Papers »

Application of acoustic emission and data mining for damage detection in power plants

Concrete shear walls are critical structural components in gravity and lateral force resisting systems. The objective of this work is to design and validate a monitoring system capable of rapid and automated damage assessment of reinforced concrete (RC) and steel-concrete (SC) composite shear walls. Monitoring the development of cracks is of large interest because their properties reflect not only the condition of concrete as material but also the condition of the entire system at structural level. The proposed system is based on a sparse array of piezoelectric transducers distributed across the wall to receive acoustic emissions (AE). The AE data are processed using advanced pattern recognition algorithms in conjunction with conventional crack mode classification to determine the so called latent clusters of shear and tensile. Furthermore, an acoustic imaging technique is developed to infer the most likely location of invisible damage behind the steel faceplates. The proposed system has been validated on two large-scale RC/SC shear walls designed for nuclear safety applications. It informs decision makers on the need for repair to ensure safe operation of the structure, minimizes the maintenance costs, and increases the operation lifetime.

Author(s):

Alireza Farhidzadeh    
Mistras
United States

Obdulia Lay    
United States

David Sinay    
United States

Richard Gostautas    
United States