1. 上海交通大学国家电投智慧能源创新学院
2. 上海交通大学核动力系统与装备教育部重点实验室
3. 上海交通大学四川研究院
4. 上海交通大学核电安全与装备全国重点实验室
5. 中广核工程有限公司
纸质出版:2025
移动端阅览
陈逸龙, 林萌, 周士祺, 等. 基于对比学习的核电厂故障诊断方法[J]. 核科学与工程, 2025,45(6):1241-1249.
陈逸龙, 林萌, 周士祺, et al. 基于对比学习的核电厂故障诊断方法[J]. 2025, 45(6): 1241-1249.
陈逸龙, 林萌, 周士祺, 等. 基于对比学习的核电厂故障诊断方法[J]. 核科学与工程, 2025,45(6):1241-1249. DOI: 10.20190/j.cnki.02580918.202506023.
陈逸龙, 林萌, 周士祺, et al. 基于对比学习的核电厂故障诊断方法[J]. 2025, 45(6): 1241-1249. DOI: 10.20190/j.cnki.02580918.202506023.
在核电厂实际运行过程中,具有高可靠性的核电厂故障数据稀少,同时故障数据需要经验丰富的专家进行标注,带标签故障数据获取成本较高,而基于有监督深度学习的故障诊断方法难以在标签稀缺情况下进行正确诊断。为减轻核电厂故障诊断方法对有标签样本的依赖,本文提出一种基于对比学习的核电厂故障诊断方法。首先从原始数据集中采集多个时间步重构成为带时序信息的故障样本。利用不同工况、不同故障的监测数据存在的天然分布差异构建正负样本对,而无需像一般的对比学习范式那样进行手动数据增强。然后输入残差卷积网络和门控循环单元构建混合神经网络从不同角度提取故障的高维特征,接着用提取的特征通过对比学习方法实现特征提取器的无监督训练,最后使用少量标签数据训练多标签分类器。相较于有监督学习的方法,在有标签数据稀少的情况下,该方法能够充分利用无真实标签的数据提升网络的特征表达能力,相比其他方法拥有更好的故障诊断性能。
邓强,王航,彭敏俊,王晓昆,徐仁义.基于时间卷积胶囊网络的核动力装置事故诊断技术研究[J].原子能科学技术,2023(02).
孙原理,宋志浩.基于卷积长短期记忆网络和人工鲸鱼算法的核反应堆运行事件诊断方法研究[J].核动力工程,2022(04).
Liu Maolong,Wei Yiwei,Wang Lang,Xiong Zhenqin,Gu Hanyang.An accident diagnosis method of pressurized water reactor based on BI-LSTM neural network[J].Progress in Nuclear Energy,2023.
Lee Gyumin,Lee Seung Jun,Lee Changyong.A convolutional neural network model for abnormality diagnosis in a nuclear power plant[J].Applied Soft Computing,2020.
Jia Liu,Gao Qin,Liu Zhaopeng,Tan Haibo,Zhou Liwei.Multidisciplinary fault diagnosis of complex engineering systems: A case study of nuclear power plants[J].International Journal of Industrial Ergonomics,2020.
Hanan A. Saeed,Hang Wang,Minjun Peng,Anwar Hussain,Amjad Nawaz.Online fault monitoring based on deep neural network & sliding window technique[J].Progress in Nuclear Energy,2020.
Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin, 150001, China,Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin, 150001, China,Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin, 150001, China,Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin, 150001, China,Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin, 150001, China,Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin, 150001, China.Research on intelligent fault diagnosis method for nuclear power plant based on correlation analysis and deep belief network[J].Progress in Nuclear Energy,2018.
Mandal Shyamapada,Santhi B.,Sridhar S.,Vinolia K.,Swaminathan P..Nuclear Power Plant Thermocouple Sensor Fault Detection and Classification using Deep Learning and Generalized Likelihood Ratio Test[J].IEEE Transactions on Nuclear Science,2017.
Kyunghyun Cho,Bart van Merrienboer,Çaglar Gülçehre,Fethi Bougares,Holger Schwenk,Yoshua Bengio.Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation.[J].CoRR,2014.
Dong Hou,Meng Lin,Zhihong Xu,Yanhua Yang.Development and application of an extensible engineering simulator for NPP DCS closed-loop test[J].Annals of nuclear energy,2011.
Jianping Ma,Jin Jiang.Applications of fault detection and diagnosis methods in nuclear power plants: A review[J].Progress in nuclear engergy,2011.
Meng Lin,Dong Hou,Pengfei Liu,Zongwei Yang,Yanhua Yang.Main control system verification and validation of NPP digital I&C system based on engineering simulator[J].Nuclear Engineering and Design,2010.
Meng Lin,Yun Su,Rui Hu,Ronghua Zhang,Yanhua Yang.Development of a thermal–hydraulic system code for simulators based on RELAP5 code[J].Nuclear Engineering and Design,2005.
Hochreiter S,Schmidhuber J.Long short-term memory.[J].Neural computation,1997(8).
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621