Browse > Article
http://dx.doi.org/10.1016/j.net.2019.12.025

Application of cost-sensitive LSTM in water level prediction for nuclear reactor pressurizer  

Zhang, Jin (Command and Control Engineering College, Army Engineering University of PLA)
Wang, Xiaolong (College of Nuclear Science and Technology, Naval University of Engineering)
Zhao, Cheng (Command and Control Engineering College, Army Engineering University of PLA)
Bai, Wei (Command and Control Engineering College, Army Engineering University of PLA)
Shen, Jun (Zhenjiang Campus, Army Military Transportation University of PLA)
Li, Yang (Command and Control Engineering College, Army Engineering University of PLA)
Pan, Zhisong (Command and Control Engineering College, Army Engineering University of PLA)
Duan, Yexin (Zhenjiang Campus, Army Military Transportation University of PLA)
Publication Information
Nuclear Engineering and Technology / v.52, no.7, 2020 , pp. 1429-1435 More about this Journal
Abstract
Applying an accurate parametric prediction model to identify abnormal or false pressurizer water levels (PWLs) is critical to the safe operation of marine pressurized water reactors (PWRs). Recently, deep-learning-based models have proved to be a powerful feature extractor to perform high-accuracy prediction. However, the effectiveness of models still suffers from two issues in PWL prediction: the correlations shifting over time between PWL and other feature parameters, and the example imbalance between fluctuation examples (minority) and stable examples (majority). To address these problems, we propose a cost-sensitive mechanism to facilitate the model to learn the feature representation of later examples and fluctuation examples. By weighting the standard mean square error loss with a cost-sensitive factor, we develop a Cost-Sensitive Long Short-Term Memory (CSLSTM) model to predict the PWL of PWRs. The overall performance of the CSLSTM is assessed by a variety of evaluation metrics with the experimental data collected from a marine PWR simulator. The comparisons with the Long Short-Term Memory (LSTM) model and the Support Vector Regression (SVR) model demonstrate the effectiveness of the CSLSTM.
Keywords
LSTM; Parameter prediction; Cost sensitive; Pressurizer; Pressurized water reactor; Time series;
Citations & Related Records
Times Cited By KSCI : 2  (Citation Analysis)
연도 인용수 순위
1 P. Baraldi, G. Gola, E. Zio, D. Roverso, M. Hoffmann, A randomized model ensemble approach for reconstructing signals from faulty sensors, Expert Syst. Appl. 38 (8) (2011) 9211-9224.   DOI
2 P. Baraldi, E. Zio, G. Gola, D. Roverso, M. Hoffmann, Two novel procedures for aggregating randomized model ensemble outcomes for robust signal reconstruction in nuclear power plants monitoring systems, Ann. Nucl. Energy 38 (2-3) (2011) 212-220.   DOI
3 P. Baraldi, A. Cammi, F. Mangili, E. Zio, An ensemble approach to sensor fault detection and signal reconstruction for nuclear system control, Ann. Nucl. Energy 37 (6) (2010) 778-790.   DOI
4 P. Baraldi, A. Cammi, F. Mangili, E. Zio, Local fusion of an ensemble of models for the reconstruction of faulty signals, IEEE Trans. Nucl. Sci. 57 (2) (2010) 793-806.   DOI
5 W. Wu, J. Wang, M. Cheng, Z. Li, Convergence analysis of online gradient method for BP neural networks, Neural Netw. 24 (1) (2011) 91-98.   DOI
6 D. Guo, W. Zhou, H. Li, M. Wang, Hierarchical LSTM for sign language translation, Conf. Artificial Intell. (AAAI) (2018) 6845-6852.
7 S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Comput. 9 (8) (1997) 1735-1780.   DOI
8 Y. Jia, J. Wu, M. Xu, Traffic flow prediction with rainfall impact using a deep learning method, J. Adv. Transp. (2017) (2017) 1-10.
9 H. Shao, H. Jiang, H. Zhang, T. Liang, Electric locomotive bearing fault diagnosis using a novel convolutional deep belief network, IEEE Trans. Ind. Electron. 65 (3) (2018) 2727-2736.   DOI
10 D. Bahdanau, K. Cho, Y. Bengio, Neural machine translation by jointly learning to align and translate, in: Int. Conf. Learn Represent. (ICLR), 2014.
11 J. Yang, J. Kim, An accident diagnosis algorithm using long short-term memory, Nucl. Eng. Technol. 50 (4) (2018) 582-588.   DOI
12 J.H. Ye, J.M. Yi, H.Y. Ji, Research on pressurizer water level control of nuclear reactor based on RBF neural network and PID controller, Int. Conf. Mach. Learn. Cybern. (2010) 1486-1489.
13 M. Marseguerra, M.E. Ricotti, E. Zio, Neural network-based fault detections in a pressurized water reactor pressurizer, Nucl. Sci. Eng. 124 (2) (1996) 339-348.   DOI
14 P.L. Bot, Human reliability data, human error and accident models-illustration through the Three Mile Island accident analysis, Reliab. Eng. Syst. Saf. 83 (2) (2014) 153-167.   DOI
15 X.L. Wang, Q. Cai, Y.Q. Chen, Abnormal parameter reconstruction for marine nuclear power plant based on multi-signal coupling, J. Shanghai Jiaot. Univ. 48 (2) (2014) 210-213.
16 W. Xiao-long, C. Qi, Z. Zi-min, C. Yu-qing, L. Shou-xiang, Prediction for water level of steam generator based on multi-signal fusion by support vector regression, Atomic Energy Sci. Technol. 47 (2013) 1806-1811.
17 Y. Lecun, Y. Bengio, G. Hinton, Deep learning, Nature 521 (7553) (2015) 436-444.   DOI
18 H. Shen, X. Liang, A time series forecasting model based on deep learning integrated algorithm with stacked autoencoders and SVR for FX prediction, Int. Conf. Artif. Neural Netw. (2016) 326-335.
19 X. Jin, C. Xu, J. Feng, Y. Wei, J. Xiong, S. Yan, Deep learning with S-shaped rectified linear activation units, Natl. Int. Conf. Artif. Intell. (AAAI) (2016) 1737-1743.
20 D.P. Kingma, J. Ba, Adam: A Method for Stochastic Optimization, Int. Conf. Learn Represent, ICLR), 2014.
21 J.C. Duchi, E. Hazan, Y. Singer, Adaptive subgradient methods for online learning and stochastic optimization, Conf. Learn. Theory (COLT) (2010) 257-269.
22 G. Hinton, L. Deng, D. Yu, Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups, IEEE Signal Process. Mag. 29 (6) (2012) 82-97.   DOI
23 F. Caliva, et al., A deep learning approach to anomaly detection in nuclear reactors, Int. Joint Conf. Neural Netw. (IJCNN) (2018) 1-8.
24 N.V. Chawla, N. Japkowicz, A. Kolcz, Editorial: special issue on learning from imbalanced data sets, Acm Sigkdd Explor. Newslett. 6 (1) (2014) 1-6.   DOI
25 Z. Wu, Y. Guo, W. Lin, A weighted deep representation learning model for imbalanced fault diagnosis in cyber-physical systems, Sensors 18 (4) (2018) 1096.   DOI
26 W. Xiao-long, C. Qi, C. Yu-qing, Study on pressurizer water level signal reconstruction based on support vector regression, Atomic Energy Sci. Technol. 47 (6) (2013) 1003-1007.   DOI
27 L. Jie, V. Valeria, Z. Enrico, R. Seraoui, A novel dynamic-weighted probabilistic support vector regression-based ensemble for prognostics of time series data, IEEE Trans. Reliab. 64 (4) (2015) 1203-1213.   DOI
28 F.D. Ribeiro, et al., Towards a deep unified framework for nuclear reactor perturbation analysis, Symp. Ser. Comput. Intell. (2018) 120-127.
29 C. Lombardi, A. Mazzola, Prediction of two-phase mixture density using artificial neural networks, Ann. Nucl. Energy 24 (17) (1997) 1373-1387.   DOI
30 S. Zaferanlouei, D. Rostamifard, S. Setayeshi, Prediction of critical heat flux using ANFIS, Ann. Nucl. Energy 37 (6) (2010) 813-821.   DOI
31 Y. Huang, J. Liu, L. Liu, Application of BP neural network in DNBR prediction, Nucl. Tech. 38 (7) (2015) 90-94.
32 A. Shaheryar, X. Yin, H. Hao, et al., A Denoising based Auto-Associative model for robust sensor monitoring in Nuclear Power, Plant. Sci. Tech. Nucl. Inst. (2016) (2016) 1-17.
33 A. Sameer, P. Baraldi, F.D. Maio, E. Zio, A novel fault detection system taking into account uncertainties in the reconstructed signals, Ann. Nucl. Energy 73 (2014) 131-144.   DOI
34 F.D. Maio, P. Baraldi, E. Zio, R. Seraoui, Fault detection in nuclear power plants components by a combination of statistical methods, IEEE Trans. Reliab. 62 (4) (2013) 833-845.   DOI