Cross-correlation based Imaging of Defects in Plate using Ultrasonic Lamb Waves Debdutta GHOSH 1* , Surendra BENIWAL 2 , Abhijit GANGULI 3 , Abhijit MUKHERJEE 4 , Shashank BISHNOI 5 1 Structual Engineering Group, CSIR-Central Building Research Institute, Roorkee 247667, India 2 Department of Aeronautics and Astronautics, Stanford University, Stanford, CA 94305, USA 3 Department of Civil Engineering, Indian Institute of Technology Tirupati, Tirupati, India 4 Department of Civil Engineering, Curtin University, WA 6102, Australia 5 Department of Civil Engineering, Indian Institute of Technology Delhi, Delhi, India *debdutta@cbri.res.in Abstract Notch like defects may occur in thin metallic structures and components. These defects are often not detected due to their smaller size and if unattended may lead to leakages, causing industrial accidents and economic losses. Therefore, early detection of these defects is essential. The current work presents preliminary studies on an ultrasonic Lamb wave based imaging technique for detection of these defects. Experimental investigations are presented on a thin plate to demonstrate the developed technique. An A0 mode dominated Lamb wave field is generated using wedge technique which suppresses the higher Lamb wave modes. Notches of different sizes are created in the plate and are imaged in the plane of plate using the Planar Synthetic Aperture Focusing Technique (P-SAFT). The resulting images are improved by performing cross correlation between the received ultrasonic field and the incident A0 mode Lamb wave field. The SAFT images constructed from the cross- correlated waveforms show improvement over their counterparts generated through the conventional SAFT algorithm. Keywords: Ultrasonic, Lamb wave, SAFT, Cross-correlation 1. Introduction In the past few decades, plate structures are widely used in aerospace, civil or mechanical structures. Defects in these structures are very common in form of crack, delamination, corrosion and fatigue related damage etc. Therefore, regular inspection is essential to assure safety of these structures. Lamb wave based techniques, have been widely used for monitoring of plates [1-3] structures. The advantage of this wave is that, it propagates through the entire thickness of the thin plate and offer comprehensive information regarding the health of the structure [4, 8]. However, Lamb waves are dispersive and multiple modes are generated in the plate, which makes the propagating field highly complex. To eliminate the effect of dispersion and complexities in the received field, the applied excitation is limited to a frequency bandwidth, wherein the dispersion is minimal and only two fundamental modes (i.e., A0 and S0) are generated [5-7]. Application of a single mode Lamb wave for detection of defects in plates has been reported by several researchers [7-11]. The defects in a plate can be detected by studying the characteristics of the wave field reflected from the defect or the transmitted wave field that has propagated through the defect [8, 12-13]. Different signal processing and imaging based techniques viz. delay and sum [14], minimum variance [15], empirical mode decomposition [16], tomographic reconstruction [17] and several triangulation techniques [18] have been used to detect the defects in plates. The triangulation techniques require limited number of sensors, however, the resolution of the image is also limited. The tomographic reconstruction improves resolution of the images of the defects; however, a large number of sensors and a significant level of computational effort is necessary for its successful implementation. More info about this article: http://www.ndt.net/?id=24355