Volume 24, Issue 4 (2024)
MCEJ 2024, 24(4): 83-94 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Saeed R, Moradloo A J. A New Method for Predicting the Seismic Behavior of Concrete Dams by Considering Lift Joints. MCEJ 2024; 24 (4) :83-94
URL: http://mcej.modares.ac.ir/article-16-72705-en.html
URL: http://mcej.modares.ac.ir/article-16-72705-en.html
1- PhD Candidate, Department of Civil Engineering, Faculty of Engineering, University of Zanjan, Zanjan, Iran. , rezvan.saeed@znu.ac.ir
2- Associate Professor, Department of Civil Engineering, Faculty of Engineering, University of Zanjan, Zanjan, Iran.
2- Associate Professor, Department of Civil Engineering, Faculty of Engineering, University of Zanjan, Zanjan, Iran.
Abstract: (408 Views)
Dams are structures whose continuous evaluation is of great importance. Due to their large scale, experimental study of concrete dams is difficult and therefore, the numerical simulation is used in the dynamic analysis of such dams more effectively. Despite the widespread use of concrete, our knowledge on its exact properties and physical behavior under different conditions is still limited, and many assumptions and simplifications are made to study the concrete behavior in most studies. This is especially complicated in mass concrete structures such as concrete dams. The presence of joints in most concrete structures is common and inevitable. Lift joints in dams cause different characteristics in vertical and horizontal planes. In fact, this is a special type of anisotropy that follows axial symmetry with respect to any vertical axis, which means that the mechanical behavior is the same in all horizontal planes. The mechanical behavior in all vertical planes passing through the axis of symmetry is also the same, however, it is different from the behavior of horizontal planes. Since the lift joints are usually ignored in the numerical analyzes of concrete dams, in the present paper, taking into account the orthotropic behavior of concrete, the concreting joints that cause weakness in specific positions and directions of the dam body are included. First, non-linear seismic analyzes were performed using FEAP finite element software, then a Fortran program was coded to predict the time history of displacement. The proposed method draws upon evolutionary algorithms inspired by Darwinian biology, which are increasingly utilized as surrogate models for various analyses. This approach relies on data-driven learning, wherein algorithms, based on training or sample data, generate a mathematical model for making predictions. The Pine Flat dam was modeled and analyzed under the Taft earthquake loading over a 20 second time interval with 0.02 second time steps. After successful training and learning, the model was compared and tested for other anisotropy ratios. The purpose of developing the program was to reduce the time required for analyzes so that by analyzing the initial seconds of seismic loading, by importing training inputs to the program, a proper prediction of the response process for the rest of the loading time could be obtained. In addition, by training the program for the isotropic and orthotropic modes, time history diagrams could be extracted for other orthotropic modes in different anisotropy ratios. According to the obtained results, the program is acceptably able to predict the graphs in a very short time. In addition, an equation for predicting the displacements in the orthotropic mode is presented. The maximum displacement of the orthotropic analysis was more than the isotropic one, and the use of isotropic material and homogeneous modeling of the dam body caused errors in the results. Therefore, considering the orthotropic properties of concrete can lead to more realistic results. The results reveal that time history plots derived from the implemented program closely resemble those from finite element analyses. The output results are remarkable, given the significantly reduced time required for predictions generated by the implemented program.
Article Type: Original Research |
Subject:
Hydraulical Structures
Received: 2023/12/1 | Accepted: 2024/02/28 | Published: 2024/10/1
Received: 2023/12/1 | Accepted: 2024/02/28 | Published: 2024/10/1
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |