Detection and localization of damage in fiber reinforced polymer bars using acoustic emission detection, micro computed tomography, and scanning electron microscopy techniques

Loading...
Thumbnail Image
Date
2018-08-15
Authors
Ghaib, Maha
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Fiber reinforced polymer (FRP) rebars have been used for reinforcing concrete structures for the last quarter of the century. The advantages of using FRP are: high strength to weight ratio, corrosion resistance, durability, and high tensile strength. It would be beneficial to have non-destructive testing (NDT) methods that have the potential to detect damage at loads well below failure. This is particularly important for FRP rebars as they do not exhibit any external signs of damage until brittle failure. However, internal damage in FRP rebars increases with increasing tensile loads. Acoustic emission (AE) signal has proved to be a useful tool for monitoring damage in FRP materials. However, FRP rebars have received little attention. In this work damage progression in different types of FRP rebars subjected to tension was studied using AE, SEM and µCT. One possible mechanism for damage progression is the growth of voids entrapped in FRP rebars during their manufacture. These voids are a weakened region within the bars. When the bar is subjected to stress these regions can grow in size. In this work the growth of the voids within the bar is interpreted as the progression of damage. Internal voids and their growth were observed using micro computed tomography (µCT) and scanning electron microscopy (SEM) and compared to AE results. Strong correlation was observed between the void volume growth determined using µCT analysis and cumulative energy of AE. This work also demonstrated improved methods used to locate the source of AE signals that occurred in FRP rebars subjected to tension. An artificial neural network (ANN) was used to reduce the uncertainty in determining the AE source location. The ANN had improved accuracy, compared to the conventional source location methods by factor of 1.6 and 3 for GFRP and CFRP samples respectively. The source location results obtained with the ANN were also compared to µCT void volume analysis along the length of the bars after tension testing.
Description
Keywords
Acoustic Emission, Carbon fiber reinforced polymer, Glass fiber reinforced polymer, Scanning Electron Microscopy, Micro Computed Tomography, Artificial Neural Networks
Citation