The identification of voids and inclusions based on the parameter level set method using boundary data

Abstract

Using the temperature information obtained from the boundary to identify internal defects of the component is significant to ensure the reliability and safety of the structure. Based on the standard finite element method and the parameter level set method (PLSM), a new inversion theoretical framework for identifying voids and inclusions is established. A heat transfer solution model for multi-material interpolation with inclusion defects is constructed to avoid the iterative process of mesh reconstruction when the defect configuration changes. The strong nucleation ability of the parameter level set and the good optimization performance of the Levenberg-Marquardt method are fully utilized. The simple and complex defects are systematically discussed in the example section. By discussing different influencing factors such as the number of measuring points, the number of knots, the measurement error, it is found that even in the case of no some prior information, the proposed method can quickly obtain relatively stable and accurate inversion results.