TY - JOUR
T1 - A New CO/CO2 Prediction Model Based on Labeled and Unlabeled Process Data for Sintering Process
AU - Zhou, Kailong
AU - Chen, Xin
AU - Wu, Min
AU - Du, Sheng
AU - Hu, Jie
AU - Nakanishi, Yosuke
N1 - Funding Information:
Manuscript received November 18, 2019; revised February 16, 2020; accepted March 26, 2020. Date of publication April 8, 2020; date of current version October 23, 2020. This work was supported in part by the National Natural Science Foundation of China under Grant 61210011, in part by the Hubei Provincial Natural Science Foundation of China under Grant 2015CFA010, in part by the 111 project under Grant B17040, and in part by the Program of China Scholarship Council under Grant 201906410061. Paper no. TII-19-5021. (Corresponding author: Xin Chen.) Kailong Zhou is with the School of Automation, China University of Geosciences, Wuhan 430074, China, with the Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems, Wuhan 430074, China, and also with the Graduate School of Envi-
Publisher Copyright:
© 2005-2012 IEEE.
PY - 2021/1
Y1 - 2021/1
N2 - To reduce energy consumption and harmful emission, it is of great significance to improve carbon efficiency in sintering process, which is able to be achieved if the carbon efficiency can be accurately predicted. In this article, the ratio of CO and CO$_2$ (CO/CO$_2$) is taken as a measurement of the carbon efficiency. As CO/CO$_2$ is hard to measure, and there exist multiple working conditions, multiple variables, and nonlinearity, a hybrid CO/CO$_2$ prediction model is devised based on the aforementioned characteristics. First, the sintering process is analyzed, and the key characteristics to predict the CO/CO$_2$ are extracted. Next, the configuration of the prediction model is given based on the analysis. The model consists by two submodels, one is to predict the state variables by an improved just-in-time learning model, combining three neural network (NN) models. The other is to predict CO/CO$_2$ with semisupervised algorithm, based on deep belief network with a combination of the three NN regression methods. Then, the configurations of the two submodels are introduced in detail. The test results based on actual running data exhibit the good performance of the model.
AB - To reduce energy consumption and harmful emission, it is of great significance to improve carbon efficiency in sintering process, which is able to be achieved if the carbon efficiency can be accurately predicted. In this article, the ratio of CO and CO$_2$ (CO/CO$_2$) is taken as a measurement of the carbon efficiency. As CO/CO$_2$ is hard to measure, and there exist multiple working conditions, multiple variables, and nonlinearity, a hybrid CO/CO$_2$ prediction model is devised based on the aforementioned characteristics. First, the sintering process is analyzed, and the key characteristics to predict the CO/CO$_2$ are extracted. Next, the configuration of the prediction model is given based on the analysis. The model consists by two submodels, one is to predict the state variables by an improved just-in-time learning model, combining three neural network (NN) models. The other is to predict CO/CO$_2$ with semisupervised algorithm, based on deep belief network with a combination of the three NN regression methods. Then, the configurations of the two submodels are introduced in detail. The test results based on actual running data exhibit the good performance of the model.
KW - CO/CO2
KW - deep belief network (DBN)
KW - improved just-in-time learning (JITL)
KW - neural network (NN)
KW - sintering process
UR - http://www.scopus.com/inward/record.url?scp=85096035372&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85096035372&partnerID=8YFLogxK
U2 - 10.1109/TII.2020.2985663
DO - 10.1109/TII.2020.2985663
M3 - Article
AN - SCOPUS:85096035372
SN - 1551-3203
VL - 17
SP - 333
EP - 345
JO - IEEE Transactions on Industrial Informatics
JF - IEEE Transactions on Industrial Informatics
IS - 1
M1 - 9061027
ER -
