output targets. We introduce the Learned Response-Field Inertia Operator (LRFIO), a no-forcing, increment-based learned surrogate that calibrates an inertial response operator from solved HEC-RAS trajectories and deploys the retained operator through closed-form native-cell rollout. LRFIO evaluates a base-case-first response hierarchy consisting of persistence, global calibrated inertia, and segmented response-field inertia. Segmentation, residual correction, and neuralized inertia are treated as learnable modeling choices, with added complexity retained only when validation evidence justifies its cost. Evaluated across four diverse HEC-RAS 2D benchmarks, LRFIO retains different response structures for different domains, demonstrating adaptive learned complexity. The selector audit shows controlled complexity with a maximum validation regret of 4.30%. During deployment, retained rollout times range from 0.003 s to 0.242 s, and the Beaver Bayou measured-solve comparison gives an estimated 2.75 x 10^4 horizon-normalized speedup over HEC-RAS. These results indicate that the current native-cell increment is a strong solver-conditioned predictive scaffold and that added response-field, neural, or spatial complexity should be retained only when empirically justified.

Problem

• water-surface elevation prediction
• HEC-RAS 2D
• surrogate modeling
• native-cell prediction
• solver consistency

Innovation

• Learned Response-Field Inertia Operator
• HEC-RAS 2D surrogate modeling
• native-cell rollout
• adaptive model complexity
• solver-consistent prediction