Abstract: The hot-dip galvanizing line (HDGL) is a typical order-driven discrete-event process in steelmaking. It has some complicated dynamic characteristics such as a large time-varying delay, strong nonlinearity, and unmeasured disturbance, all of which lead to the difficulty of an online coating weight controller design. We propose a novel neural network based control system to solve these problems. The proposed method has been successfully applied to a real production line at VaLin LY Steel Co., Loudi, China. The industrial application results show the effectiveness and efficiency of the proposed method, including significant reductions in the variance of the coating weight and the transition time.

基于神经网络的热镀锌层厚控制系统的研究与应用

[1]Adams J, Miles LB, Parker DJ, et al., 1996. Coating mass control on No. 2 galvanizing line at LTV steel’s Indiana Harbor works. Iron Steel Eng, 73(1):123-131.

[2]Bloch G, Sirou F, Eustache V, et al., 1997. Neural intelligent control for a steel plant. IEEE Trans Neur Netw, 8(4): 910-918.

[3]Elsaadawy EA, Hanumanth GS, Balthazaar AKS, et al., 2007. Coating weight model for the continuous hot-dip galvanizing process. Metall Mater Trans B, 38(3):413-424.

[4]Fei J, Zhang Y, Wang JS, et al., 2016. Development and application of coating thickness control system for cold rolling continuous galvanizing line. Iron Steel, 51(5): 57-61 (in Chinese).

[5]Guelton N, Lerouge A, 2010. Coating weight control on ArcelorMittal’s galvanizing line at Florange Works. Contr Eng Pract, 18(10):1220-1229.

[6]Guelton N, Lopès C, Sordini H, 2016. Cross coating weight control by electromagnetic strip stabilization at the continuous galvanizing line of ArcelorMittal Florange. Metall Mater Trans B, 47(4):2666-2680.

[7]Jordan CE, Goggins KM, Benscoter AO, et al., 1993. Metallographic preparation technique for hot-dip galvanized and galvannealed coatings on steel. Mater Charact, 31(2):107-114.

[8]Lu YZ, 1996. Industrial Intelligent Control: Fundamentals and Applications. Wiley, New York, USA.

[9]Lu YZ, Markward SW, 1997. Development and application of an integrated neural system for an HDCL. IEEE Trans Neur Netw, 8(6):1328-1337.

[10]Marder AR, 2000. The metallurgy of zinc-coated steel. Prog Mater Sci, 45(3):191-271.

[11]Martínez-de-Pisón FJ, Pernía A, Jiménez-Macías EB, et al., 2010. Overall model of the dynamic behaviour of the steel strip in an annealing heating furnace on a hot-dip galvanizing line. Rev Metal, 46(5):405-420.

[12]Martínez-de-Pisón FJ, Celorrio L, Pérez-de-la-Parte M, et al., 2011. Optimising annealing process on hot dip galvanising line based on robust predictive models adjusted with genetic algorithms. Iron Steel, 38(3):218-228.

[13]Pal D, Datta A, Sahay SS, 2006. An efficient model for batch annealing using a neural network. Mater Manuf Process, 21(5):567-572.

[14]Sanz-García A, Fernández-Ceniceros J, Fernández-Martínez R, et al., 2012. Methodology based on genetic optimisation to develop overall parsimony models for predicting temperature settings on annealing furnace. Iron Steel, 41(2):87-98.

[15]Shin KT, Park HD, Chung WK, 2006. Synthesis method for the modelling and robust control of coating weight at galvanizing process. ISIJ Int, 46(10):1442-1451.

[16]Thornton JA, Graff HF, 1976. An analytical description of the jet finishing process for hot-dip metallic coatings on strip. Metall Trans B, 7(4):607-618.

[17]Tu CV, Wood DH, 1996. Wall pressure and shear stress measurements beneath an impinging jet. Exp Therm Fluid Sci, 13(4):364-373.

[18]Warwick K, Rees D, 1988. Industrial Digital Control Systems. IET, London, England.

[19]Yu W, Li XO, 2008. Optimization of crude oil blending with neural networks and bias-update scheme. Eng Intell Syst, 16(1):28-37.

[20]Zhang Y, Shao FQ, Wang JS, et al., 2011. Adaptive control of coating weight for continuous hot-dip galvanizing. J Northeast Univ (Nat Sci), 32(11):1525-1528 (in Chinese).