摘要
高速线材水冷控制系统是高速线材自动化控制的重要组成部分,是实现轧线温度的智能控制,提高高线产品质量的关键。利用先进的计算机技术来完善水冷控制系统,可以提高高速线材产品质量,从而提高线材生产制造业的产品档次,实现传统钢铁企业的产业升级。在目前竞争激烈的高线材领域,高质量高规格的产品,可以开拓更广阔海内外的市场。因此,研究如何提高高速线材水冷控制系统的性能,特别是如何提高水冷控制系统的可靠性和控制精度具有很重要的现实意义。
本文针对武汉钢铁集团公司大型轧钢厂在高速线材生产线中存在的水冷控制系统的可靠性差,轧线温度波动范围大等问题进行研究。应用智能计算理论及方法对上述工业控制系统进行系统辨识、建模以及优化,解决了高速线材生产线水冷控制系统中存在的温度控制精度波动范围大和温度控制系统可靠性差的问题,具体研究内容如下：
针对高速线材生产线水冷控制系统中存在的温度控制精度波动范围大的问题,提出了一种具有学习自适应性且能逼近任意复杂函数的前馈多层感知器,通过建立基于前馈网络水冷控制系统的优化模型。通过大量采样SMS水冷控制系统的输入输出数据,在数据特征分析与预处理的基础上建立训练样本集,以高线水冷系统的温度控制环节为对象,以高线水冷系统的入口温度作为输入,精轧机出口温度作为输出,构建水冷系统的温度输入输出神经网络模型,实现了高线水冷控制系统的自适应温度控制。在仿真数据和实际数据上,实践表明,该系统投入应用后,轧线温度的控制精度有显著提高,轧线精轧机入口温度的波动范围由±60℃降低至±28℃,精轧机出口温度的波动范围由±30℃降低至±15℃。
针对现有高速线材生产线水冷控制系统中实时性和精度难以平衡的问题,提出了两类基于膜算法的高线水冷控制系统优化模型。膜算法是一类受细胞结构和功能启发的分布式并行计算算法,是智能优化的一个新领域。近年来,膜算法在智能优化领域受到学者的广泛关注。本文利用膜算法优化高速线材生产线水冷控制系统,特别是对其BP神经网络的权值与阈值进行优化,在大量采集整个系统运行时通讯数据的基础上,通过膜算法训练,利用MATLAB工具箱构造的BP神经网络的权值和阈值,达到提高水冷控制系统的运行稳定性与控制精度的目的。在MATLAB软件仿真上,表明经膜算法优化后的水冷控制系统在满足工业实时性要求的同时,大幅度提高了输出精度。
针对现有高速线材生产线的水冷控制系统中控制系统可靠性较差的问题,提出了一种基于贝叶斯概率网络的水冷控制系统故障判断及容错处理模型。通过贝叶斯网络自动判断采集数据的范围以及数据有效性,在线调整神经网络训练样本的数据集合,通过离线重新训练网络权值,并异步更新网络权值数据,从而在保证水冷闭环控制系统实时性的前提下,提高控制系统的自适应性与可靠性。理论表明,采用基于贝叶斯概率网络的水冷控制系统,大大提高了该系统运行的可靠性和控制精度。
根据基于前馈网络水冷控制系统的优化模型,对武钢大型轧钢厂的高速线材水冷控制系统的控制软件进行优化,从实际数据结果表明,优化后的水冷控制系统提高了运行的可靠性和控制精度,实现了产业升级。
The water cooling control system takes an important role in automatic control of high-speed wire rod mill. Since it realizes the intelligent control of temperature of rolling line, and ensures the quality of production, the water cooling control system is essential for the improvement of rolling speed. The use of advanced computer technology in the water cooling control system can enhance the monitoring, operation and automatic control level of the process of the wire rod production, thereby raising the efficiency and quality of the production. The above indicates that the improvement of the system of high-speed wire rod mills, especially the reliability and the precision of control is very meaningful.
This dissertation has carried out in-depth research into the question of the low reliability and the high range of the temperature wave in the water cooling control system in high speed steel wire production line. The work has improved the reliability of the system and the precision of the temperature control.
For the problem of the accuracy of temperature control in the water cooling control system in high speed steel wire production line, by introducing the feedforward multilayer perceptions that can approach complex function and has the ability of self-learning adaptive, we established an artificial neural network based optimization model in the water cooling control system and finally improved the accuracy of temperature control in water-cooling control system. Through a large number of input/output data samples of SMS water-cooling control system, we established the training sample set on the basis of data characteristics analysis and preprocessing, realized water-cooling control optimization system based on the Feed Forward Network (FFN). By the automatic judgment of the data gathering range and data validity with the guarantee of the real time, we established online adjustment method of the data sample collection for the neural network training in the water-cooled closed-loop control system, cooperate with offline training of network weights, and asynchronous update data network of weights. All these efforts results in improvement of the adaptability and reliability of the control system. The rolling line finishing mill inlet temperature range reduced from±60℃down to±28℃, the fluctuation range of the finishing mill exit temperature reduced from±15℃to±30℃.
For the difficult to balance the accuracy and the real-time of temperature control in the water cooling control system, we utilized two types of membrane algorithm to improve artificial neural networks based optimization model in the water cooling control system. The membrane algorithm is a distributed parallel computing algorithm inspired by the cell structure and function, which is a new area of intelligent optimization. In recent years, membrane algorithm attracts the attention of scholars in the field of intelligent optimization. Experimental results showed that our methods not only meet the real time of the industrial requirements, but also greatly improve the accuracy temperature control. On the basis of data characteristics analysis and preprocessing, we introduced a method using the MATLAB toolbox to train the BP neural network weights and thresholds with the help of membrane algorithm. On site operation shows that the use of membrane algorithm greatly enhanced the stability and control accuracy of the water cooling control system.
For the Robustness of the water cooling control system of high speed steel wire production line, we established the cooling system fault diagnosis and fault-tolerant model based on Bayesian probability network theory. Determine the range of data collection and data validation, Bayesian networks automatically adjust the neural network training sample data collection, offline re-training the network weights, and asynchronous updating of the weights of the network data, thus ensuring a water-cooled closed-loop control system in real time the premise, the self-adaptability and reliability of the network. On site operation shows that the use of Bayesian probability network optimization system greatly enhanced the reliability and control accuracy of the water cooling control system.
According to the artificial neural network based optimization model in the water cooling control system, optimizing the control software of water cooling control system for Wuhan iron and steel company, from the results of the actual data, it is shown that optimized water cooling control system can improve the operational reliability and control accuracy and realized the industrial upgrading.
引文
[1]李怡萍.简述线材轧机的发展概况.山西机械,2002,12：91-93.
[2]Zhang YT, Wang CZ, et al.. Prediction of carbon concentration and ferrite volume fraction of hot-rolled steel strip during laminar cooling. Acta Metallurgica Sinica,2007,20:251-257.
[3]张云鹏.我国高速线材生产的现状及新技术的应用.包钢科技,2008,4：4-6.
[4]Matsuyama Y, Honda T, Yamamura H, et al.. Automated solder joint inspection system using optical 3-D image detection. Proceedings 3rd IEEE Workshop on Applications of Computer Vision,1996,2:116-122.
[5]Vafaie H, Bourbakis NG A tree grammar scheme for generation and recognition of simple texture paths in pictures. Proceedings of the IEEE International Symposium on Intelligent Control,1988:201-206.
[6]张冬花,韩超,宋烈滨,詹道平,朱敏杰.高线水冷温度闭环自动控制系统.第七届中国钢铁年会论文集(下),2009,2：178-182.
[7]Trahanias P, Skordalakis E. Syntactic pattern recognition of the ECG. IEEE Transactions on Pattern Analysis and Machine Intelligence,1990:648-657.
[8]黄祺.神经网络在宝钢连铸漏钢预报系统中的应用.宝钢技术,1999,1：41-45.
[9]Poggio T, Girosi F. Networks for approximation and learning. Proceedings of the IEEE,1990,78:1481-1497.
[10]Moody J, Darken C. Fast learning in networks of locally-tuned processing units. Neural computers,1991,1:281-294.
[11]蒋迪华,文言金,王祥明,廖年.涟钢5#热风炉燃烧过程神经网络模糊控制系统.涟钢科技与管理,2002,3：34-38.
[12]Juan A, Vidal E. On the use of normalized edit distances and an efficient k-NN search technique (k-AESA) for fast and accurate string classification. Pattern Recognition, Proceedings 15th International Conference, Sept,2000,2:676-679.
[13]龙浔,高速线材活套控制系统的研究及应用.东北大学硕士论文,2005-01-01.
[14]Netten H, Young IT, et al.. Automation of fluorescent dot counting in cell nuclei. Proceedings of the 12th IAPR International Conference on Pattern Recognition,1994,1:84-87.
[15]Sulzmann A, Jacot J.3D computer graphics based interface to real microscopic worlds for μ-robot telemanipulation and position control. IEEE International Conference on Systems Man and Cybernetics Intelligent Systems for the 21st Century,1995,1:286-291.
[16]樊春玲,孙四通,金志华.贝叶斯神经网络建模预测方法及其应用.中国惯性技术学报,2009,(01)：21-24.
[17]Adiga PS Umesh, Chaudhuri BB, Rodenacker K. Semi-automatic segmentation of tissue cells from confocal microscope images. Proceedings of the 13th International Conference on Pattern Recognition,1996,3:494-497.
[18]Slater R, Healey Q Sheu P, et al.. A machine vision system for the automated classification and counting of neurons in 3-D brain tissue samples. Proceedings 3rd IEEE Workshop on Applications of Computer Vision,1996:224-229.
[19]Merkle RC. Making smaller faster cheaper computers. Proceedings of the IEEE 1998,11:2384-2386.
[20]侯昌磊.掌纹识别技术研究及应用,成都理工大学硕士论文,2003-04-01.
[21]Slater D, Healey G, Sheu P, et al.. A novel use of color computer vision methods for the quantification of neurons in 3-D brain tissue samples. IEEE International Conference on Systems Man and Cybernetics 1998,5:4411-4416.
[22]Ohba K, Ortega C, Tanie K, et al.. Real-time micro observation technique for tele-micro-operation. Proceedings of 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems,2000,1:647-652.
[23]秦英杰,王云波,张冬花.高速线材温度控制穿水冷控系统.2010年全国轧钢生产技术会议文集,2010,447-449.
[24]Bronson E. Syntactic pattern recognition of discrete utterances. Acoustics Speech and Signal Processing IEEE International Conference on ICASSP,1983, 8:719-722.
[25]Chan KP. Learning templates from fuzzy examples in structural pattern recognition. Proceedings of the Third IEEE Conference on IEEE World Congress on Computational Intelligence,1994,1:608-613.
[26]Chen CH. On the relationships between statistical pattern recognition and artificial neural networks. IEEE International Conference on Systems Man and Cybernetics,1990:182-183.
[27]曹剑叶,王云波,张冬花.高速线材生产线-水冷精确控制方法.全国高速线材生产技术交流会论文集,2007, 4：201-206.
[28]NeaPolitan R.E., Probabilistic Reasoning in Expert Systems. John Wiley & Sons, NY,1990.
[29]Buhmann J, Kuhnel H. Unsupervised and supervised data clustering with competitive neural networks. Neural Networks, International Joint Conference, June,1992,4:796-801.
[30]梁勋国,贾涛,矫志杰,王国栋,刘相华.基于贝叶斯方法的神经网络应用于冷轧轧制力预报.钢铁研究学报,2008,10：132-136.
[31]CowellR. G, DawidA. E, Lauritzen S. L., et al. Probabilistic Networks and Expert Systems. Springer,1999.
[32]彭开香,余尚志.基于贝叶斯神经网络的带钢厚度预测与控制.北京科技大学学报,2010,2：72-76.
[33]Bouckaert R. Bayesian belief networks from construction to inference. The Netherlands:Department of Computer Science, Utrecht University,1995.
[34]贾涛,刘振宇,胡恒法,王国栋.基于贝叶斯神经网络的SPA-H热轧板力学性能预测.东北大学学报(自然科学版),2008,4：81-93.
[35]Funahashi KI. On the approximate realization of continuous mappings by neural networks. Neural Networks,1989,2:183-192.
[36]Cybenko G Approximation by superpositions of a sigmoidal function. Mathematics of Control Signals and Systems,1989,2:303-314.
[37]孙岩.贝叶斯网络结构学习算法研究与应用.大连理工大学博士学位论文,2010-12-1.
[38]Peterson C. Explorations of the mean field theory learning algorithm. Hartman Eric Neural Networks,1989,2:475-494.
[39]Rumelhart DE, Hinton GE, Williams RJ. Learning internal representation by error propagation. Parallel Distributed Processing vol.1:foundations, MIT Press, Cambridge,1986:318-362.
[40]Karayiannis NB. Learning algorithms for reformulated radial basis neural networks. Neural Networks Proceedings,1998,3:2230-2235.
[41]Tan S, Hao J, Vandewalle J. Anew learning algorithm for RBF neural networks with applications to nonlinear system identification. Circuits and Systems, 1995,3:1708-1711.
[42]邵继业,王日新,徐敏强.贝叶斯网络在模型诊断中的应用.吉林大学学报(工学版),2010,1：9-22.
[43]Teodorescu HN, Bonciu C. Learning algorithm for RBF networks as features extractors. Knowledge-Based Intelligent Electronic Systems,1997,1:201-208.
[44]Paun Gh, Computing with membranes. Journal of Computer and System Sciences,2000,1(61):108-143. See also Turku Centre for Computer Science-TUCS Report No.208,1998.
[45]Leporati A, Pagani D. A membrane algorithm for the min storage problem. Lecture Notes in Computer Sci.,2006,4361:443-462.
[46]Tomassini M. Spatially Structured Evolutionary Algorithms. Springer,2005.
[47]Nishida T Y. Membrane algorithms. Lecture Notes in Computer Sci.,2006, 3850:55-66.
[48]Nishida T Y. Membrane algorithm with Brownian subalgorithm and genetic subalgorithm. Intern. J. Found. Computer Sci.,2007,18:1353-1360.
[49]Nishida T Y, Shiotani T, Takahashi Y. Membrane algorithm solving job-shop scheduling problems. Pre-Proceedings of WMC9 (P. Frisco et al., eds.), Edinburgh, UK,2008,363-370.
[50]Zaharie D, Ciobanu G. Distributed evolutionary algorithms inspired by membranes in solving continuous optimization problems. Lecture Notes in Computer Sci.,2006,4361:536-553.
[51]Kohonen T. On the significance of internal representations in neural networks. Artificial Neural Networks,1989:150-155.
[52]Wang L, Qi FH. Adaptive fuzzy Kohonen clustering network for image segmentation. Neural Networks,1999,4:2664-2667.
[53]袁晓峰,陈淑红,许化龙.贝叶斯网络推理的一种高精度仿真算法.系统仿真学报,2009,1：29-32.
[54]Kohonen T. Improved versions of learning vector quantization. Neural Networks,1990,1:545-550.
[55]Kohonen T. The self-organizing map. Proceedings of the IEEE.1990,78: 1464-1480.
[56]孙艳霞.纹理分析在遥感图像识别中的应用.计算机应用技术,2005,3：116-119.
[57]Chaiyaratana N, Zalzala, A.M.S. Evolving hybrid RBF-MLP networks using combined genetic/unsupervised/supervised learning control. UKACC International Conference,1998,1:330-335.
[58]Chen S, Grant PM, Cowan CN. Orthogonal least squares algorithm for training multi-output radial basis function networks. Artificial Neural Networks,1991: 336-339.
[59]Fu KS. Step towards unification of syntactic and statistical pattern recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence,1986: 398-404.
[60]彭开香,赖春山.基于信息融合的带钢厚度预测控制.北京科技大学学报,2011,(03)：1-11.
[61]Gokstorp M. Computing depth from out-of-focus blur using a local frequency representation. Proceedings of the International Conference on Pattern Recognition,1994:153-158.
[62]Marr D, Hildreth E. Theory of edge detection. Proceedings of the Royal Society B,1980,207:187-217.
[63]Smith TG, Jr M, et al.. Edge detection in images using Marr-Hildreth filtering techniques. Journal of Neuroscience Methods,1988,26:75-81.
[64]He ZQ, Siyal. Edge detection with BP neural networks. Signal Processing Proceedings,1998,2:1382-1384.
[65]金炜,俞建定.基于图像特征与面向对象BP算法的边缘检测.计算机工程,2002,4：45-48.
[66]Lo YS, Pei SC. Color image segmentation using local histogram and self-organization of Kohonen feature map. Image Processing,1999,3:232-235.
[67]Egmont PM, De RD, Handels H. Image processing with neural networks-A review. Pattern Recognition,2002:2279-2301.
[68]Elghazali M, Abdelbaki HE, Fathi AS. Texture segmentation using Kohonen self organizing feature map, Proceedings of the Nineteenth National Radio Science Conference,2002:309-317.
[69]Tarkov MS, Mun YS, et al.. Mapping adaptive fuzzy Kohonen clustering network onto distributed image processing system. Parallel Computing,2002, 28:1239-1256.
[70]Ray B, Ray M, Basuray A. A microprocessor-based on-line optical transfer function (OTF) measurement system. IEEE Transactions on Instrumentation and Measurement,1992,41:381-384.
[71]Daubechies I. Wavelet transform, time-frequency localization and signal analysis. IEEE 1988,25:42-43.
[72]Mallat S, Zhong S F. Signal characterization from multiscale edges. Proceedings of International Conference on Pattern Recognition,1990: 891-896.
[73]Daubechies I. Wavelets:an overview, with recent applications. Proceedings of the IEEE International Symposium on Information Theory,1995,2:5-6.
[74]周炯般.信息理论基础.人民邮电出版社,1989,1：56-64.
[75]Pun T. A new method for gray-level picture thresholding using the entropy of the histogram. Signal Processing,1980,2:223-237.
[76]Pun T. Entropic thresholding:A new approach. Comp. Graphics and Image Proc,1981,16:210-239.
[77]Johannsen G, Bille J. A threshold selection method using information measures. International Conference on Pattern Recognition,1982,3:140-143.
[78]Kapur JN, Sahoo PK, Wong AK. A new method for gray-level picture thresholding using the entropy of the histogram. Image Process,1985,29: 273-285.
[79]王洪彬.BP神经网络在热轧带钢卷取温度控制中的应用.重型机械,2008,6：16-18.
[80]Nikhil RP, Sankar KP. Entropic thresholding. Signal Processing,1989,16: 97-108.
[81]Abutaleb AS. Automatic thresholding of grey-level pictures using two-dimensional entropies. Pattern Recognition,1989,47:22-32.
[82]Brink AD. Thresholding of digital images using two-dimensional entropies. Pattern Recognition,1992,25:803-808.
[83]乔俊飞,韩红桂.RBF神经网络的结构动态优化设计.自动化学报,2010,6：45-49.
[84]Han B, Wang SS, Li CS. Analysis of causes of ski-up during rough rolling of Fe-1.6%Si non-oriented silicon steel. Research on Iron and Steel,2010,38: 15-17.
[85]Zhang KJ, Shao C. Computer simulation of temperature drop in plate before rolling and during rough rolling. Journal of Iron and Steel Research,2004,16: 38-41.
[86]Jiang H, Huang ZY, Yin GQ. Microstructures and mechanical properties of strehgth heavy section h-beam after rough rolling. Shanghai Metals,2009,31: 40-43.
[87]柴天佑,王笑波.RBF神经网络在加速冷却控制系统中的应用.自动化学报,2000,2：86-89.
[88]Ma T, Peng YH. Study on online temperature model in rough rolling zone of hot rolling strip and its application. Metallurgical Industry Automation,2009, 33:43-46.
[89]Shen BZ, Zhou J, Han ZQ. Numerical simulation on temperature field of hot strip in rough rolling process. Journal of Iron and Steel Research,2003,15: 10-13.
[90]王建梅,覃文忠.基于L-M算法的BP神经网络分类器.武汉大学学报(信息科学版),2005,10：107-110.
[91]Ding JG, Jiao JM, Zan P. Hot strip width prediction during rough rolling with pso-neural network based on fuzzy clustering. Journal of Northeastern University (Natural Science),2007,28:1282-1285.
[92]Ma WB, Yuan M. Optimization and perfection of the rough rolling module for ferrite sus430. Baosteel Technology,2009,06:56-60.
[93]项文强,张华,王姮,解兴哲.基于L-M算法的BP网络在变压器故障诊断中的应用.电力系统保护与控制,2011,8：62-65.
[94]Zhang QH, Benweniste A. Wavelet network. IEEE Transactions on Neural Networks,1992,3:889-898.
[95]Zhang J, Walter GG, Miao Y, et al.. Wavelet neural networks for function learning. IEEE Transactions on Signal Processing,1995,43:1485-1497.
[96]Ciftcioglu O, Durmisevic S, Sariyildiz S. Multiresolutional training of RBF networks for enhanced approximation. IFSA World Congress and 20th NAFIPS International Conference,2001,2:732-737.
[97]李鹏.基于贝叶斯理论的神经网络算法研究.光机电信息,2011,1：23-26.
[98]Lai LL, Vaseekar E, Subasinghe H, et al.. Application of wavelet transform and neural networks to fault location of a teed-circuit. Time-scale and Time-Frequency Analysis and Applications, IEEE Seminar,2000,5:11-15.
[99]Xu JH, Ho DW. Adaptive wavelet networks for nonlinear system identification, American Control Conference,1999,5:3472-3473.
[100]邱德润,朱明旱.BP算法在求解非线性问题中的应用研究.湖南文理学院学报(自然科学版),2006,2：32-35.
[101]Mac K. Bayesian interpolation. Neural Computation,1992,4:415-447.
[102]David JC. Bayesian neural networks and density networks. Nuclear Instruments and Methods in Physics Research,1994,354:73-80.
[103]魏东,张明廉,蒋志坚,孙明.基于贝叶斯方法的神经网络非线性模型辨识.计算机工程与应用,2005,11：53-56.
[104]Wang JC, Lu WX. Recurrent neural network based ecganalysis. Chinese Journal of Biomedical Engineering.1995,3:85-89.
[105]Deng QC, Wang CD, Wang HB. A New Intrusion Detection System Based on Multilayer Perceptrons Neural Network. E-Product E-Service and E-Entertainment (ICEEE), International Conference,2010,4:175-178.
[106]廖永红,李洛,黄战.基于内容的IP包过滤实现技术,计算机安全,2007,7：13-15.
[107]Chen ZQ, Yin LF, Chen CH. The functional survival of melanopsin-containing retinal ganglion cells in RCS rats during retinal degeneration. Proceedings of ⅩⅧ International Congress for Eye Research, 2008,2:201-206.
[108]Deng QC, Wang CD, Chang QG. Intrusion detection method based on pattern matching and feed-forward neural network. Journal of Tianjin University of Technology,2010,5:284-286.
[109]廖永红,李洛,黄战.基于Windows200x的WDM体系的IP过滤实现技术,计算机工程与设计,2007,7(22)：5425-5427.
[110]Hu JH, Zhou YT, Guo RB. A new model and implementation of network intrusion detection. Journal of Hunan University (Natural Sciences),2007,1: 48-57.
[111]Shi XL, Li X, Xu J. Edge detection algorithm based on forward sensing network. Computer Engineering,2005,8:53-63.
[112]李卫东,柳焯,郭玉红,宋家骅.基于电力系统运行模式及人工神经网络的潮流并行算法.电力系统自动化,1997,5：185-189.
[113]De C, Fernando J. An evolutionary immune network for data clustering. Proceedings of the IEEE SBRN,2000,11:84-89.
[114]James MN. A fast nearest-neighbor algorithm based on a principal axis search tree. IEEE Transactions on Pattern Analysis and Machine Intelligence,2001,23: 964-976.
[115]Lai WC. A VLSI neural processor for image data compression using self-organization networks. IEEE Transactions on Neural Networks,1993,3: 533-542.
[116]Qiao JF, Han HG Optimal structure design for RBFNN structure. ACTC Automatica Sinica,2010,36:866-872.
[117]廖永红,黄战.基于IP包内容的Windows包过滤技术的实现,电脑与信息技术,2007,7：4-7.
[118]Willis MJ, Montague GA, Tham MT, et al.. Artificial neural network in process engineering. Control Theory and Applications,1991,138:256-266.
[119]Xiao BX, Wang XW, Liu YF. Application and simulation research for boiler superheated system identification based on MPSO-RBF hybrid optimization strategy. Journal of System Simulation,2007,19:1382-1389.
[120]Kathryn N. Cultivating diversity. Collection Management,2003,7:42-45.