Assessment of damage evolution of concrete beams strengthened with BFRP sheets with acoustic emission and unsupervised machine learning

Abstract

Basalt fiber reinforced polymer (BFRP) is an emerging material that can be used to strengthen or repair concrete structures. This paper experimentally investigates the flexural failure process of concrete beams strengthened by BFRP sheets under four-point loading conditions using acoustic emission (AE)-based nondestructive evaluation and unsupervised machine learning techniques. Both damaged beams and repaired beams were tested and compared with intact beams regarding their mechanical performance and AE signature. Test results showed that BFRP sheets enhanced the load-bearing capacity of the beams significantly, and AE signals were sensitive to the initiation and development of damages. The damage process and mechanisms of different beams have been investigated by analyzing AE signals regarding various AE parameters, and particularly, the failure process and damage mechanisms were investigated via b-value and k-means clustering analyses. This research provides theoretical guidance for evaluating the reinforcing effect of BFRP sheets for concrete beams based on AE techniques.