الفهرس 
Types of fiberOnly 14 pages are availabe for public view
 |  | from | 215 |  |  |
| | | from | 215 | | |
Abstract
Fiber reinforced polymers (FRP) are a composite material made of fibers embedded in a polymeric resin. Only recently has it been introduced as a suitable replacement for steel in the civil engineering field. Not only does it address the problem of electromagnetic interference resulting in avoiding steel corrosion but FRP materials also have several advantages among steel bars such as high tensile strength, high strength to weight ratio, high resistance to corrosion and fatigue that make them a suitable substitute for structural reinforcement.
Accurate prediction of the ultimate flexural strength of reinforced concrete beams strengthened by FRP is important for developing a reliable method. Several researchers and several empirical formulas based on experimental results have been developed; some of these formulas are used in design codes; some of them are not accurate. The capacity of strengthened beams is affected by several variables; such as FRP type, compressive strength of concrete, the reinforcement ratio, the type of used FRP and end anchorage. Hence, it was proposed to use neural network analysis, to count for all effective parameters. The neural network (NN) model was developed using previous experimental data on flexural failure of RC beams strengthened by carbon, glass and aramid FRP. It was concluded that NN can predict, to a good degree of accuracy, the ultimate moment capacity and the debonding strain of RC beams strengthened by FRP. NN model results were compared to the results predicted by the models in the American, Canadian, Italian and Egyptian code equations.
Abstract
ii
However, as the bond length effect on the debonding strain is not taken into consideration in current Egyptian Code equation, a new model was developed using the neural network analysis to count for the bond length effect.
Keywords: FRP; RC beams; strengthening; bond characteristics; bond length; composite materials; neural networks