Two-dimensional curvature mode shape method based on wavelets and Teager energy for damage detection in plates

Vibration-based damage detection in plates has been investigated by various methods relying on mode shapes, among which the 20 curvature mode shape is a damage feature attracting much attention of researchers. Unlike the sound understanding of the use of the 1D curvature mode shape for detecting damage in beams, however, use of the 20 curvature mode shape to detect damage in plates is not yet well elucidated, major unresolved issues including lack of clarity about the mechanism of characterizing damage, susceptibility to noise, and insensitivity to sight damage. These deficiencies severely hamper use of the 20 curvature mode shape to portray damage in plates. To deal with these deficiencies, the mechanism of using 20 curvature mode shape to depict damage is analytically clarified in light of thin plate theory. On the basis of this clarification, a synergy between wavelet transform and a Teager energy operator is proposed to tackle the other deficiencies of susceptibility to noise and insensitivity to sight damage, leading an enhanced 20 curvature mode shape. The efficacy of the enhanced 20 curvature mode shape is numerically demonstrated using finite element simulations and experimentally validated through noncontact measurement by a scanning laser vibrometer, whereby its advances of clear mechanism of characterizing damage, robustness against noise, and sensitivity to slight damage are sufficiently corroborated. (C) 2015 Elsevier Ltd. All rights reserved.