Soft Computing

, Volume 23, Issue 19, pp 9551–9571 | Cite as

Furnace operation optimization with hybrid model based on mechanism and data analytics

  • Qiong Xia
  • Xianpeng WangEmail author
  • Lixin Tang
Methodologies and Application


The operation optimization problem of furnace process (OOPFP) is to obtain an optimal setting of furnace temperatures to make the reheated slabs have suitable temperature distribution with minimum energy consumption and oxide loss. Since the furnace process is complex, dynamic, and nonlinear, it is difficult to establish a precise process model for the OOPFP. In this paper, firstly, a novel hybrid modeling method combining temperature mechanism with data analytics is proposed to map the relationship of the temperature variables in the furnace process more accurately. The main idea of the hybrid modeling is to compensate for the deviation of original temperature mechanism model through the least squares support vector machine. Then, an improved differential evolution (DE) with feasible region dynamic adjustment strategy and population size gradual shrinking strategy is proposed to solve the OOPFP under hybrid model. Finally, extensive numerical experiments are carried out to demonstrate the effectiveness and accuracy of the hybrid model, and evaluate the performance of the improved DE on solving OOPFP. Experimental results show that the hybrid model is more precise than the original mechanism one, and the proposed DE outperforms other representative algorithms, respectively.


Differential evolution Hybrid modeling Least squares support vector machine Operation optimization Reheating furnace 



This research is supported by the National Key Research and Development Program of China (2016YFB0901900), the Major Program of National Natural Science Foundation of China (71790614), the Fund for Innovative Research Groups of the National Natural Science Foundation of China (71621061), the Major International Joint Research Project of the National Natural Science Foundation of China (71520107004), the Fund for the National Natural Science Foundation of China (61374203, 61573086), and the 111 Project (B16009).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.


  1. Axelberg PGV, Gu YH, Bollen MHJ (2007) Support vector machine for classification of voltage disturbances. IEEE Trans Power Deliv 22(3):1297–1303CrossRefGoogle Scholar
  2. Cao LJ, Tay FEH (2003) Support vector machine with adaptive parameters in financial time series forecasting. IEEE Trans Neural Netw 14(6):1506–1518CrossRefGoogle Scholar
  3. Cao LL, Sun YP, Wu HY (2011) Grey-box modeling based on fundamental genes for nonlinear system. J Central South Univ Sci Technol 42(2):414–418Google Scholar
  4. Das S, Suganthan PN (2011) Differential evolution: a survey of the state-of-the-art. IEEE Trans Evol Comput 15(1):4–31CrossRefGoogle Scholar
  5. Dong Y, Zhao R (2017) Solve train stowage planning problem of steel coil using a pointer-based discrete differential evolution. J Prod Res, Int. Google Scholar
  6. Dong B, Li ZX, Rahman SMM, Vega R (2016) A hybrid model approach for forecasting future residential electricity consumption. Energy Build 117:341–351CrossRefGoogle Scholar
  7. El-Naqa I, Yang YY, Wernick MN et al (2002) A support vector machine approach for detection of microcalcifications in mammograms. In: IEEE international conference on image processing, Nishikawa, pp 953–956Google Scholar
  8. Guo XC, Yang JH, Wu CG et al (2008) A novel LS-SVMS hyper-parameter selection based on particle swarm optimization. Neurocomputing 71:3211–3215CrossRefGoogle Scholar
  9. Li QY, Zhou H, Lin AP et al (2011) Prediction of ash fusion temperature based on grid search and support vector machine. J Zhejiang Univ Sci A 45(12):2181–2187Google Scholar
  10. Liang JJ, Qu BY, Suganthan PN (2014) Problem definitions and evaluation criteria for the CEC 2014 special session and competition on single objective real-parameter numerical optimization. In: 2014 IEEE Congress on Evolutionary Computation.Google Scholar
  11. Lin SW, Ying KC et al (2008) Particle swarm optimization for parameter determination and feature selection of support vector machines. Expert Syst Appl 35:1817–1824CrossRefGoogle Scholar
  12. Liu DW, Lin XH, Ghosh D (2007) Semiparametric regression of multidimensional genetic pathway data: least-squares kernel machines and linear mixed models. Biometrics 63:1079–1088MathSciNetCrossRefzbMATHGoogle Scholar
  13. Luo N, Li N, Qian F (2007) Support vector machine and mechanism model hybrid modeling method and its application to polyester production process. Polyest Ind 20(2):12–18Google Scholar
  14. Mao WT, Yan GR et al (2011) Model selection for least squares support vector regressions based on small-word strategy. Expert Syst Appl 38:3227–3237CrossRefGoogle Scholar
  15. Mathias MA, Mohamed C, Alain B (2011) Semisupervised learning using Bayesian interpretation: application to LS-SVM. IEEE Trans Neural Netw 22(4):513–524CrossRefGoogle Scholar
  16. Michalak M (2011) Adaptive kernel approach to the time series prediction. Pattern Anal Appl 14(3):283–293MathSciNetCrossRefGoogle Scholar
  17. Pajares G, Cruz JMDL (2004) On combining support vector machines and simulated annealing in stereovision matching. IEEE Trans Syst Man Cybern Part B Cybern 34(4):1646–1657CrossRefGoogle Scholar
  18. Rong JG, Chen F (2005) Data fusion for nutrient solution components detection based on SVM. Comput Simul 22(11):110–112Google Scholar
  19. Storn R, Price K (1995) A simple and efficient adaptive scheme for global optimization over continuous spaces. In: International computer science institute, Berkeley, CA, Technical Report TR–95–012Google Scholar
  20. Suykens JAK (2001) Nonlinear modelling and support vector machines. In: IEEE instrumentation and measurement technology conference, pp 287–294Google Scholar
  21. Suykens JAK, Vandewalle J (1999) Least squares support vector machine classifiers. Neural Process Lett 9(3):293–300CrossRefGoogle Scholar
  22. Tang ZH, Yang Y (2014) Two-stage particle swarm optimization-based nonlinear model predictive control method for reheating furnace process. ISIJ Int 54(8):1836–1842CrossRefGoogle Scholar
  23. Tang LX, Zhao Y, Liu JY (2014) An improved differential evolution algorithm for practical dynamic scheduling in steelmaking-continuous casting production. IEEE Trans Evol Comput 18(2):209–225CrossRefGoogle Scholar
  24. Tang LX, Dong Y, Liu JY (2015) Differential evolution with an individual-dependent mechanism. IEEE Trans Evol Comput 19(4):560–574CrossRefGoogle Scholar
  25. Tang LX, Wang XP, Dong ZM (2018) Adaptive multi-objective differential evolution with reference axis vicinity mechanism. IEEE Trans Cyber. Google Scholar
  26. Tian HX, Mao ZZ, Wang Y (2008) Hybrid modeling of molten steel temperature prediction in LF. ISIJ Int 48(1):58–62CrossRefGoogle Scholar
  27. Vapnik VN (1995) The nature of statistical learning theory. Springer, New YorkCrossRefzbMATHGoogle Scholar
  28. Wang XP, Tang LX (2016) An adaptive multi-population differential evolution algorithm for continuous multi-objective optimization. Inf Sci 348:124–141CrossRefGoogle Scholar
  29. Wang XF, Chen JD, Liu C, Pan F (2010) Hybrid modeling of penicillin fermentation process based on least square support vector machine. Chem Eng Res Des 88:415–420CrossRefGoogle Scholar
  30. Xia Q, Wang XP, Tang LX (2016) Operation optimization of slab reheating process based on differential evolution. ISIJ Int 56(11):2006–2015CrossRefGoogle Scholar
  31. Xing J, Jiang AN et al (2010) On the DE-SVM-based forecast of rock stratum drillability. J Northeast Univ (Nat Sci) 31(9):1345–1348Google Scholar
  32. Xu G, Yu H et al (2005) Modeling fermentation process based on hybrid support vector machine-kinetics mechainsm. J Chem Ind 56(4):653–658Google Scholar
  33. Xu CJ, Huang H, Ye SJ (2014) A differential evolution with replacement strategy for real-parameter numerical optimization. In: IEEE congress on evolutionary computation, Beijing, July 6–11Google Scholar
  34. Zhao ZW, Yang JM et al (2016) A differential evolution algorithm with self-adaptive strategy and control parameters based on symmetric Latin hypercube design for unconstrained optimization problems. Eur J Oper Res 250:30–45MathSciNetCrossRefzbMATHGoogle Scholar
  35. Zhou LL, Zheng ZG et al (2009) Research on work roll wear model based on combination of mechanics and work condition about four-high tandem cold mill. J Yanshan Univ 33(4):284–293Google Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Liaoning Engineering Laboratory of Operations Analytics and Optimization for Smart IndustryNortheastern UniversityShenyangChina
  2. 2.State Key Laboratory of Synthetical Automation for Process IndustriesNortheastern UniversityShenyangChina
  3. 3.Liaoning Key Laboratory of Manufacturing System and LogisticsNortheastern UniversityShenyangChina
  4. 4.Institute of Industrial and Systems EngineeringNortheastern UniversityShenyangChina

Personalised recommendations