International Concrete Abstracts Portal

The structural deterioration of aging infrastructure systems is becoming an increasingly important issue worldwide. To compound the problem, economic strains limit the resources available for the repair or replacement of such systems. Over the past several decades, structural health monitoring (SHM) has proven to be a cost-effective method for the detection and evaluation of damage in structures. Visual inspection and condition rating is one of the most commonly applied SHM techniques, but the effectiveness of SHM varies depending on the availability and experience of qualified personnel and largely qualitative damage evaluations. Simply supported three-dimensional reinforced concrete T-beams with varying geometric, material, and cracking properties were modeled using Abaqus finite element (FE) analysis software. Up to five cracks were considered in each beam, and the ratios of stiffness between cracked and healthy beams with the same geometric and material parameters were measured at nine equidistant nodes along the beam. A feedforward ANN utilizing backpropagation learning algorithms was then trained on the FE model database with beam properties and nodal stiffness ratios serving as inputs for the neural network model. The outputs consisted of the predicted parameters of location, depth, and width of up to five cracks. This inverse problem is very difficult or impossible to solve with the training done by the Artificial Neural Network. One ANN was trained to predict the parameters of the cracks using the full database of FE simulations. The damage prediction ANN achieved fair prediction accuracies, with coefficients of determination (R2) equal to 0.42. This result was the outcome of the no uniqueness in the prediction of this inverse analysis. Nevertheless, this ANN model provides a rough estimate of the cracking type and damage content in bridge girders once the nodal stiffness ratios are measured by applying a field vehicle loading and measuring the deflection using a theodolite. A touch-enabled user interface was developed to allow the ANN model to predict the crack configurations. The application was given the acronym DRY BEAM, for Damage Recognition Yielding Bridge Evaluation After Monitoring.

Many modeling approaches in engineering are based on physical principles. The input and output relationships are developed using physical laws (e.g., Newton's laws of motion and conservation of mass and energy). However, in many situations, the development of physically-based models requires simplifying assumptions due to the complicated nature of the systems, which could lead to a large degree of uncertainty. In these situations, data can be used to formulate models by detecting relationships between the system’s variables (inputs and outputs) without explicitly knowing the physical behavior of the system. Therefore, there is a paradigm shift from physically-based models to data-driven models. The objective of this study is to develop a drift capacity prediction model for structural walls with special boundary elements using the extreme gradient boosting (XGBoost) machine learning algorithm. The resulting prediction model is compared with the recently developed empirical model presented in literature i.e., the Abdullah & Wallace (2019) model. The results reveal the proposed model’s superior predictive capabilities relative to the empirical model.

This paper presents the development of a data-driven deep neural network (DNN) for evaluating the fire resistance time of fiber-reinforced polymer (FRP) strengthened concrete beams. The model was trained for a scaled and unscaled dataset. For this, a comprehensive dataset of FRP-strengthened concrete beams with different geometry, insulation configuration, applied loading, and material characteristics was compiled. The DNN structure was selected after an extensive hyperparameter tuning in conjunction with ten-fold cross-validation scheme. The effect of different input parameters on the fire resistance prediction was analyzed. The DNN model developed using scaled data provides a reasonably accurate estimate, of the fire resistance of FRP-strengthened concrete beams with an R2 value of almost 92%. The developed model is further utilized to evaluate the impact of different parameters on fire resistance prediction for FRP-strengthened concrete beams. Results from the analysis indicate the thermal properties of insulation play an important role in determining the fire resistance of FRP-strengthened concrete beams.

The bond between reinforcing bars and concrete is an important property that determines the performance of reinforced concrete structures. Accurate prediction of the bond strength is essential for ensuring the safety and economy of the structures. This paper proposes an artificial neural network for predicting the bond strength between straight deformed reinforcing bars and concrete under tensile load. The neural network was trained using a large dataset of test results from the ACI Committee 408 database. A robust ten-fold cross-validation method was employed for evaluating network performance and finding optimal network parameters. Hyperparameter tuning was carried out to establish the optimal network hyperparameters. The relative impact of the neural network input parameters on the bond strength was evaluated using the SHAP method. The developed neural network with the optimal hyperparameters shows a good agreement with the test results. Its accuracy exceeds the accuracy of the descriptive equations.

Artificial Intelligence has one of the most efficient methods for solving engineering and materials problems because of its impressive performance and can reach higher accuracy. The Deep Learning theory is an approach based on Deep Neural Networks for establishing numerical analysis and value predictions. This paper proposes a fresh approach, using a Deep Learning model for predicting the compressive strength in a particular concrete just based on non-destructive test measurements (NDTs). The model proposed is an attractive alternative to estimate the resistance of compressive strength in any structure, just taking data like ultrasonic pulse velocity, electrical resistivity, and resonance frequencies. The present work employs data science techniques to find the correlation values between the NDTs and the compressive strength effort and realized broad numerical exploration about concrete performance. An amount of 285 specimens of concrete were monitored during this research. The model proposed contains 600 neurons and uses a Rectified Linear Unit and Sigmoid as activation functions where the NDTs were established as the input data. The dataset was segmented into two groups: train and test. In order to evaluate the model, the authors tested it in a validation set with different concrete features, achieving an accuracy of 94%.