Industrial System Development for Volumetric Integrity Verification and Analysis

Currently, internal integrity verifications of a part are achieved on a slice-by-slice basis. Due to the curvy nature of the part and because flaw features can spread in any location and in any direction, 2-D image analysis may produce inaccurate measurements and low-sensitivity flaw detectability which can lead to ambiguities in decision making. By utilizing the 3-D spatial correlations presented in the volumetric data set, accurate measurement and enhanced flaw detectability can be achieved by volumetric processing and analysis. For instance, cracks are very difficult to detect using 2-D processing when X-ray projection plane cuts through the cracking direction of an elongated flaw as shown in Figure 2(a). This is due to the low signal-to-noise ratio of the data on a single slice basis. However, because the signals are correlated in the complete 3-D space, enhanced flaw detectability can be easily obtained using 3-D volumetric processing and analysis. In addition, accurate wall thickness measurements can be achieved only when the normal of the projection plane is perpendicular to the surface normal of the inspected wall. However, in practice, this requirement cannot be met easily due to the curvy nature of the part. Inaccurate measurements may result when wall thickness measurements are performed on a slice-by-slice basis, as shown in Figure 2(b). In order to achieve full-coverage inspection to ensure material integrity and geometry accuracy, true 3-D analysis is required. A modality-independent voxel processing demonstrator has been designed and developed to achieve full-volume inspection. A three-dimensional voxel organized data structure can be constructed from a stack of two-dimensional CT slices, reconstructed from a three-dimensional cone-beam configuration, or constructed from three-dimensional ultrasonic wave data.