摘要
[目的/意义]在核电站等涉辐射环境中,辐射工作许可证(RWP)作为确保作业安全的关键环节,面临着审批数量庞大、人工审核耗时等问题.旨在通过构建基于RoBERTa的数据驱动决策系统,利用先进的自然语言处理技术和数据治理策略,提高RWP审批的效率与精准度,进而加强辐射安全管理,保障作业人员安全.[方法/过程]提出了一种融合深度学习与数据治理的方法,借助RoBERTa模型的强大预训练能力,结合核电场景下的历史数据,开发了一套智能化审批系统,能够自动完成RWP的风险评估与审批流程.[结果/结论]系统能显著提升审批效率,预计可辅助处理90%以上RWP审批流程,为核电站的数字化转型提供了有力的技术支持,同时也强化了网络空间的安全治理.
[Purpose/Significance]In radiation-intensive environments such as nuclear power plants,Radiation Work Permits(RWPs)play a crucial role in ensuring the safety of operations.However,the sheer volume of RWPs and the time-consuming nature of manual reviews present significant challenges.This work aims to develop an intelligent decision-making system based on RoBERTa and data governance strategies to enhance the efficiency and accuracy of RWP approvals,thereby strengthening radiation safety management and personnel safety.[Method/Process]We propose a method combining deep learning with data governance,leveraging the robust pre-trained capabilities of the RoBERTa model and historical data specific to nuclear scenarios to develop an automated approval system for RWPs.[Results/Conclusion]Experimental results demonstrate that this system significantly improves the efficiency of approvals,and is expected to support over 90%of RWP(Radiation Work Permit)approval procedures,providing strong technical support for the digital transformation of nuclear power plants and enhancing cybersecurity governance.
作者
牛朝辉
陈晓月
Niu Chaohui;Chen Xiaoyue(Sanmen Nuclear Power Co.,Ltd.,Zhejiang Sanmen 317112;The Third Research Institute of Ministry of Public Security,Shanghai 200120)
出处
《网络空间安全》
2024年第4期134-140,共7页
Cyberspace Security
作者简介
牛朝辉(1986-),男,汉族,浙江台州人,兰州大学,学士,三门核电有限公司,高级工程师,主要研究方向和关注领域:辐射防护管理和技术;陈晓月(1997-),女,汉族,江苏盐城人,西安电子科技大学,硕士,公安部第三研究所,研究实习员,主要研究方向和关注领域:图像识别与目标检测、智慧安管。
