Forecasting and optimal probabilistic scheduling of surplus gas systems in iron and steel industry

Abstract

To make full use of the gas resource, stabilize the pipe network pressure, and obtain higher economic benefits in the iron and steel industry, the surplus gas prediction and scheduling models were proposed. Before applying the forecasting techniques, a support vector classifier was first used to classify the data, and then the filtering was used to create separate trend and volatility sequences. After forecasting, the Markov chain transition probability matrix was introduced to adjust the residual. Simulation results using surplus gas data from an iron and steel enterprise demonstrate that the constructed SVC-HP-ENN-LSSVM-MC prediction model prediction is accurate, and that the classification accuracy is high under different conditions. Based on this, the scheduling model was constructed for surplus gas operating, and it has been used to investigate the comprehensive measures for managing the operational probabilistic risk and optimize the economic benefit at various working conditions and implementations. It has extended the concepts of traditional surplus gas dispatching systems, and provides a method for enterprises to determine optimal schedules.

Keywords

surplus gas prediction / probabilistic scheduling / iron and steel enterprise / HP filter / Elman neural network (ENN) / least squares support vector machine (LSSVM) / Markov chain

References

[1] ZhangQ, CaiJ-j, WangJ-j, WuF-z, DongHui. Reasonable utilization of byproduct gases in metallurgical industry [J]. The Proceedings of the China Association for Science and Technology, 2007, 4(2): 464-468

[2] ZhangX-p, ZhaoJ, WangW, FengW-m, ChenW-chang. Multi-output least squares support-vector-machine for level prediction in Linz Donaniz gas holder [J]. Control Theory & Applications, 2010, 27(11): 1463-1470

[3] ZhangX-p, ZhaoJ, WangW, CongL-q, FengW-m, ChenW-chang. COG holder level prediction model based on least square support vector machine and its application [J]. Control and Decision, 2010, 25(8): 1178-1188

[4] ZhangX-p, ZhaoJ, WangW, CongL-q, FengW-min. An optimal method for prediction and adjustment on byproduct gas holder in steel industry [J]. Expert Systems with Applications, 2011, 38: 4588-4599

[5] DuttaG, SinhaG P, RoyP, MitterN. A linear programming model for distribution of electrical energy in a steel plant [J]. International Transactions in Operational Research, 2003, 1(1): 17-29

[6] KimJ H, YiH S, HanC. Plant wide optimal byproduct gas distribution and holder level control in the iron and steel making process [J]. Korean Journal of Chemical Engineering, 2003, 20(3): 429-435

[7] KimJ H, YiH S, HanC. A novel MILP model for plant wide multiperiod optimization of byproduct gas supply system in the iron and steel making process [J]. Chemical Engineering Research Design, 2003, 81(8): 1015-1025

[8] JeongC, ChuY H, HanC, YoonE S. Gasholder level control based on time-series analysis and process heuristics[J]. Korean Journal of Chemical Engineering, 2011, 28(1): 16

[9] ChaH, HanX-s, WangY, ZhangLi. Study of power system probabilistic dispatching with security-economy coordination [J]. Proceedings of the Chinese Society for Electrical Engineering, 2009, 29(13): 16-21

[10] LiZ-y, WuW-lin. Classification of power quality combined disturbances based on phase space reconstruction and support vector machines [J]. Journal of Zhejiang University, 2008, 9(2): 173-181

[11] YuanC-ming. Forecasting exchange rates: The multi-state Markov-switching model with smoothing [J]. International Review of Economics and Finance, 2011, 20: 342-362

[12] LiuL-m, WangA-n, ShaM, ZhaoF-yun. Multi-class classification methods of cost-conscious LS-SVM for fault diagnosis of blast furnace [J]. Journal of Iron and Steel Research, International, 2011, 18(10): 17-23

[13] BaoY-k, LiuZ-t, GuoL, WangWen. Forecasting stock composite index by fuzzy support vector machines regression [J]. Machine Learning and Cybernetics, 2005, 6: 3535-3540

[14] ChenX-j, LiY, RobertH, ZhangY-qing. Type-2 fuzzy logic-based classifier fusion for support vector machines [J]. Applied Soft Computing, 2008, 8(3): 1222-1231

[15] TorbenM P. The Hodrick-Prescott filter, the Slutzky effect, and the distortionary effect of filters [J]. Journal of Economic Dynamics & Control, 2001, 25: 1081-1101

[16] MaravallA, AnaR. Temporal aggregation, systematic sampling, and the Hodrick-Prescott filter [J]. Computational Statistics & Data Analysis, 2007, 52: 975-998

[17] ZhaoJ, ZhuX-l, WangW, LiuYing. Extended Kalman filter-based Elman networks for industrial time series prediction with GPU acceleration [J]. Neurocomputing, 2013, 118(22): 215

[18] WanW, XuH, ZhangW-h, HuX-c, DengGang. Questionnaires-based skin attribute prediction using Elman neural network [J]. Neurocomputing, 2011, 74(17): 2834-2841

[19] LiJ-y, MengX-feng. Temperature decoupling control of double-level air flow field dynamic vacuum system based on neural network and prediction principle [J]. Engineering Applications of Artificial Intelligence, 2013, 26(4): 1237-1245

[20] LiouC-y, HuangJ-c, YangW-chie. Modeling word perception using the Elman network [J]. Neurocomputing, 2008, 71(16/17/18): 3150-3157

[21] ChangY-w, WangY-c, LiuT, WangZ-jie. Fault diagnosis of a mine hoist using PCA and SVM techniques [J]. Journal of China University of Mining & Technology, 2008, 18(3): 327-331

[22] HongW-chiang. Hybrid evolutionary algorithms in a SVR based electric load forecasting model [J]. International Journal of Electrical Power and Energy Systems, 2009, 31(7/8): 409-417

[23] HongW-chiang. Chaotic particle swarm optimization algorithm in a support vector regression electric load forecasting model [J]. Energy Conversion and Management, 2009, 50(1): 105-117

[24] HongW-chiang. Electric load forecasting by support vector model [J]. Applied Mathematical Modelling, 2009, 33(5): 2444-2454

[25] PaiP-f, HongW-chiang. Support vector machines with simulated annealing algorithms in electricity load forecasting [J]. Energy Conversion and Management, 2005, 46(17): 2669-2688

[26] PaiP-f, HongW-chiang. Forecasting regional electric load based on recurrent support vector machines with genetic algorithms [J]. Electric Power Systems Research, 2005, 74(3): 417-425

[27] HongW-chiang. Application of chaotic ant swarm optimization in electric load forecasting [J]. Energy Policy, 2010, 38(10): 5830-5839

[28] HongW-chiang. Rainfall forecasting by technological machine learning models [J]. Applied Mathematics and Computation, 2008, 200(1): 41-57

[29] HongW-c, PaiP-feng. Potential assessment of the support vector regression technique in rainfall forecasting [J]. Water Resour Manag, 2007, 21(2): 495-513

[30] SUYKENS J A K, GESTEL T V, BRABANTER J D, MOOR D, VANDEWALLE J. Least squares support vector machines [M]. Singapore: World Scientific Pub. Co., 2002: 117-144.

[31] WangS, YuL, TangL, WangS-yang. A novel seasonal decomposition based least squares support vector regression ensemble learning approach for hydropower supply forecasting in China [J]. Energy, 2011, 36(11): 6542-6555

[32] SamuiP, KothariD P. Utilization of a least square support vector machine (LSSVM) for slope stability analysis [J]. Scientia Iranica, 2011, 18(1): 53-58

[33] TangL, YuL, WangS, LiJian. Novel hybrid ensemble learning paradigm for nuclear energy supply forecasting [J]. Applied Energy, 2012, 93(5): 432-443

[34] ZhaoJ, LiuY, ZhangX-ping. A MKL based on-line prediction for gasholder level in steel industry [J]. Control Engineering Practice, 2012, 20(6): 629-641

[35] VapnikV NThe nature of statistical learning theory [M], 1999, New York, Springer: 181-190

[36] SuykensJ A K, VandewalleJLeast square support vector machine classifiers [M], 1995, Berlin, Neural Processing Letters: 293-300