Monitoring crack propagation in reinforced concrete shear walls by acoustic emission

In the last two decades, several efforts have been made to monitor the cracking behavior in RC structures. A technique that shows promise is acoustic emission (AE). This paper presents the results of an experimental study aimed at monitoring fracture processes in a large-scale RC shear wall using one of the most important AE parameters, that is, the b-value. The specimen was subjected to a displacement controlled reversed cyclic loading. A Gaussian filter is proposed to improve the interpretation of b-value data obtained during the test. In addition, a cluster analysis based on the k-means is presented to automatically classify the signal into tensile and shear cluster. Finally, a new algorithm called Sifted b-value (Sb) analysis is introduced to monitor the evolution of each crack mode. The proposed approach is capable to identify the initial yielding and eventually provide an early warning for the planning and implementation of remedial action to the structure at a point at which it is less expensive and invasive than when its structural performance has been seriously compromised.