Integrated Inspection Solution using X-Ray Computed Tomography for Nuclear Power Plant Maintenance

Abstract

Nuclear reactors need regular maintenance using specific maintenance tools. This paper presents an integrated and automated inspection solution for tools used in reactor maintenance. The proposed solution is based on the utilization of X-ray Computed Tomography (CT) to scan the maintenance tools before and after utilization in reactor maintenance and detect any defect. The proposed automated solution is called the CT-based Integrity Monitoring System (CTIMS). The proposed solution includes 2D and 3D algorithms and deep learning models to analyze images obtained from the CT scans and perform a series of preprocessing to align different scans of the same tool despite differences in tool placement. The solution offers human-in-the-loop features to integrate human experience and annotation and labeling of different parts of the tool. The database is used to manage tools, scans, and defects details. The automated defect detection algorithms detect defects such as missing parts, damaged screws, or shifted components.