Field Validation of a Multi-Resolution ConvLSTM Framework for Retaining Wall Deformation Prediction

arXiv cs.LG Papers

Summary

This paper validates a multi-resolution ConvLSTM framework for predicting retaining wall deformation during staged excavation, using field data from 11 sites in South Korea and achieving an average MAE of 1.4 mm.

arXiv:2606.05556v1 Announce Type: new Abstract: This study presents a comprehensive field validation of a multi-resolution Convolutional Long Short-Term Memory (ConvLSTM) framework for predicting retaining wall deformation during staged excavation. The framework is trained on Gaussian noise-augmented numerical simulations and integrates ConvLSTM models operating at different temporal resolutions through a stacking ensemble strategy. The proposed framework is validated using field monitoring data from 34 inclinometers across 11 excavation sites in South Korea. Site-wise prediction performance is systematically evaluated using multiple evaluation metrics, with analyses of the influence of temporal deformation irregularity and spatiotemporal prediction characteristics on model performance. The results demonstrate that the framework predicts retaining wall deformation associated with up to 5.0 m of additional excavation with an average mean absolute error of 1.4 mm and a coefficient of determination of 0.93 across the excavation sites. These results indicate that the framework, although trained exclusively on numerically simulated and augmented database, can be effectively applied to diverse field excavation conditions and achieve a reliable level of prediction accuracy in practical retaining wall deformation prediction.
Original Article
View Cached Full Text

Cached at: 06/05/26, 08:11 AM

# Field Validation of a Multi-Resolution ConvLSTM Framework for Retaining Wall Deformation Prediction
Source: [https://arxiv.org/abs/2606.05556](https://arxiv.org/abs/2606.05556)
[View PDF](https://arxiv.org/pdf/2606.05556)

> Abstract:This study presents a comprehensive field validation of a multi\-resolution Convolutional Long Short\-Term Memory \(ConvLSTM\) framework for predicting retaining wall deformation during staged excavation\. The framework is trained on Gaussian noise\-augmented numerical simulations and integrates ConvLSTM models operating at different temporal resolutions through a stacking ensemble strategy\. The proposed framework is validated using field monitoring data from 34 inclinometers across 11 excavation sites in South Korea\. Site\-wise prediction performance is systematically evaluated using multiple evaluation metrics, with analyses of the influence of temporal deformation irregularity and spatiotemporal prediction characteristics on model performance\. The results demonstrate that the framework predicts retaining wall deformation associated with up to 5\.0 m of additional excavation with an average mean absolute error of 1\.4 mm and a coefficient of determination of 0\.93 across the excavation sites\. These results indicate that the framework, although trained exclusively on numerically simulated and augmented database, can be effectively applied to diverse field excavation conditions and achieve a reliable level of prediction accuracy in practical retaining wall deformation prediction\.

## Submission history

From: Jihoon Kim Mr\. \[[view email](https://arxiv.org/show-email/29274497/2606.05556)\] **\[v1\]**Thu, 4 Jun 2026 01:09:41 UTC \(5,491 KB\)

Similar Articles

Beyond Surface Statistics: Robust Conformal Prediction for LLMs via Internal Representations

arXiv cs.CL

This paper proposes a conformal prediction framework for LLMs that leverages internal representations rather than output-level statistics, introducing Layer-Wise Information (LI) scores as nonconformity measures to improve validity-efficiency trade-offs under distribution shift. The method demonstrates stronger robustness to calibration-deployment mismatch compared to text-level baselines across QA benchmarks.

Evaluating Transformer and LSTM Frameworks for Prediction in Ungauged Basins

arXiv cs.AI

This paper evaluates encoder-only Transformer and LSTM models for streamflow prediction in ungauged basins using NOAA's National Water Model simulations. Results show LSTM outperforms Transformer, and incorporating downstream information significantly improves prediction skill across both architectures.

LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

Papers with Code Trending

LeWorldModel introduces a stable, end-to-end Joint-Embedding Predictive Architecture that trains directly from pixels with minimal hyperparameters and provable anti-collapse guarantees. It achieves significant speedups in planning compared to foundation models while maintaining competitive performance on robotic manipulation tasks.