Dr Minfei Liang

Abstract:

This study presents an integrated finite-element–machine-learning framework for predicting early-age stress evolution in concrete materials/structures by combining an enhanced thermo-chemo-mechanical (TCM) model, deep sequential learning (DSL), and active learning (AL). The proposed TCM model incorporates experimentally informed viscoelasticity, a stable exponential creep–relaxation conversion, and an efficient exponential algorithm for the Maxwell-chain formulation in finite element analysis, which is further validated by a temperature stress testing machine. This model generates high-fidelity stress–time data across diverse mixtures, temperatures, and structural configurations. These simulations are used to train a Gated Recurrent Unit with Monte Carlo Dropout (GRU-MCD) model, whose predictive performance surpasses conventional point-wise approaches such as Light Gradient Boosting Machine and Gaussian Process Regression, yielding higher accuracy with reduced overfitting. The AL strategy further enhances efficiency by enabling the GRU-MCD model to achieve the accuracy of ∼900 Latin Hypercube samples using only ∼200 samples selected by active learning. Although demonstrated on a wall–base structure, the proposed framework is general and applicable to other cementitious material or structural systems, providing an effective tool for cracking-risk evaluation, reliability analysis, and the design of low-carbon concrete structures.

Abstract:

Temperature Stress Testing Machine (TSTM) is a universal testing tool for many properties relevant to early-age cracking of cementitious materials. However, the complexity of TSTMs require heavy lab work and thus hinders a more thorough parametric study on a range of cementitious materials. This study presents the development and validation of a Mini-TSTM for efficiently testing the autogenous deformation (AD), viscoelastic properties, and their combined results, the early-age stress (EAS). The setup was validated through systematic tests of EAS, AD, elastic modulus, and creep. Besides, the heating/cooling capability of the setup was examined by tests of coefficient of thermal expansion by temperature cycles. The results of EAS correspond well to that of AD, which qualitatively validates the developed setup. To quantitatively validate the setup, a classical viscoelastic model was built, based on the scenario of a 1-D uniaxial restraint test, to predict the EAS results with the tested AD, elastic modulus, and creep of the same cementitious material as the input. The predicted EAS matched the testing results of Mini-TSTM with good accuracy in 6 different cases. The viscoelastic model also provided quantitative explanations for why variations in early AD do not influence the EAS results. The testing and modelling results together validate the developed Mini-TSTM setup as an efficient tool for studying early-age cracking of cementitious materials. At the end, the potential limitations of the Mini-TSTM are discussed and its applicability for concrete with aggregate size up to 22 mm is demonstrated.