Crack identification by the electric potential CT inverse analyses incorporating optimization techniques

Abstract

The present authors proposed the electric potential CT (Computed Tomography) method for measuring two- and three-dimensional cracks in a body from the potential distributions observed on surfaces of the body. By using the boundary element equations and applying the concept of inverse approaches, two crack identification schemes were formulated the inverse boundary-integral-equation method and the least residual method. Numerical simulations and experiments showed that the proposed method was useful for the quantitative determination of location, size and shape of two- and three-dimensional cracks. In the present paper optimization techniques are incorporated in the least residual method to achieve an efficient determination of the location, size and shape of cracks. Vectorization of DO loops in the program of inverse analysis is also made to speed up the computation on a supercomputer. A numerical simulation on the identification of a three-dimensional corner crack demonstrates that an efficient and high-speed identification of cracks can be made by the inverse analysis incorporating optimization procedures on a supercomputer.