Description: At the present time, nuclear pellet inspection is performed manually using naked eyes for judgment and decisionmaking on accepting or rejecting pellets. This current practice of pellet inspection is tedious and subject to inconsistencies and error. Furthermore, unnecessary re-fabrication of pellets is costly and the presence of low quality pellets in a fuel assembly is unacceptable. To improve the quality control in nuclear fuel fabrication plants, an automated pellet inspection system based on advanced techniques is needed. Such a system addresses the following concerns of the current manual inspection method: (1) the reliability of inspection due to typical human errors, (2) radiation exposure to the workers, and (3) speed of inspection and its economical impact. The goal of this research is to develop an automated nuclear fuel pellet inspection system which is based on pellet video (photographic) images and uses artificial intelligence techniques.
Date: November 1, 1999
Creator: Keyvan, S.
Item Type: Refine your search to only Report
Partner: UNT Libraries Government Documents Department