Engineering infrastructure includes a large number of concrete structures. The main binder used in concrete is Portland cement, which reacts with water when mixing. During the chemical reaction (cement hydration) heat is released causing non-uniform temperature distribution and subsequent internal stresses that may crack the concrete. Therefore, accurate prediction of concrete temperature evolution in design and control of concrete temperature in construction are crucial tasks for engineers. This paper presents the development of a finite difference (FD) model for predicting thermal behavior of concrete structures during cement hydration. A formwork or insulation layer that covers the outer of the concrete was accounted for in the model. The temperature profiles calculated by the proposed model show close agreement with those measured from a concrete bridge footing and a concrete cube. The Particle Swarm Optimization (PSO) algorithm was adopted to compute an “equivalent convection coefficient” for the convection boundary condition in the simplified model for several common insulation materials. The results show that using the simplified model along with an equivalent convection coefficient can give a significant reduction in the computation time while the accuracy is still maintained. In addition, an equation for estimating an equivalent convection coefficient for a given insulation material was developed for design and practical purposes. The proposed method can be used in the design of concrete structures and control of concrete temperature to prevent early-age thermal cracking thus ensuring the expected durability and service life. From the obtained results, varying equivalent convection values are also recommended to be used for different insulation materials.

Recently, researches have been used Artificial Neural Network (ANN) to predict the early-age thermal cracking of rectangle piers. But ANN has not resulted for different types of concrete piers. This article presents an evaluation of the early-age thermal characteristics of mass concrete piers with four distinct cross-sectional shapes. A finite element (FE) model was employed to estimate the maximum temperature, thermal stress, and cracking potential of the concrete pier at its early age. To investigate the impact of various pier geometries on the thermal cracking potential, different pier geometries were considered. In this study, an ANN model was utilized to predict the maximum temperature and decrease the risk of cracking in mass concrete piers at early age. The database of thermal mass concrete piers used in this study comprises 128 results obtained from the FE model. The results of the analysis indicate that the ANN model can predict early-age thermal parameters, and cracking risk in early-age concrete piers with good accuracy and help to the designer to choose the appropriate size in minimizing cracks on the pier concrete.

Tạp chí Khoa học Giao thông vận tải - Trường Đại học Giao thông vận tải;Transport and Communications Science Journal under University of Transport and Communications was established in 2002 and launched online version in 2019.

Sửa chữa và tăng cường kết cấu mặt cầu bê tông cốt thép đang là đối tượng của nhiều nghiên cứu, tuy nhiên hiện nay chưa có nhiều nghiên cứu về sửa chữa kết cấu này bằng vật liệu bê tông cường độ siêu cao (UHPC). Bài báo trình bày kết quả nghiên cứu thực nghiệm ứng xử chịu uốn của 03 mẫu bản: 01 mẫu bản bê tông cốt thép (BTCT), 02 mẫu cùng chiều cao với mẫu BTCT có thay thế 03 cm lớp bê tông bảo vệ bằng lớp UHPC neo liên kết hoặc không với bản bê tông hiện hữu. Kết quả thí nghiệm uốn 04 điểm cho thấy việc tăng cường UHPC ở lớp phủ bê tông bề mặt sẽ tăng độ võng tới hạn lên 133%, mô men uốn phá hoại so với mẫu đối chứng lên 31,4% đồng thời làm thay đổi đặc điểm phá hoại của mẫu. Đối với mẫu không có liên hết với bản bê tông hiện hữu, mẫu bị phá hoại với mô men uốn nhỏ hơn và xuất hiện hiện tượng bong tách 02 lớp này. Nghiên cứu này đã mở ra một hướng nghiên cứu mới và tiềm năng cho việc cải thiện và gia tăng tuổi thọ của cầu bê tông cốt thép sử dụng UHPC

This study develops and evaluates machine learning (ML) models to predict the axial load capacity (Pu) of reinforced concrete (RC) columns strengthened with ultra-high-performance concrete (UHPC) jackets. A comprehensive experimental database containing 105 test samples with 17 key input parameters was compiled from the literature, representing the most extensive dataset of UHPC-jacketed RC columns to date. Using this database, a machine learning (ML) framework was established to predict the ultimate axial load capacity, employing six models: Extremely Randomized Trees (ER) model, K-Nearest Neighbors (KNN), Light Gradient Boosting Machine (LightGBM), Xgboost, CatBoost, and Cascade Forward Neural Networks (CFNNs). The CatBoost model achieved the best performance with R² = 0.983, MAE = 177 kN, and RMSE = 211 kN, significantly outperforming traditional design codes such as ACI 318 and EC2. In addition to high predictive accuracy, SHAP analysis was conducted to interpret the influence of each parameter, providing new insights into the mechanical behavior and governing factors of UHPC-jacketed RC columns. These findings highlight the capability of advanced ML to capture complex nonlinear effects more effectively than traditional methods. The proposed framework not only provides new insights into the mechanics of UHPC–RC columns but also offers a reliable predictive tool to support safer and more efficient design for strengthening.