Various data driven soft sensor models have been established for online prediction of the silicon content in blast furnace ironmaking processes. However, two main disadvantages still remain in these empirical models. First, most of traditional outlier detection methods for preprocessing the data samples assume that they (approximately) follow a Gaussian distribution and thus may be invalid for some situations. To address this problem, a support vector clustering (SVC) based efficient outlier detection method is proposed whereby the process nonlinearity and non-Gaussianity can be better handled. Second, only using a single global model is insufficient to capture all the process characteristics, especially for those complicated regions. In this paper, a reliable just-in-time modelling method is proposed. The SVC outlier detection is integrated into the just-in-time-based local modelling method to enhance the reliability of quality prediction. A healthier relevant data set is constructed to build a more reliable local prediction model. Moreover, the historical data set is updated repetitively in a reasonable way. The superiority of the proposed method is demonstrated and compared with other soft sensors in terms of online prediction of the silicon content in an industrial blast furnace in China.
Get full access to this article
View all access options for this article.
References
1.
SugawaraKMorimotoKSugawaraT and DranoffJS: ‘Dynamic behavior of iron forms in rapid reduction of carbon-coated iron ore’, AIChE J., 1999, 45, 574–580.
2.
BiXGTorrsselK and WijkO: ‘Prediction of the blast furnace process by a mathematical model’, ISIJ Int., 1992, 32, 481–488.
3.
NogamiHChuMS and YajiJ: ‘Multi-dimensional transient mathematical simulator of blast furnace process based on multi-fluid and kinetic theories’, Comput. Chem. Eng., 2005, 29, 2438–2448.
4.
NishiokaKMaedaT and ShimizuM: ‘A three-dimensional mathematical modelling of drainage behavior in blast furnace hearth’, ISIJ Int., 2005, 45, 669–676.
5.
RadhakrishnanVR and RamKM: ‘Mathematical model for predictive control of the bell-less top charging system of a blast furnace’, J. Process Contr., 2001, 11, 565–586.
6.
UedaSNatsuiSNogamiHYagiJI and TatsuroA: ‘Recent progress and future perspective on mathematical modelling of blast furnace’, ISIJ Int., 2010, 50, 914–923.
7.
SinghHSridharNV and DeoB: ‘Artificial neural nets for prediction of silicon content of blast furnace hot metal’, Steel Res. Int., 1996, 67, 521–527.
8.
RadhakrishnanVR and MohamedAR: ‘Neural networks for the identification and control of blast furnace hot metal quality’, J. Process Contr., 2000, 10, 509–524.
9.
ChenJ: ‘A predictive system for blast furnaces by integrating a neural network with qualitative analysis’, Eng. Appl. Artif. Intell., 2001, 14, 77–85.
10.
JimenezJMochonJde AyalaiJS and ObesoF: ‘Blast furnace hot metal temperature prediction through neural networks-based models’, ISIJ Int., 2004, 44, 573–580.
11.
SaxenH and PetterssonF: ‘Nonlinear prediction of the hot metal silicon content in the blast furnace’, ISIJ Int., 2007, 47, 1732–1737.
12.
PetterssonFChakrabortiN and SaxénH: ‘A genetic algorithm based multi objective neural net applied to noisy blast furnace data’, Appl. Soft Comput., 2007, 7, 387–397.
13.
NurkkalaAPetterssonF and SaxenH: ‘Nonlinear modelling method applied to prediction of hot metal silicon in the ironmaking blast furnace’, Industr. Eng. Chem. Res., 2011, 50, 9236–9248.
14.
HaoXJShenFMDuGShenYS and XieZ: ‘A blast furnace prediction model combining neural network with partial least square regression’, Steel Res. Int., 2005, 76, 694–699.
15.
BhattacharyaT: ‘Prediction of silicon content in blast furnace hot metal using partial least squares (PLS)’, ISIJ Int., 2005, 45, 1943–1945.
16.
MartinRDObesoFMochonJBareaR and JimenezJ: ‘Hot metal temperature prediction in blast furnace using advanced model based on fuzzy logic tools’, Ironmak. Steelmak., 2007, 34, 241–247.
17.
WallerM and SaxenH: ‘On the development of predictive models with applications to a metallurgical process’, Industr. Eng. Chem. Res., 2000, 39, 982–988.
18.
GaoCHZhouZM and ChenJM: ‘Assessing the predictability for blast furnace system through nonlinear time series analysis’, Industr. Eng. Chem. Res., 2008, 47, 3037–3045.
19.
WallerM and SaxenH: ‘Application of nonlinear time series analysis to the prediction of silicon content of pig iron’, ISIJ Int., 2002, 42, 316–318.
20.
MiyanoTKimotoSShibutaHNakashimaKIkenagaY and AiharaK: ‘Time series analysis and prediction on complex dynamical behavior observed in a blast furnace’, Phys. D, 2000, 135, 305–330.
21.
ZengJS and GaoCH: ‘Improvement of identification of blast furnace ironmaking process by outlier detection and missing value imputation’, J. Process Contr., 2009, 19, 1519–1528.
22.
JianLGaoCH and XiaZQ: ‘A sliding-window smooth support vector regression model for nonlinear blast furnace system’, Steel Res. Int., 2011, 82, 169–179.
23.
GaoCHZengJS and LuoSH: ‘Modelling of the thermal state change of blast furnace hearth with support vector machines’, IEEE Trans. Industr. Electron., 2012, 59, 1134–1145.
24.
JianLShenSQ and SongYQ: ‘Improving the solution of least squares support vector machines with application to a blast furnace system’, J. Appl. Math., 2012, Article ID 949654.
25.
GaoCHChenJMZengJSLiuXY and SunYX: ‘A chaos-based iterated multistep predictor for blast furnace ironmaking process’, AIChE J., 2009, 55, 947–962.
26.
GaoCHZengJS and ZhouZM: ‘Identification of multiscale nature and multiple dynamics of the blast furnace system from operating data’, AIChE J., 2011, 57, 3448–3458.
27.
ChuYX and GaoCH: ‘Data-based multiscale modelling for blast furnace system’, AIChE J., 2014, 60, 2197–2210.
28.
LuoSHGaoCHZengJS and HuangJ: ‘Blast furnace system modelling by multivariate phase space reconstruction and neural networks’, Asian J. Contr., 2013, 15, 553–561.
29.
NurkkalaAPetterssonF and SaxenH: ‘A study of blast furnace dynamics using multiple autoregressive vector models’, ISIJ Int., 2012, 52, 1763–1770.
30.
SaxenHGaoCH and GaoZW: Data-Driven time discrete models for dynamic prediction of the hot metal silicon content in the blast furnace—a review,IEEE Trans. Industr. Inform., 2013, 9, 2213–2225.
31.
KanoM and NakagawaY: ‘Data-based process monitoring, process control, and quality improvement: Recent developments and applications in steel industry’, Comput. Chem. Eng., 2008, 32, 12–24.
32.
KadlecPGabrysB and StrandtS: ‘Data-driven soft sensors in the process industry’, Comput. Chem. Eng., 2009, 33, 795.
33.
VapnikVN: ‘The nature of statistical learning theory’; 1995, New York, Springer-Verlag.
34.
SuykensJAKVan GestelTde BrabanterJde MoorB and VandewalleJ: ‘Least squares support vector machines’; 2002, World Singapore, Scientific.
35.
ScholkopfB and SmolaAJ: ‘Learning with kernels, support vector machines’; 2002, Cambridge, MA.
36.
CzekajTWuW and WalczakB: ‘About kernel latent variable approaches and SVM’, J. Chemom., 2005, 19, 341–354.
37.
LiuYHuNPWangHQ and LiP: ‘Soft chemical analyzer development using adaptive least-squares support vector regression with selective pruning and variable moving window size’, Industr. Eng. Chem. Res., 2009, 48, 5731–5741.
38.
ChiangLHPellRJ and SeasholtzMB: ‘Exploring process data with the use of robust outlier detection algorithms’, J. Process Contr., 2003, 13, 437–449.
39.
LiuHShahSL and JiangW: ‘Online outlier detection and data cleaning’, Comput. Chem. Eng., 2004, 28, 1635–1647.
40.
Ben-HurAHornDSiegelmannH and VapnikV: ‘Support vector clustering’, J. Mach. Learn. Res., 2002, 2, 125–137.
41.
ChiangJH and HaoPY: ‘A new kernel-based fuzzy clustering approach: support vector clustering with cell growing’, IEEE Trans. Fuzzy Syst., 2003, 11, 518–527.
42.
WangJS and ChiangJC: ‘A cluster validity measure with outlier detection for support vector clustering’, IEEE Trans. Systems Man Cybern B: Cybern., 2008, 38, 78–89.
BontempiGBirattariM and BersiniH: ‘Lazy learning for local modelling and control design’, Int. J. Contr., 1999, 72, 643–658.
45.
ChengC and ChiuMS: ‘A new data-based methodology for nonlinear process modelling’, Chem. Eng. Sci., 2004, 59, 2801–2810.
46.
FujiwaraKKanoMHasebeS and TakinamiA: ‘Soft-sensor development using correlation-based just-in-time modelling’, AIChE J., 2009, 55, 1754–1765.
47.
GeZQ and SongZH: ‘A comparative study of just-in-time-learning based methods for online soft sensor modelling’, Chemom. Intell. Labor. Syst., 2010, 104, 306–317.
48.
LiuYGaoZLLiP and WangHQ: ‘Just-in-time kernel learning with adaptive parameter selection for soft sensor modelling of batch processes’, Industr. Eng. Chem. Res., 2012, 51, 4313–4327.
49.
ChenKJiJWangHQLiuY and SongZH: ‘Adaptive local kernel-based learning for soft sensor modelling of nonlinear processes’, Chem. Eng. Res. Des., 2011, 89, 2117–2124.
50.
CawleyGC and TalbotNLC: ‘Fast exact leave-one-out cross validation of sparse least-squares support vector machines’, Neural Networks, 2004, 17, 1467–1475.
51.
HuangDWangCD and LaiJH: ‘Incremental support vector clustering with outlier detection’, Proc. 21st Int. Conf. on ‘Pattern recognition (ICPR)’, 2012, 2339–2342.
52.
ChenT and RenJH: ‘Bagging for Gaussian process regression’, Neurocomputing, 2009, 72, 1605–1610.
53.
ChenJChanLLT and ChengN: ‘Gaussian process regression based optimal design of combustion systems using flame images’, Appl. Energy, 2013, 111, 153–160.
54.
LiuY and ChenJ: ‘Integrated soft sensor using just-in-time support vector regression and probabilistic analysis for quality prediction of multi-grade processes’, J. Process Contr., 2013, 23, 793–804.
