An integrated modeling method for prediction of sulfur content in agglomerate

Abstract

Based on the idea of fusing modeling, an integrated prediction model for sintering process was proposed. A framework for sulfur content prediction was established, which integrated multi modeling ways together, including mathematical model combined with neural network(NN), rule model based on empirical knowledge and modelchoosing coordinator. Via metallurgic mechanism analysis and material balance computation, a mathematical model calculated the sulfur content in agglomerate by the material balance equation with some parameters predicted by NN method. In the other model, the relationship between sulfur content and key factors was described in the form of expert rules. The model-choosing coordinator based on fuzzy logic was introduced to decide the weight of result of each model according to process conditions. The model was tested by industrial application data and produced a relatively satisfactory prediction error. The model also preferably reflected the varying tendency of sulfur content in agglomerate as the evidence of its prediction performance.

Keywords

prediction model / integrated modeling / neural network / material balance / expert rule

References

[1] WANG Ya-lin, GUI Wei-hua, CHEN Xiao-fang, et al. Neural network modeling for composition prediction of Pb-Zn sinter in imperial smelting process[A]. Proceedings of the Annual Chinese Automation Conference[C]. Derby, England, 1999: 21–24.

[2] EfstathiouJExpert Systems In Process Control[M], 1989, Essex, Longman Scientific & Technical: 20-27

[3] WuMin, GuiWei-hua. An expert control and fault diagnosis scheme for the leaching zinc process [J]. Control Theory & Applications, 2001, 18(6): 861-866

[4] HaganM T, DemuthH B, BealeM HNeural Network Design [M], 1996, Boston, PWS Publishing Co: 5-17

[5] ChenS, BillingsS A. Neural networks for nonlinear dynamic system modeling and identification [J]. Int J Control, 1992, 56(2): 319-346

[6] WangYa-lin, GuiWei-hua, YangChun-hua, WuMin. Adaptive supervised distributed neural networks and its industrial application [J]. Control and Decision, 2001, 16(5): 549-552(in Chinese)

[7] WangHong-wei, MaGuang-fu. Nonlinear system modeling via fuzzy logic rules [J]. Control Theory & Applications, 2000, 17(3): 419-422(in Chinese)

[8] DengHui, SunZen-qi, SunFu-chun. The fuzzy cluster identification algorithm [J]. Control Theory & Applications, 2001, 18(2): 171-175(in Chinese)

[9] TangJia-fu, WangDing-wei. Fuzzy optimization theory and methodology: a survey [J]. Control Theory & Applications, 2000, 17(2): 159-164(in Chinese)

[10] WuMin, NakanoM, SheJin-hua. A model-based expert control strategy using neural networks for the coal blending process in an iron and steel plant [J]. Expert Systems with Applications, 1999, 16(6): 271-281

[11] WANG Ya-lin, GUI Wei-hua, YANG Chun-hua, et al. A multi-objective optimization for the mixture ratio of Pb-Zn sintering process based on intelligent integrated model [A]. The Third Chinese World Congress on Intelligent Control and Intelligent Automation[C]. Hefei, China, 2000: 92–97.

[12] HiroshiT. Extracting rules from trained neural networks [J]. IEEE Transactions on Neural Network, 2000, 11(2): 377-389

[13] MedskerL RHybrid Neural Networks and Expert Systems[M], 1994, Boston, Kluwer Academic Publisher: 31-54

[14] HaganM T, MenhajM. Training feed forward network with the Marquardt algorithm [J]. IEEE Trans. on neural network, 1994, 5(6): 989-993

[15] LiJie, HanZheng-zhi. the learning error function of neural network and its generalization [J]. Control and Decision, 2000, 15(1): 95-97(in Chinese)