https://ceej.aut.ac.ir/ Amirkabir Journal of Civil Engineering en daily 1 Fri, 20 Feb 2026 00:00:00 +0330 Fri, 20 Feb 2026 00:00:00 +0330 https://ceej.aut.ac.ir/article_6005.html Piano key weirs are a new type of non-linear labyrinth weirs. Due to the high efficiency of these weirs in flow passage, it is of great importance to investigate local scour and the solution to reduce it. In the present study, different block geometries were used in the tunneled outlet keys of the rectangular piano key weir type B. The piano key weir used is 0.20 m high and has three cycles (three outlet keys, two inlet keys, and two inlet half keys). Rectangular, trapezoidal, and cylindrical block geometries were installed in each outlet key of the weir and on the tunnel structure. The tunnel structure prevented the flow exiting from the inlet and outlet keys from mixing. The blocks direct the flow and the maximum scour depth to a distance further from the weir toe, acting as a barrier to the flow and reducing the velocity of the flow exiting from the keys. A further distance of the maximum scour depth from the weir toe can reduce the risk of the weir overturning. In rectangular blocks, the maximum scour depth decreases, and its location is further from the weir toe. Also, as the densimetric Froude number increases, the flow rate rises and the tailwater decreases; the maximum scour depth also increases. The range of the densimetric Froude number in the present study varies between 0.43 and 0.55. Finally, dimensional analysis was used to generalize the results to nature and other models of piano key weirs. https://ceej.aut.ac.ir/article_6006.html This research proposes a hybrid numerical method, for modeling the development of artificial ground freezing in soil containing subsurface inhomogeneities. In this approach, by combining the Boundary Element Method (BEM) and employing time-independent fundamental solutions, appropriate boundary integral equations are developed. The volume integrals arising from the presence of dynamic terms are incorporated into the equations using the Finite Element Method (FEM). For this purpose, a type of boundary-finite element, combining a quadratic boundary element with a three-node triangular finite element, was developed, and the solvable forms of the final equations were presented. Subsequently, by implementing the hybrid method into a computational algorithm, its accuracy and efficiency was evaluated and validated by solving several benchmark examples. Finally, in a parametric study, the application of the hybrid method for modeling the development of artificial freezing in saturated soil containing a inhomogeneities in the form of an unsaturated rock mass is described. The effects of varying the cross-sectional area of the inhomogeneities and its distance from the freeze pipe were evaluated. The results of the parametric study indicate that the presence of the inhomogeneities reduces the volume of the freeze bulb by up to 20%. Furthermore, inhomogeneities with a circular cross-section were more limited the development of freezing compared to a square shape. https://ceej.aut.ac.ir/article_6011.html This study investigates the effectiveness of non-destructive ultrasonic testing for detecting and classifying damage in cement mortar specimens with different mix designs. To this end, a set of specimens with varying cement content (CC) were prepared and tested in both undamaged and damaged states. Ultrasonic signal data, after preprocessing and the extraction of statistical and time–frequency features, were used as inputs to three machine-learning algorithms: K-Nearest Neighbors (KNN), Naïve Bayes, and Random Forest. Analysis results indicated that the Naïve Bayes classifier, owing to its ability to handle independent and uncorrelated features of the ultrasonic data, outperformed the other methods and yielded the highest classification performance, achieving an accuracy of 99.00 ± 3.16% and a recall of 97.50 ± 7.91%. The main innovation of this research is the combination of machine-learning approaches with non-destructive methods to analyze energy variations in ultrasonic signals and to enable early damage detection in cement mortars with different cement content ratios. This data-driven, reproducible analytical framework not only improves the accuracy and reliability of damage identification but is also extensible for health monitoring and continuous assessment of concrete structures at practical scales, serving as an effective tool to enhance the durability and safety of civil infrastructure. https://ceej.aut.ac.ir/article_6013.html In this study, a real-time adaptive control framework with a two-layer architecture is proposed for mitigating the seismic responses of tall buildings exhibiting nonlinear behavior. The proposed structure consists of a baseline adaptive controller and a supervisory hyper-adaptive layer that adjusts the adaptation parameters online using a filtered extremum-seeking algorithm. The baseline controller generates the control signal based on output-feedback adaptive laws, while the hyper-adaptive layer determines the direction and rate of updating the controller gains by estimating the instantaneous approximate gradient of the performance function. This mechanism, without relying on an accurate structural model or requiring prior identification, enables dynamic adaptation to sudden or gradual changes in system dynamics, nonlinear behaviors such as hysteresis, and structural uncertainties. The iRT-SAC framework is designed to guarantee boundedness of the adaptive parameters and closed-loop stability under challenging operating conditions, while minimizing the need for manual retuning or remodeling. To evaluate its performance, the proposed method is implemented on the 20-story benchmark structure of Ohtori et al. and compared with reference controllers including LQG, H_∞, fuzzy control, and Clipped-LQG. The results demonstrate approximately a 60% reduction in interstory drift and preservation of closed-loop stability in the presence of severe uncertainties and nonlinear behavior. Overall, iRT-SAC represents a model-light, robust, and practically deployable adaptive approach for tall and complex structures. https://ceej.aut.ac.ir/article_6016.html In recent years, the use of steel-reinforced concrete (SRC) columns, including composite connections, in high-rise structures has increased significantly. One of the key issues in connection design is the proper detailing of the interface between the steel section embedded in the concrete column and the reinforced concrete beam. This study experimentally investigates the seismic performance of the connection between a reinforced concrete beam with a transition steel part (TP) and an SRC column, and compares it with other types of connections. Three experimental specimens were constructed: (i) a reinforced concrete beam (RC)–column joint, (ii) an reinforced concrete beam–steel reinforced concrete (SRC) column joint, and (iii) an reinforced concrete beam with a transition part (TP)–SRC column joint. All specimens were subjected to cyclic lateral loading on the beams and axial loading on the columns. The main experimental parameters include the effects of the steel section on load-bearing capacity, ductility, and stiffness, as well as the influence of the transitional part on ductility and stiffness. The results indicate that the incorporation of the TP substantially enhances the performance of the beam–column joint. Specifically, lateral load capacity increased by 1% and 15% in tension, and by 20% and 8% in compression, compared to the reinforced concrete beam–column and reinforced concrete beam–SRC column specimens, respectively. Additionally, ductility improved by 44% and 24%, respectively, compared to the same specimens. The results highlight the effectiveness of the TP in enhancing the load-bearing capacity and ductility of RC beam–SRC column joints. https://ceej.aut.ac.ir/article_6017.html In this study, the damage of a type of marble under uniaxial compressive loading was investigated using grain-based finite element simulations and a damage constitutive model. Accordingly, 40 Voronoi networks with different structural configurations and low and high joint densities were generated using the Phase2 software. Then, based on the calibrated parameters, uniaxial compression tests were numerically simulated on the generated networks. Finally, by utilizing the stress-strain curves derived from the numerical simulations and extracting the initial and damage moduli, the damage variable values were calculated as a function of strain. The results indicated that with an increase in the Voronoi joint density of the networks, the average values of both the initial and damage moduli decreased. Next, it was observed that with increasing strain up to a certain threshold, the damage variable increased gradually with a gentle slope. However, after reaching a critical strain value, the damage variable increased sharply with a very steep slope. This point, at which the damage begins to increase suddenly and rapidly, represents the onset of significant damage experienced by the rock specimen under uniaxial compressive conditions. Moreover, analysis of the failure patterns showed that tensile failures in networks with high Voronoi density occurred more frequently than in those with low density. Therefore, it can be concluded that the internal grain-based structure of rocks plays a crucial role in determining the extent of internal damage, failure patterns, and the dominant failure modes of rocks under uniaxial compressive loading conditions. https://ceej.aut.ac.ir/article_6023.html Both hydroxylation and surface oxidation have primary roles in steel nanofiber mechanical properties. The chemical reactions, resulting from contact with water vapor, oxygen, and corrosive environments, lead to the alteration of the atomic composition of the surface of the nanofiber and form layers whose properties are different from pure steel. In the present paper, a study of the effects of these processes on compressive and tensile mechanical properties of steel nanofibers through molecular dynamics methods using the reactive force field potential (ReaxFF) and core-shell modeling approach has been discussed. Simulations are performed using LAMMPS software with a quasi-static incremental loading scheme to minimize dynamic stresses. It has been found that higher oxide layer thickness reduces Young's modulus, yield stress, and ultimate strength of the nanofibers. Most notably, a 20% oxide layer thickening can reduce Young's modulus by up to 40% and yield stress by up to 34%. Hydroxylation causes these values to become even lower due to its ability to create weaker and less stable bonds. The analysis of the stress-strain curve indicates that the layers of oxide and hydroxide facilitate stress concentration and increase material failure. Experimental evidence corroborates the simulation results and demonstrates the high accuracy of the numerical model. The findings of the present study indicate that when steel nanofibers are exposed to the alkaline condition of concrete, widespread yield stress and Young's modulus reduction will be witnessed, which should be accounted for in the application of such components. https://ceej.aut.ac.ir/article_6024.html This study proposes an integrated framework that combines dynamic modeling, numerical optimization, and machine learning classification to predict the optimal design parameters of Tuned Liquid Mass Dampers (TLMDs). Two primary outputs—the optimal frequency ratio and optimal damping ratio—were analyzed using six classification models: Logistic Regression, Decision Tree, Random Forest, K Nearest Neighbors (KNN), Support Vector Machine (SVM), and Naive Bayes. Two structural configurations were examined: a single-story and a five-story shear building, each equipped with rooftop TLMDs mounted on elastomeric pads. Dynamic responses were obtained for six earthquake records using time history analysis, with liquid motion modeled by the Housner model. Optimal elastomeric pad parameters for various tank configurations were determined via the Pattern Search algorithm. The results revealed that for the optimal frequency ratio in the single-story structure, KNN and Random Forest achieved the highest F1 score (~0.73), whereas in the five-story building, prediction accuracy declined and Naive Bayes performed best (~0.68). Regarding the optimal damping ratio, Naive Bayes excelled in both structures, particularly in the five-story model. Confusion matrix analysis indicated that most errors occurred in the intermediate class, primarily due to feature overlap. By significantly reducing computational time and eliminating the need for exhaustive numerical simulations, the proposed data-driven methodology supports reliable decision-making in both preliminary and detailed stages of TLMD design. Moreover, the framework is extendable to other passive vibration control devices and more complex structural systems, advancing the concept of intelligent, efficient, and precise design tools in structural engineering. https://ceej.aut.ac.ir/article_6026.html Fly ash is produced as a byproduct of the coal combustion process in thermal power plants. Fly ash consists of very fine and microscopic particles, typically composed of mineral compounds such as silicon dioxide, aluminum oxide, and iron oxide. These compounds make fly ash suitable for use in various industries, particularly in the construction industry. Applications of fly ash include additives in concrete, fillers in asphalt, production of bricks and concrete blocks, and pollutant absorption. As a pozzolanic material, fly ash helps reduce carbon dioxide emissions in the cement production process. In this study, a comprehensive database of previous studies on fly ash concrete was initially collected. This data included 599 samples from credible laboratory studies. The gathered dataset consisted of various input variables, including the water-to-cement ratio, amount of fly ash, cement content, coarse aggregate amount, fine aggregate amount, superplasticizer content, and curing age of the concrete. To predict the compressive strength of the concrete, various machine learning algorithms were utilized, including Genetic Programming (GP), Adaptive Neuro-Fuzzy Inference System (ANFIS), Multi-Layer Perceptron (MLP), Radial Basis Function Neural Network (RBF), Kriging, and Extreme Learning Machine (ELM). Furthermore, the accuracy of each model was evaluated using statistical indices, and the best model was identified. The results show that different machine learning models exhibit varying performances in predicting compressive strength. In particular, the Kriging method, with a correlation coefficient of 0.96, was selected as the best model.