Master of Science in Engineering (MSE)
Civil and Environmental Engineering
First Committee Member
Aly M. Said
Second Committee Member
Pramen P. Shrestha
Third Committee Member
Fourth Committee Member
Number of Pages
The shear strength and deformation capacities of reinforced concrete (RC) columns are governed by a multitude of variables related to material properties of the steel and concrete used in the design and construction of the columns. Predicting performance of RC columns using design variables is a complex, non-linear problem. The prediction of shear strength and ductility for these types of structural members has historically been performed using empirically or semi-empirically derived formulae based on experimental results. The introduction of cyclical lateral loading, such as the forces imposed on a structure during an earthquake, can result in severe degradation of shear strength and ductility as load cycles continue. This can increase the complexity of predicting performance even further, as shear failure of the column occurs at relatively low deformations and can significantly affect the ability of the structure to resist lateral loading. Most existing models consider monotonic loading only and do not address this at all, which can result in extremely poor structural performance in a seismic event when compared to performance predictions.
Artificial neural network; Buildings, Reinforced concrete – Earthquake effects; Earthquakes; Genetic algorithm; Hysteresis; Lateral loads; Neural networks (Computer science); Seismic response; Shear (Mechanics); Shear failure
Civil Engineering | Geotechnical Engineering | Structural Materials
Gordon, Nicholas, "Prediction of Shear Strength and Ductility of Cyclically Loaded Reinforced Concrete Columns Using Artificial Intelligence" (2015). UNLV Theses, Dissertations, Professional Papers, and Capstones. 2356.
Available for download on Thursday, May 31, 2018