Nuclear Power AI Applications: Status, Challenges and Opportunities
-
摘要: 近年来,人工智能技术被广泛应用于核电领域,以促进核电厂通过实现自诊断、自寻优、自适应,最终达到提高生产效率、降低运行成本、提高运行安全性的目的。本文介绍了在核电领域经常使用的人工智能技术,总结了其在智慧矿山、智能设计、智能制造和智能运维4个核工业典型应用场景中的研究现状,最后,从数据样本、网络安全、深度学习的解释性3个方面分析了人工智能技术在核电领域应用的挑战和发展趋势。Abstract: In recent years, artificial intelligence (AI) technology has been widely used in the field of nuclear power to promote nuclear power plants to achieve the goal of improving production efficiency, reducing operating costs and improving operating safety through self diagnosis, self optimization and self adaptation. This paper introduces the AI technology often used in the nuclear power field, summarizes its research status in four typical application scenarios of the nuclear industry, namely, intelligent mine, intelligent design, intelligent manufacturing and intelligent operation and maintenance. Finally, it analyzes the challenges and development trends of the application of AI technology in the nuclear power field from three aspects: data samples, network security, and the explanatory nature of deep learning.
-
表 1 现有可信人工智能相关战略
Table 1. Existing Trustworthy AI-related Strategies
国家 相关战略文件 美国 《促进政府使用可信人工智能》行政命令,促进公众接受并信任政府在决策中使用人工智能技术;《迈向识别和管理人工智能偏见的标准》对人工智能偏见问题进行分析,为人工智能风险管理框架提供指导 中国 《国家新一代人工智能标准体系建设指南》给出安全与隐私保护及伦理相关标准;《促进可信人工智能发展倡议》提出践行科技向善、发展以人为本的可信人工智能 欧盟成员国 《人工智能白皮书》提出人工智能“可信生态系统”;《可信人工智能伦理指南草案》提出可信人工智能框架;《可信赖人工智能道德准则》提出实现可信赖人工智能全生命周期的框架;《可信人工智能的政策和投资建议》提出33项具体建议,将可信人工智能变为提高个人和社会福祉的手段 -
[1] BERNARD J A. Applications of artificial intelligence to reactor and plant control[J]. Nuclear Engineering and Design, 1989, 113(2): 219-227. doi: 10.1016/0029-5493(89)90073-3 [2] SUMAN S. Artificial intelligence in nuclear industry: chimera or solution?[J]. Journal of Cleaner Production, 2021, 278: 124022. doi: 10.1016/j.jclepro.2020.124022 [3] LI Y X. Deep reinforcement learning: an overview[DB/OL]. (2018-10-15)[2022-09-08]. https://arxiv.org/abs/1701.07274. [4] 杨思明,单征,丁煜,等. 深度强化学习研究综述[J]. 计算机工程,2021, 47(12): 19-29. doi: 10.19678/j.issn.1000-3428.0061116 [5] 中国电子学会. 中国机器人产业发展报告[EB/OL]. (2022-08-23)[2022-09-08]. https://www.sohu.com/a/580740245_121123919. [6] 张明林,刘洋,吴建勇,等. 中国铀矿地质勘查信息化建设现状及“十四五”发展思路[J]. 世界核地质科学,2021, 38(3): 287-294. doi: 10.3969/j.issn.1672-0636.2021.03.001 [7] 伊广林. 人工智能在地质勘探中的应用[J]. 测井技术,1984(5): 6-13. doi: 10.16489/j.issn.1004-1338.1984.05.002 [8] 陈玉民,李国清,何吉平,等. 山东黄金数字矿山建设实践[J]. 中国矿业,2011, 20(3): 10-14. doi: 10.3969/j.issn.1004-4051.2011.03.003 [9] CONG T L, SU G H, QIU S Z, et al. Applications of ANNs in flow and heat transfer problems in nuclear engineering: a review work[J]. Progress in Nuclear Energy, 2013, 62: 54-71. doi: 10.1016/j.pnucene.2012.09.003 [10] YUAN P, DENG J, PAN L M, et al. Air-water two-phase flow regime and transition criteria in vertical upward narrow rectangular channels[J]. Progress in Nuclear Energy, 2021, 136: 103750. doi: 10.1016/j.pnucene.2021.103750 [11] ZHAO B T, SU Y X. Artificial neural network-based modeling of pressure drop coefficient for cyclone separators[J]. Chemical Engineering Research and Design, 2010, 88(5-6): 606-613. doi: 10.1016/j.cherd.2009.11.010 [12] SERRA P L S, MASOTTI P H F, ROCHA M S, et al. Two-phase flow void fraction estimation based on bubble image segmentation using Randomized Hough Transform with Neural Network (RHTN)[J]. Progress in Nuclear Energy, 2020, 118: 103133. doi: 10.1016/j.pnucene.2019.103133 [13] JIANG B T, LIU Y N. A brief review of computational intelligence techniques for critical heat flux prediction[C]//2018 26th International Conference on Nuclear Engineering. London: American Society of Mechanical Engineers, 2018: V06BT08A050. [14] WEI H M, SU G H, TIAN W X, et al. Study on the onset of nucleate boiling in narrow annular channel by genetic neural network[J]. International Communications in Heat and Mass Transfer, 2010, 37(6): 596-599. doi: 10.1016/j.icheatmasstransfer.2009.11.017 [15] ZHOU L W, GARG D, QIU Y, et al. Machine learning algorithms to predict flow condensation heat transfer coefficient in mini/micro-channel utilizing universal data[J]. International Journal of Heat and Mass Transfer, 2020, 162: 120351. doi: 10.1016/j.ijheatmasstransfer.2020.120351 [16] CECILIA MARTÍN-DEL-CAMPO A,MIGUEL-NGEL PALOMERA-PÉREZ B,JUAN-LUIS FRANOIS A. Advanced and flexible genetic algorithms for BWR fuel loading pattern optimization[J]. Annals of Nuclear Energy, 2009, 36(10): 1553-1559. doi: 10.1016/j.anucene.2009.07.013 [17] SAFARZADEH O, ZOLFAGHARI A, ZANGIAN M, et al. Pattern optimization of PWR reactor using hybrid parallel Artificial Bee Colony[J]. Annals of Nuclear Energy, 2014, 63: 295-301. doi: 10.1016/j.anucene.2013.08.011 [18] LIN C, HUNG S C. Automatic multi-cycle reload design of pressurized water reactor using particle swarm optimization algorithm and local search[J]. Annals of Nuclear Energy, 2013, 59: 255-260. doi: 10.1016/j.anucene.2013.04.013 [19] DE LIMA A M M, SCHIRRU R, DA SILVA F C, et al. A nuclear reactor core fuel reload optimization using artificial ant colony connective networks[J]. Annals of Nuclear Energy, 2008, 35(9): 1606-1612. doi: 10.1016/j.anucene.2008.03.002 [20] AKBARI R, ABBASI M, FAGHIHI F, et al. A novel multi-objective optimization method, imperialist competitive algorithm, for fuel loading pattern of nuclear reactors[J]. Progress in Nuclear Energy, 2018, 108: 391-397. doi: 10.1016/j.pnucene.2018.06.016 [21] PAZIRANDEH A, TAYEFI S. Optimizing the fuel management in a VVER-1000 reactor using an artificial neural network[J]. Annals of Nuclear Energy, 2012, 42: 112-118. doi: 10.1016/j.anucene.2011.12.010 [22] 雷铠灰,曹良志,万承辉,等. 基于深度卷积神经网络的堆芯换料方案性能评价研究[J]. 原子能科学技术,2021, 55(2): 279-285. doi: 10.7538/yzk.2020.youxian.0111 [23] KUMAR A, TSVETKOV P V. A new approach to nuclear reactor design optimization using genetic algorithms and regression analysis[J]. Annals of Nuclear Energy, 2015, 85: 27-35. doi: 10.1016/j.anucene.2015.04.028 [24] 韦子豪,王端,王东东,等. 神经网络-遗传复合算法在压水堆堆芯换料设计中的应用[J]. 原子能科学技术,2020, 54(5): 825-834. doi: 10.7538/yzk.2019.youxian.0788 [25] 刘东,罗琦,唐雷,等. 基于PINN深度机器学习技术求解多维中子学扩散方程[J]. 核动力工程,2022, 43(2): 1-8. doi: 10.13832/j.jnpe.2022.02.0001 [26] 大众日报. 核电装备智能制造又添利器[EB/OL]. (2020-12-17)[2022-09-08]. http://paper.dzwww.com/dzrb/content/20201217/Articel17003MT.htm. [27] 中国原子能科学研究院. 原子能院智能装备和机器人研发团队成功中标中科院强辐射巡测机器人采购项目[EB/OL]. (2019-12-03)[2022-09-08]. http://www.ciae.ac.cn/newsContent.jsp?RID=2998. [28] FRENCH R, MARIN-REYES H, KAPELLMANN-ZAFRA G, et al. Development of an intelligent robotic additive manufacturing cell for the nuclear industry[C]//International Conference on Applied Human Factors and Ergonomics. Washington: Springer, 2019: 3-13. [29] CHEN J, PATTON R J. Robust model-based fault diagnosis for dynamic systems[M]. Germany: Springer Science & Business Media, 2012. [30] NABESHIMA K, SUZUDO T, SUZUKI K, et al. Real-time nuclear power plant monitoring with neural network[J]. Journal of Nuclear Science and Technology, 1998, 35(2): 93-100. doi: 10.1080/18811248.1998.9733829 [31] LIU Y K, XIE C L, PENG M J, et al. Improvement of fault diagnosis efficiency in nuclear power plants using hybrid intelligence approach[J]. Progress in Nuclear Energy, 2014, 76: 122-136. doi: 10.1016/j.pnucene.2014.05.001 [32] AGARWAL V, ALAMANIOTIS M, TSOUKALAS L H. Predictive based monitoring of nuclear plant component degradation using support vector regression: NL/CON-14-32980[R]. Idaho Falls, United States: Idaho National Lab. , 2015. [33] LIU J, SERAOUI R, VITELLI V, et al. Nuclear power plant components condition monitoring by probabilistic support vector machine[J]. Annals of Nuclear Energy, 2013, 56: 23-33. doi: 10.1016/j.anucene.2013.01.005 [34] 陈涵瀛. 核电站热工水力系统工况预测与诊断方法研究[D]. 哈尔滨: 哈尔滨工程大学, 2018. [35] MO K, LEE S J, SEONG P H. A dynamic neural network aggregation model for transient diagnosis in nuclear power plants[J]. Progress in Nuclear Energy, 2007, 49(3): 262-272. doi: 10.1016/j.pnucene.2007.01.002 [36] BARTAL Y, LIN J, UHRIG R E. Nuclear power plant transient diagnostics using artificial neural networks that allow "don't-know" classifications[J]. Nuclear Technology, 1995, 110(3): 436-449. doi: 10.13182/NT95-A35112 [37] LI X, FU X M, XIONG F R, et al. Deep learning-based unsupervised representation clustering methodology for automatic nuclear reactor operating transient identification[J]. Knowledge-Based Systems, 2020, 204: 106178. doi: 10.1016/j.knosys.2020.106178 [38] 赵琛,沈杰,李思颖. 水下核电机器人应用现状与技术发展分析[J]. 自动化技术与应用,2019, 38(11): 94-98. doi: 10.3969/j.issn.1003-7241.2019.11.022 [39] 田冰. 喷涂机器人在核电维修领域的应用分析[J]. 现代工业经济和信息化,2021, 11(10): 142-143,148. doi: 10.16525/j.cnki.14-1362/n.2021.10.054 [40] 国家原子能机构. 中广核成功研制核电站蒸汽发生器爬墙机器人[EB/OL]. (2017-05-24)[2022-09-08]. http://www.caea.gov.cn/n6760338/n6760342/c6829999/content.html. [41] 许勇,蔡云泽,宋林. 基于数据驱动的核电设备状态评估研究综述[J]. 上海交通大学学报,2022, 56(3): 267-278. [42] 刘桐,顾小清. 走向可解释性:打开教育中人工智能的“黑盒”[J]. 中国电化教育,2022(5): 82-90. doi: 10.3969/j.issn.1006-9860.2022.05.012 [43] SZEGEDY C, ZAREMBA W, SUTSKEVER I, et al. Intriguing properties of neural networks[DB/OL]. (2013-12-21)[2022-09-08]. https://arxiv.org/abs/1312.6199v4. [44] ALCORN M A, LI Q, GONG Z T, et al. Strike (with) a pose: neural networks are easily fooled by strange poses of familiar objects[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE, 2019: 4845-4854. [45] 李明慧,江沛佩,王骞,等. 针对深度学习模型的对抗性攻击与防御[J]. 计算机研究与发展,2021, 58(5): 909-926. doi: 10.7544/issn1000-1239.2021.20200920 [46] 陈磊,李雅静. 人工智能系统安全综述[J]. 信息通信技术与政策,2021, 47(8): 56-63. doi: 10.12267/j.issn.2096-5931.2021.08.009 [47] EYKHOLT K, EVTIMOV I, FERNANDES E, et al. Robust physical-world attacks on deep learning visual classification[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 1625-1634. [48] 张林超,张欣海. 可信任的人工智能系统及安全框架浅析[J]. 中国电子科学研究院学报,2019, 14(12): 1253-1258. doi: 10.3969/j.issn.1673-5692.2019.12.006 [49] 王国胤,傅顺,杨洁,等. 基于多粒度认知的智能计算研究[J]. 计算机学报,2022, 45(6): 1161-1175. doi: 10.11897/SP.J.1016.2022.01161 [50] BAO H N, WANG G Y, LI S, et al. Multi-granularity visual explanations for CNN[J]. Knowledge-Based Systems, 2022, 253: 109474. doi: 10.1016/j.knosys.2022.109474 [51] 王国胤,代劲,李昊. 基于多粒度认知计算的生产安全管理与决策[J]. 中国科学基金,2021, 35(5): 752-758. doi: 10.16262/j.cnki.1000-8217.2021.05.012 [52] 于洪,何德牛,王国胤,等. 大数据智能决策[J]. 自动化学报,2020, 46(5): 878-896. doi: 10.16383/j.aas.c180861 [53] 孔祥维,唐鑫泽,王子明. 人工智能决策可解释性的研究综述[J]. 系统工程理论与实践,2021, 41(2): 524-536. doi: 10.12011/SETP2020-1536 [54] MACHLEV R, HEISTRENE L, PERL M, et al. Explainable artificial intelligence (XAI) techniques for energy and power systems: review, challenges and opportunities[J]. Energy and AI, 2022, 9: 100169. doi: 10.1016/j.egyai.2022.100169 [55] 中国信息通信研究院, 京东探索研究院. 可信人工智能白皮书[Z]. 上海: 2021世界人工智能大会-可信AI论坛, 2021-07-09. [56] 何积丰. 安全可信人工智能[J]. 信息安全与通信保密,2019(10): 5-8. [57] 何积丰. 智能制造与安全可信人工智能[J]. 信息安全与通信保密,2020(12): 2-6. doi: 10.3969/j.issn.1009-8054.2020.12.001 -
表(1)
计量
- 文章访问数: 9062
- HTML全文浏览量: 583
- PDF下载量: 1052
- 被引次数: 0