Advertisement

Dynamic forecasting and optimal scheduling of by-product gases in integrated iron and steel works

  • Qi ZhangEmail author
  • Hui Li
  • Jia-lin Ma
  • Hua-yan Xu
  • Bo-yang Yu
  • Gang Wang
  • Shan Jiang
Original Paper
  • 58 Downloads

Abstract

The by-product gases, which are generated in ironmaking, coking and steelmaking processes, can be used as fuel for the metallurgical processes and on-site power plants. However, if the supply and demand of by-product gases are imbalanced, gas flaring may occur, leading to energy wastage and environmental pollution. Therefore, optimal scheduling of by-product gases is important in iron and steel works. A BP_LSSVM model, which combines back-propagation (BP) neural network and least squares support vector machine (LSSVM), and an improved mixed integer linear programming model were proposed to forecast the surplus gases and allocate them optimally. To maximize energy utilization, the stability of gas holders and boilers was considered and a concise heuristic procedure was proposed to assign penalties for boilers and gas holders. Moreover, the optimal level of gas holder was studied to enhance the stability of the gas system. Compared to the manual operation, the optimal results showed that the electricity generated by the power plant increased by 2.93% in normal condition and by 22.2% in overhaul condition. The proposed model minimizes the total cost by optimizing the boiler load with less adjustment frequency and the stability of gas holders and can be used as a guidance in dynamic forecasting and optimal scheduling of by-product gases in integrated iron and steel works.

Keywords

Iron and steel works Back-propagation neural network Least squares support vector machine Mixed integer linear programming Dynamic forecasting Optimal scheduling 

List of symbols

B

{b | Boilers}

G

{g | Gases}

GH

{k | Gas holders}

P

{t | Periods}

TB

{tb | Turbines}

\(a_{tb}\)

First regression parameter from steam production of turbine tb

\(b_{tb}\)

Second regression parameter from steam production of turbine tb

\(C_{\text{c}}\)

Unit price for coal (RMB/t)

\(C_{\text{e}}\)

Unit price for electricity (RMB/kWh)

\(C_{g}\)

Economic benefit of gas g by generating electricity (RMB/m3)

\(C_{\text{w}}\)

Unit price for water (RMB/t)

\(C_{{1\,{\text{kJ}}}}\)

Economic benefit of 1 kJ thermal energy by generating electricity (RMB/kJ)

\(D_{st}^{\text{pro}}\)

Amount of steam supplied to steelmaking process (m3/h)

\(D_{st}\)

Steam demand in steelmaking process (m3/h)

Etb, t

Electricity generated by turbine tb in period t (RMB/kWh)

\(f_{b}^{ \hbox{min} }\)

Minimum heat value of boiler b combustion in period t (kJ/m3)

\(f_{b}^{ \hbox{max} }\)

Maximum heat value of boiler b combustion in period t (kJ/m3)

\(f_{t,b}\)

Heat value of mixed fuel used by boiler b in period t (kJ/m3)

\(F\)

Total energy consumption released by fuel combustion (kJ)

\(F_{{{b}},{t}}^{\text{v}}\)

Heat supplied by fuel to boiler b in period t (kJ/h)

\(F_{t}^{g}\)

Surplus of gas g in period t (m3/h)

\(h_{k}^{ \hbox{min} }\)

Lowest level of gas holder k (m3)

\(h_{k}^{ \hbox{max} }\)

Highest level of gas holder k (m3)

\(h_{t,k}\)

Level of gas holder k in period t (m3)

\(H_{g}\)

Heat value of gas g (kJ/m3)

\(H_{st}\)

Enthalpy of steam (kJ/t)

\(H_{\text{w}}\)

Enthalpy of water (kJ/t)

\(I\)

Binary variables that indicate boiler load fluctuation

\(T_{b}^{ \hbox{min} }\)

Minimum load of boiler b in period t (m3/h)

\(T_{b}^{ \hbox{max} }\)

Maximum load of boiler b in period t (m3/h)

\(T_{t,tb}\)

Steam get into steam turbine tb in period t (t/h)

\(T_{t,tb}^{tb}\)

Steam consumed by turbine tb in period t (t/h)

\(T_{{i,t,tb}}^{\text{dem}}\)

Sum of steam i consumed in steelmaking process (t/h)

\(T_{t,b}\)

Production of steam in boiler b in period t (t/h)

\(v_{b,t}^{{{\text{c}},{ \hbox{min} }}}\)

Minimum coal consumption of boiler b in period t (t/h)

\(v_{b,t}^{{{\text{c}},{ \hbox{max} }}}\)

Maximum coal consumption of boiler b in period t (t/h)

\(v_{b,t}^{\text{c}}\)

Coal consumption of boiler b in period t (t/h)

\(v_{b,t}^{\text{w}}\)

Water consumption of boiler b in period t (t/h)

\(v_{k}\)

Maximum running speed of piston in gas holder k (m3/h)

\(v_{t}^{{{\text{w}},{\text{cnd}}}}\)

Condensate water recovered from steam turbine in period t (m3/h)

\(W_{b}\)

Load fluctuation penalty factor for boiler b (RMB/times)

\(W_{g}\)

Penalty factor of gas flaring for gas g (RMB/h)

\(W_{\text{h}}\)

Penalty factor for gas holder above optimal level but in stable level (RMB/m3)

\(W_{\text{hh}}\)

Penalty factor for gas holder in high level (RMB/m3)

\(W_{\text{l}}\)

Penalty factor for gas holder below optimal level but in stable level (RMB/m3)

\(W_{\text{ll}}\)

Penalty factor for gas holder in low level (RMB/m3)

\(W_{t,b}\)

Total water flowing into boiler b in period t (m3/h)

\(\Delta B_{t}^{g}\)

Gas consumption of boiler b in period t (m3/h)

\(\Delta D_{t}^{{g}}\)

Gas flaring of by-product gas g (m3/h)

\(\Delta D_{t}^{{{g}},{ \hbox{max} }}\)

Maximum gas flaring of gas g in period t (m3/h)

\(\Delta L_{t}^{g}\)

Amount of gas g entering gas tank within period t (m3/h)

\(\Delta V_{\text{h}}\)

Amounts of by-product gases above optimal levels (m3/h)

\(\Delta V_{\text{hh}}\)

Amounts of by-product gases that deviate from high level (m3/h)

\(\Delta V_{\text{l}}\)

Amounts of by-product gases below optimal levels (m3/h)

\(\Delta V_{\text{ll}}\)

Amounts of by-product gases that deviate from low level (m3/h)

\(\Delta V_{t,k}\)

Level variation of gas holder k during period t (m3)

\(\Delta t\)

Duration of a period (h)

\(\eta_{b}\)

Efficiency of boiler b

Notes

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 51874095) and by the National Key Research and Development Program (Project Nos. 2016YFB0601305 and 2016YFB0601301). The authors gratefully acknowledge the reviewers and editors for their fruitful comments.

References

  1. [1]
    National Bureau of Statistics, China Statistical Yearbook 2016, China Statistics Press, Beijing. http://www.stats.gov.cn/tjsj/ndsj/2016/indexch.htm (Accessed: 11 Feb. 2017).
  2. [2]
    Q. Zhang, X.Y. Zhao, H.Y. Lu, T.J. Ni, Y. Li, Appl. Energy 191 (2017) 502–520.CrossRefGoogle Scholar
  3. [3]
    Q. Zhang, Y. Li, J. Xu, G.Y. Jia, J. Clean. Prod. 172 (2018) 709–723.CrossRefGoogle Scholar
  4. [4]
    M. Larsson, C. Wang, J. Dahl, Appl. Therm. Eng. 26 (2006) 1353–1361.CrossRefGoogle Scholar
  5. [5]
    Q. Zhang, J.J. Cai, J.J. Wang, H. Dong, X.L. Zhang, Iron and Steel 44 (2009) No. 12, 95–99.Google Scholar
  6. [6]
    Z. Liu, X. Mao, J. Tu, M. Jaccard, J. Environ. Manag. 144 (2014) 135–142.CrossRefGoogle Scholar
  7. [7]
    W.R. Morrow III, A. Hasanbeigi, J. Sathaye, T. Xu, J. Clean. Prod. 65 (2014) 131–141.CrossRefGoogle Scholar
  8. [8]
    H.N. Kong, J. Iron Steel Res. Int. 22 (2015) 681–685.CrossRefGoogle Scholar
  9. [9]
    K. Makinen, T. Kymalainen, J. Junttila, in: Industrial Energy Technology Conference, Energy Systems Laboratory, Texas A&M University, Texas, USA, 2012, pp. 4640–4645.Google Scholar
  10. [10]
    J.H Kim, C. Han, Ind. Eng. Chem. Res. 40 (2001) 1928–1939.CrossRefGoogle Scholar
  11. [11]
    J.H. Kim, H.S. Yi, C. Han, Chem. Eng. Res. Des. 81 (2003) 1015–1025.CrossRefGoogle Scholar
  12. [12]
    N. Wang, R. Chen, in: 2012 International Conference on ICT Convergence, IEEE, Jeju Island, South Korea, 2012, pp. 615–619.Google Scholar
  13. [13]
    F. Pettersson, N. Chakraborti, H. Saxén, Appl. Soft Comp. 7 (2007) 387–397.CrossRefGoogle Scholar
  14. [14]
    N.A. Zambri, A. Mohamed, M.Z.C. Wanik, Int. J. Electr. Power Energy Syst. 67 (2015) 179–190.CrossRefGoogle Scholar
  15. [15]
    J. Zhao, W. Wang, K. Sun, Y. Liu, IEEE Trans. Autom. Sci. Eng. 11 (2014) 1149–1154.Google Scholar
  16. [16]
    R.E. Markland, J. Oper. Manag. 1 (1980) 95–102.CrossRefGoogle Scholar
  17. [17]
    K. Akimoto, N. Sannomiya, Y. Nishikawa, T. Tsuda, Automatica 27 (1991) 513–518.CrossRefGoogle Scholar
  18. [18]
    J.H. Kim, H.S. Yi, C. Han, Korean J. Chem. Eng. 20 (2003) 429–435.CrossRefGoogle Scholar
  19. [19]
    Q. Zhang, W. Ti, J.J. Cai, T. Du, A.H. Wang, J. Iron Steel Res. Int. 18 (2011) No. 8, 37–41.CrossRefGoogle Scholar
  20. [20]
    X.C. Zhao, B. Bai, X. Lu, Q. Shi, J.H. Han, Appl. Energy 148 (2015) 142–158.CrossRefGoogle Scholar
  21. [21]
    X.C. Zhao, H. Bai, Q. Shi, X. Lu, Z.H. Zhang, Appl. Energy 195 (2017) 100–113.CrossRefGoogle Scholar
  22. [22]
    V.B. de Oliveira Junior, J.G.C. Pena, J.L.F. Salles, Appl. Energy 164 (2016) 462–474.Google Scholar
  23. [23]
    J.H. Yang, J.J. Cai, W.Q. Sun, J. Huang, Appl. Therm. Eng. 115 (2017) 586–596.CrossRefGoogle Scholar
  24. [24]
    J.H. Yang, J.J. Cai, W.Q. Sun, J.Y. Liu, J. Iron Steel Res. Int. 22 (2015) 408–413.CrossRefGoogle Scholar
  25. [25]
    J. Zhao, X. Zhu, W. Wang, Y. Liu, Neurocomputing 118 (2013) 215–224.CrossRefGoogle Scholar
  26. [26]
    Z. Han, Y. Liu, J. Zhao, W. Wang, Control Eng. Pract. 20 (2012) 1400–1409.CrossRefGoogle Scholar
  27. [27]
    L. Li, H.J. Li, J. Cent. South Univ. 22 (2015) 1437–1447.CrossRefGoogle Scholar

Copyright information

© China Iron and Steel Research Institute Group 2019

Authors and Affiliations

  1. 1.State Environmental Protection Key Laboratory of Eco-IndustryNortheastern UniversityShenyangChina
  2. 2.State Key Laboratory of Hybrid Process Industry Automation Systems and Equipment TechnologyAutomation Research and Design Institute of Metallurgical IndustryBeijingChina
  3. 3.Software CollegeNortheastern UniversityShenyangChina
  4. 4.Shougang Jingtang United Iron & Steel Co., Ltd.TangshanChina
  5. 5.China United Engineering CompanyHangzhouChina

Personalised recommendations