内蒙古电力技术2025，Vol.43Issue(3)：39-43,5.DOI:10.19929/j.cnki.nmgdljs.2025.0033
人工智能技术在辐照后核燃料组件外观目视检查中的应用
Application of Artificial Intelligence Technology in Visual Inspection of Irradiated Nuclear Fuel Assembly Appearance
摘要
Abstract
In order to improve the accuracy of the appearance inspection of nuclear fuel assemblies and ensure the safety of nuclear fuel,artificial intelligence(AI)technology is used to develop a defect recognition system,and AI technology is applied to the visual inspection of the appearance of nuclear fuel assemblies.The system uses a generative image template method to remove thermal turbulence,and enhances the effective information in the image through an image enhancement algorithm.During the inspection,the defect samples are uploaded to the cloud computing platform for image training,and the training model is obtained.The automatic recognition of defects is realized by machine vision.The system can identify and classify six typical defects and display confidence.In order to further verify the effectiveness and accuracy of the proposed method,this method is used to inspect the appearance of nuclear fuel assemblies during the overhaul of a third-generation nuclear power station in China.The results show that the method can deal with complex and changeable environments such as high irradiation and thermal disturbance,and has high real-time detection accuracy,which can meet the visual inspection requirements of nuclear fuel assemblies during the overhaul of nuclear power station.关键词
核燃料检查/人工智能/缺陷识别/图像增强/图像训练/目视检查/机器视觉Key words
nuclear fuel inspection/artificial intelligence(AI)/defect identification/image enhancement/image training/visual inspection/machine vision分类
能源科技引用本文复制引用
章济..人工智能技术在辐照后核燃料组件外观目视检查中的应用[J].内蒙古电力技术,2025,43(3):39-43,5.基金项目
国核电站运行服务技术有限公司科技项目"燃料组件四面外观检查控制显示与识别系统"(GHYX-NB-2011) （GHYX-NB-2011）
