Auto-Regressive U-Net for Full-Field Prediction of Shrinkage-Induced Damage in Concrete

2025-09-24

AI Summary

This study addresses the challenge of predicting spatiotemporal damage evolution induced by concrete drying shrinkage at the mesoscale. It elucidates the mechanistic influence of aggregate geometric features—namely shape, size, and spatial distribution—on effective shrinkage strain and stiffness degradation. A novel dual-network architecture is proposed: an autoregressive U-Net predicts sequential damage fields from microstructural images and initial shrinkage fields, while a parallel CNN regresses effective shrinkage strain and residual elastic modulus. Trained end-to-end on synthetic microstructure datasets, the model achieves high accuracy (prediction error < 5%) and accelerates computation by two to three orders of magnitude relative to conventional mesoscale simulations. The framework provides an interpretable, efficient, and design-integrated paradigm for modeling shrinkage-induced damage mechanisms—enabling direct incorporation into mixture proportioning optimization workflows.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Computer Vision: Learning & Optimization for CV
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

This paper introduces a deep learning approach for predicting time-dependent full-field damage in concrete. The study uses an auto-regressive U-Net model to predict the evolution of the scalar damage field in a unit cell given microstructural geometry and evolution of an imposed shrinkage profile. By sequentially using the predicted damage output as input for subsequent predictions, the model facilitates the continuous assessment of damage progression. Complementarily, a convolutional neural network (CNN) utilises the damage estimations to forecast key mechanical properties, including observed shrinkage and residual stiffness. The proposed dual-network architecture demonstrates high computational efficiency and robust predictive performance on the synthesised datasets. The approach reduces the computational load traditionally associated with full-field damage evaluations and is used to gain insights into the relationship between aggregate properties, such as shape, size, and distribution, and the effective shrinkage and reduction in stiffness. Ultimately, this can help to optimize concrete mix designs, leading to improved durability and reduced internal damage.

Problem

• Predicting time-dependent full-field damage evolution in concrete microstructures
• Reducing computational load associated with traditional damage evaluation methods
• Understanding how aggregate properties affect shrinkage and stiffness reduction

Innovation

• Auto-regressive U-Net predicts full-field damage evolution
• Convolutional neural network forecasts key mechanical properties
• Dual-network architecture enables efficient damage assessment