الفهرس 
CHAPTER (1) 1 INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 180 |  |  |
| | | from | 180 | | |
Abstract
2 ABSTRACT
This research investigates the shear capacity of wide-shallow reinforced concrete beams, often referred to as wide, shallow, or banded beams, recognized for their application in ribbed slabs for efficient formwork. The study sets out to provide a comprehensive overview and identify the gaps in understanding the considered beam type, using six design codes and around fifty prior research endeavors. Despite discrepancies in code definitions, a prevalent neglect of shear reinforcement contribution is observed. The analysis reveals a lack of theoretical studies and predictive models, especially concerning Finite Element Method (FEM) and Artificial Intelligence (AI) applications.
To experimentally explore shear capacity, fourteen specimens, including a control, undergo three-point load testing, with varying parameters such as aspect ratio, concrete strength, and compression steel ratio. Notably, shear reinforcement proves crucial, contributing a minimum of 55% to the total shear capacity. Aspect ratio and concrete strength significantly influence shear capacity, while the compression steel ratio enhances stiffness and ductility.
Further investigations into shear reinforcement amount and arrangement were conducted. Results emphasize the pivotal role of shear reinforcement, with transverse spacing influencing shear capacity more than longitudinal spacing. Design code comparisons yield varying degrees of accuracy, with the Japanese code (JSCE) closest to experimental outcomes. Moreover, Genetic Programming (GP) and Evolutionary Polynomial Regression (EPR) predictive equations exhibit accuracies of 78% and 86%, respectively.
Transitioning to predictive methodologies, the study leverages Finite Element Analysis (FEA) and machine learning, validating a detailed FEM against experimental data and conducting a parametric study. Three AI-based prediction equations GP, EPR, and Artificial Neural Network (ANN) are developed. Comparative analysis underscores the superior accuracy of the ANN model, exceeding 99%. Sensitivity analysis highlights the influential role of concrete strength and beam aspect ratio on shear strength. In conclusion, this research demonstrates the synergy of FEA and machine learning in predicting shear strength in wide-shallow reinforced concrete beams, providing valuable insights for architectural and engineering practices. The findings underscore the significance of concrete strength and beam geometry in understanding shear behavior, enriching the foundation for structural design.