A novel method to control gasholder levels in an iron and steel company with accurate prediction of the future trend is presented. Although various gasholders are used to recycle by-product gases generated during the ironmaking, coke-burning, and steel-making processes, the capacities of these gasholders are insufficient to handle large amounts of gases. To overcome this problem, tight control of the gasholder level should be maintained by predicting their anticipated changes. However, the current prediction logic cannot show satisfactory results due to the lack of characterization of the relevant processes. In the proposed method, time-series modeling and heuristics from industrial operators are used to correctly reflect the process characteristics and deal with unexpected process delays. By applying the proposed method to an off-line data set, a significant reduction in the discrepancy between predicted values and_x000D_ actual values was observed. The method is expected to be adopted in the prediction system of POSCO.

Level Control Time-series Modeling Prediction Iron and Steel Making Process Heuristics

Makkonen HT, Heino J, Laitila L, Hiltunen A, Poylio E, Harkki J, Res. Conserv. Recycl., 35, 77 (2002)
Worrell E, Price L, Martin N, Farla J, Schaeffer R, Energy Policy, 25(7-9), 727 (1997)
Kim K, Fuel Energy Abstr., 39, 234 (1998)
Sridhar S, Sichen D, Pal UB, Seetharaman S, JOM., 54, 30 (2002)
Bojic M, Mourdoukoutas P, Appl. Thermal Eng., 20, 963 (2000)
Prokop L, Kohut K, Fuel Energy Abstr., 39, 188 (1998)
Markland RE, J. Oper. Manage., 1, 95 (1980)
Box GEP, Jenkins GM, Reinsel GC, Time series analysis: Forecasting and control, Prentice Hall, NJ (1994)
Pandit SM and Wu S, Time series and system analysis, with applications, Wiley, New York (1983)