Flux-measuring approach of high temperature metal liquid based on BP neural networks

Abstract

A soft-measuring approach is presented to measure the flux of liquid zinc with high temperature and causticity. By constructing mathematical model based on neural networks, weighing the mass of liquid zinc, the flux of liquid zinc is acquired indirectly, the measuring on line and flux control are realized. Simulation results and industrial practice demonstrate that the relative error between the estimated flux value and practical measured flux value is lower than 1.5%, meeting the need of industrial process.

Keywords

flux / high temperature metal liquid / flux-measuring / neural networks

References

[1] LehtokangasM. Modelling with constructive backpropagation [J]. IEEE Trans neural networkss, 1999, 10(12): 707-716

[2] ReyesJ S, DuroR J. Influence of noise on discrete time backpropagation trained networks [J]. Neural Computer, 2001, 41(1): 67-89

[3] WangZheng’ou, LinChen. A fast learning algorithm of feedforward neural networkss and its application to system identification[J]. Acta Automatic Sinica, 1997, 23(6): 728-735(in Chinese)

[4] TanHao, LiLi-yuan, ChenWei-nan. Burner flame recognition based on backpropagation neural networks identification of burning state of boiler hearth flame on BP network[J]. Acta Automatic Sinica, 1998, 24(5): 667-670(in Chinese)

[5] GuoChuang-xin, JingLei, LiangNian-sheng, et al.. A robust BP algorithm and its application on identification of non-linear mechanics system [J]. Information and Control, 1996, 25(6): 354-360(in Chinese)

[6] LaiXiao-ping, ZhouHong-xing. Self-learning neurons and self-learning BP networks [J]. Information and Control, 1999, 28(6): 407-415(in Chinese)

[7] LiGuo-jun, ZhanBo-qiongEngineering Liquid Mechanics [M], 1991, Wuhan, Huzhong University of Science and Technology Press

[8] LiuDing, WanBai-wu. Model identification and steady-state optimizing control of wind system for air preheater [J]. Acta Automatic Sinica, 1999, 25(6): 321-324(in Chinese)

[9] WangChao-jun, LiuYan-ming. An optimization controller in milling compacting genetic algorithm with neural networkss [J]. Control Theory and Applications, 1999, 16(4): 607-610(in Chinese)

[10] WangYin, RongGang. A self-organizing neural-network-based fuzzy system [J]. Control Theory and Application, 1999, 16(6): 455-457