基于大数据的燃料棒破损诊断方法研究  

Research on Fuel Rod Damage Diagnosis Method Based on Big Data

作　　者：景福庭[1] 吕焕文[1] 唐松乾[1] 魏江林 李兰[1] 夏明明 Jing Futing;Lyu Huanwen;Tang Songqian;Wei Jianglin;Li Lan;Xia Mingming(National Key Laboratory of Nuclear Reactor Technology,Nuclear Power Institute of China,Chengdu,610213,China)

机构地区：[1]中国核动力研究设计院核反应堆技术全国重点实验室,成都610213

出　　处：《核动力工程》2024年第S2期150-155,共6页Nuclear Power Engineering

摘　　要：核电厂燃料棒破损诊断(FRDD)是核电业主和核安全监管关注的重要问题。将大数据和近邻算法用于燃料棒破损诊断,并开发了压水堆燃料棒破损诊断软件,采用核电厂燃料棒破损的运行实例和理论例题进行了软件验证。验证结果为:①燃料棒破口尺寸类别分析方面,80%的分析结果与例题保持一致;②破损燃料棒数目分析方面,分析结果与例题的最大偏差为1根。基于大数据和近邻算法的压水堆燃料棒破损诊断方法,可以给出更接近实际破损情形的诊断结果,及时发现燃料棒的破损以及破损状态的变化,为反应堆燃料棒破损后的运行决策和辐射防护提供可靠依据,在确保安全的前提下提升核电厂运行的经济性。Fuel rod damage diagnosis(FRDD)in nuclear power plants is a critical issue of concern for nuclear power plant operators and nuclear safety regulators.The application of big data and the nearest neighbor algorithm to fuel rod damage diagnosis has led to the development of a PWR(Pressurized Water Reactor)fuel rod damage diagnosis software.The software has been validated using operational cases of fuel rod damage in nuclear power plants and theoretical examples.The validation results are as follows:①In terms of category analysis of fuel rod breach size,80%of the analysis results are consistent with the theoretical examples;②In terms of the analysis of the damaged fuel rod number,the maximum deviation from the theoretical examples is one rod.The FRDD methodology for PWRs,based on big data and the nearest neighbor algorithm,provides diagnostic results that are closer to the actual damage scenario.This allows for the timely detection of fuel rod damage and changes in the damage state,providing a reliable basis for operational decision-making and radiation protection after fuel rod damage.This approach can enhance the economic efficiency of nuclear power plant operations while ensuring safety.

关 键 词：燃料棒破损 诊断方法 大数据 

分 类 号：TL375.5[核科学技术—核技术及应用]