Ultrasonic Geometrical Full Waveform Inversion of Defects for Nondestructive Evaluation
Abstract
Ultrasonic characterization of internal defects, such as cracks and voids, provides vital information for evaluating the remaining lifetime of critical engineering components used in the nuclear and aerospace industry. Various inverse imaging methods have been applied including total focusing method, diffraction tomography and parametric manifold mapping. In this talk, the speaker and his collaborator propose a geometrical full waveform inversion (GFWI) method to reconstruct defects. Different from conventional full waveform inversion (FWI) aiming to recover a spatial map of material properties of the subsurface medium, GFWI seeks to recover the shape of internal defects (stress-free boundary condition) by minimizing the mismatch between the synthetic and measurement data. GFWI is based on a carefully designed formulation to calculate the deformation of the defect’s boundary, following its negative geometrical boundary gradient. The defect will gradually evolve to the true geometry by distorting the shapes of elements attached at the boundary. Multiple graphical processing units (GPUs) are executed to significantly enhance the efficiency of inversion. Both numerical and experimental results using an ultrasonic phased array will be given. In addition, defects are often found located near the boundary of the engineering component where the stresses are highly concentrated. The speaker will show that multiple scattering events can be utilized automatically during the optimization to enhance the inversion.
About the speaker
Prof. Fan Shi obtained his MS and PhD, both in Ultrasonic Nondestructive Evaluation, from the Georgia Institute of Technology in 2012 and the Imperial College London in 2015 respectively. He continued his research as a Postdoctoral Research Associate at the Non-Destructive Evaluation (NDE) Research Group at the Imperial College London before joining the Hong Kong University of Science and Technology (HKUST) in 2019. He is currently an Assistant Professor in the Department of Mechanical and Aerospace Engineering at HKUST.
Prof. Shi has strong motivation towards the solution of real industrial problems for ultrasonic NDE, achieved through rigorous research of the fundamental wave phenomena. His current research interests include ultrasonic imaging, NDE, elastic wave scattering, material characterisation, ultrasonic guided wave and metamaterials. He has extensive collaborations with nuclear industries including EDF Energy, Rolls-Royce Submarines, BAE Systems, John Wood Group and the UK National Nuclear Laboratory, to solve practical NDE problems and improve inspection reliability and qualification.
About the program
For more information, please refer to the program website at http://iasprogram.ust.hk/inverseproblems.
