Show
Abstract
Abstract Details
Prediction of Bond Strength Between Fiber Reinforced Polymers and Concrete Using Artificial Neural Network
Major Topic: Concrete Structures|سازه های بتنی
Abstract
Application of sheets of Fiber Reinforced Polymers (FRP) for external reinforcement of Reinforced Concrete (RC) is a common strengthening method these days. The FRP sheets widely replaced traditional steel plates for external reinforcement of RC structures due to their various positive characteristics, most importantly the relatively high bonding strength between FRP and concrete, which is a very important factor at joints. Various studies have been performed on FRP strengthening of RC members and different models are proposed to estimate the bond strength between FRP and concrete. These predictive models estimate bond strength as a factor of different parameters such as concrete compressive strength, concrete and FRP moduli of elasticity, width of RC member and FRP sheet and etc. However, some of these relations do not provide enough accurate estimations. The main objective of the present study is to develop predictive model using Artificial Neural Networks (ANN) for prediction of bond strength. For this purpose, a relatively large database of experimental results is collected from the literature and used for training the ANN. The accuracy of the predictions made by the developed network is evaluated using part of the collected experimental database. The attained results prove that the ANN is an effective method for prediction of FRP-to-concrete bond strength that provide high accuracy and lower model error ( ) compared to many of the existing predictive models.
Keywords
Bond strength; FRP; Tensile test; Bonding stress; Artificial Neural Network