To address problems of high cost, complicated process and low accuracy of oxygen content measurement in flue gas of coal-fired power plant, a method based on long short-term memory (LSTM) network is proposed in this paper to replace oxygen sensor to estimate oxygen content in flue gas of boilers. Specifically, first, the LSTM model was built with the Keras deep learning framework, and the accuracy of the model was further improved by selecting appropriate super-parameters through experiments. Secondly, the flue gas oxygen content, as the leading variable, was combined with the mechanism and boiler process primary auxiliary variables. Based on the actual production data collected from a coal-fired power plant in Yulin, China, the data sets were preprocessed. Moreover, a selection model of auxiliary variables based on grey relational analysis is proposed to construct a new data set and divide the training set and testing set. Finally, this model is compared with the traditional soft-sensing modelling methods (i.e. the methods based on support vector machine and BP neural network). The RMSE of LSTM model is 4.51% lower than that of GA-SVM model and 3.55% lower than that of PSO-BP model. The conclusion shows that the oxygen content model based on LSTM has better generalization and has certain industrial value.
CorniaMBaraldiLSerraGCucchiaraR (2018) Predicting human eye fixations via an LSTM-based saliency attentive model. IEEE Transactions on Image Processing27(10): 5142–5154.
2.
FortunaLGrazianiSRizzoAXibiliaMG (2007) Soft sensors for monitoring and control of industrial processes. Springer Science & Business Media.
3.
FujiwaraKKanoMHasebeSTakinamiA (2009) Soft-sensor development using correlation-based just-in-time modeling. AIChE Journal55(7): 1754–1765.
4.
García-PedrajasNHervás-MartínezCOrtiz-BoyerD (2005) Cooperative coevolution of artificial neural network ensembles for pattern classification. IEEE Transactions on Evolutionary Computation9(3): 271–302.
5.
GengZXuMZhuQ (2019) Research and application of soft measurement model for complex chemical processes based on deep learning. Journal of Chemical Industry70(02): 144–151.
6.
GersFASchmidhuberJCumminsF (1999) Learning to forget: Continual prediction with lstm. Neural Computation12(10): 2451–2471.
7.
GonzagaJMeleiroLACKiangCMaciel FilhoR (2009) Ann-based soft-sensor for real-time process monitoring and control of an industrial polymerization process. Computers & Chemical Engineering33(1): 43–49.
8.
GreffKSrivastavaRKKoutníkJSteunebrinkBRSchmidhuberJ (2016) LSTM: A search space odyssey. IEEE Transactions on Neural Networks and Learning Systems28(10): 2222–2232.
9.
HeYWangYYangQLinW (2017) Industry development and frontier technology roadmap of thermal power generation. Proceedings of the CSEE37(13): 3787–3794.
10.
HuJDingB (2019a) Dynamic output feedback predictive control with one free control move for the Takagi-Sugeno model with bounded disturbance. IEEE Transactions on Fuzzy Systems27(3): 462–473.
11.
HuJDingB (2019b) Output feedback robust MPC for linear systems with norm-bounded model uncertainty and disturbance. Automatica. Https://doi.org/10.1016/j.automatica.2019.07.002.
12.
JiangZJiaQGuanX (2019) A review of multi-temporal-and-spatial-scale wind power forecasting method. Acta Automatica Sinica45(1): 51–71.
13.
KadlecPGabrysBStrandtS (2009) Data-driven soft sensors in the process industry. Computers & Chemical Engineering33(4): 795–814.
14.
LeCunYBengioYHintonG (2015) Deep learning. Nature521(7553): 436–444.
15.
LiJZhaoKChenXZhangY (2017a) Combined optimization of optimal oxygen content and second air distribution in 600 mw coal-fired unit. Proceedings of the CSEE37: 4422–4429.
16.
LiJZhaoKNiuC (2017b) GA-SVM-based soft-sensor model for oxygen content in flue gas of utility boilers. Thermal Power Generation46(4): 63–69.
17.
LiuSCaiHCaoYYangY (2011) Advance in grey incidence analysis modelling. In: 2011 IEEE International Conference on Systems, Man, and Cybernetics. IEEE, 1886–1890.
18.
LvYDuanYKangWLiZWangFY (2014) Traffic flow prediction with big data: A deep learning approach. IEEE Transactions on Intelligent Transportation Systems16(2): 865–873.
19.
MonnerDReggiaJA (2012) A generalized LSTM-like training algorithm for second-order recurrent neural networks. Neural Networks25: 70–83.
20.
MopuriKRGargUBabuRV (2018) CNN fixations: An unraveling approach to visualize the discriminative image regions. IEEE Transactions on Image Processing28(5): 2116–2125.
21.
PanHMiW (2020) An adaptive decoder design based on the receding horizon optimization in BMI system. Cognitive Neurodynamics. Https://doi.org/10.1007/s11571-019-09567-4.
22.
PanHMiWLeiXDengJ (2020a) A closed-loop brain-machine interface framework design for motor rehabilitation. Biomedical Signal Processing and Control. Https://doi.org/10.1016/j.bspc.2020.101877.
23.
PanHMiWLeiXZhongW (2020b) A closed-loop bmi system design based on the improved SJIT model and the network of Izhikevich neurons. Neurocomputing. Https://doi.org/10.1016/j.neucom.2020.03.047.
24.
RenCAnNWangJLiLHuBShangD (2014) Optimal parameters selection for BP neural network based on particle swarm optimization: A case study of wind speed forecasting. Knowledge-based Systems56: 226–239.
25.
ShinHCRothHRGaoMLuLXuZNoguesIYaoJMolluraDSummersRM (2016) Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE Transactions on Medical Imaging35(5): 1285–1298.
26.
SongLLiJunLManZ (2012) Prediction for short-term traffic flow based on modified PSO optimized BP neural network. Systems Engineering – Theory & Practice32(9): 2045–2049.
27.
SunYWangYLiuXYangCZhangZGuiWChenXZhuB (2017) A novel Bayesian inference soft sensor for real-time statistic learning modeling for industrial polypropylene melt index prediction. Journal of Applied Polymer Science134(40): 45384.
28.
WangPFengXLiWPingXYuW (2017) Robust RHC for wheeled vehicles with bounded disturbances. International Journal of Robust and Nonlinear Control29(7): 2063–2081. Https://doi.org/10.1002/rnc.4478.
29.
YangZGuoXChenZHuangYZhangY (2018) RNN-stega: Linguistic steganography based on recurrent neural networks. IEEE Transactions on Information Forensics and Security14(5): 1280–1295.
30.
YaoLGeZ (2017) Deep learning of semisupervised process data with hierarchical extreme learning machine and soft sensor application. IEEE Transactions on Industrial Electronics65(2): 1490–1498.
31.
ZhangDLindholmGRatnaweeraH (2018a) Use long short-term memory to enhance internet of things for combined sewer overflow monitoring. Journal of Hydrology556: 409–418.
32.
ZhangYXiongRHeHPechtMG (2018b) Long short-term memory recurrent neural network for remaining useful life prediction of lithium-ion batteries. IEEE Transactions on Vehicular Technology67(7): 5695–5705.
