LIBSVM回归算法在焦炭强度预测中的应用
|
摘要
随着高炉大型化和富氧喷吹技术的发展,焦炭在高炉中的骨架作用变得日益重要,建立一种适用性好、准确度高的焦炭强度预测方法对钢铁企业具有重要意义。支持向量机是一种基于统计学理论发展而来的机器学习方法,它在解决小样本、非线性和高维识别中表现出许多特有的优势。采用基于遗传算法参数寻优的支持向量机方法预测焦炭的冷态强度和热性质,通过对比2因素和5因素的预测结果的偏差大小,得知在用支持向量机方法进行建模时,若在配合煤的挥发分Vdaf和黏结指数G的基础上加入配合煤灰分Ad、细度以及焦炉平均温度这3个指标,则能得到更接近实际值的预测结果。
With the development of large-scale blast furnaces and oxygen-enriched injection technologies,the role of coke in the blast furnace has been increasingly important. Establishing a coke quality prediction method with good applicability and high accuracy may be important for iron and steel enterprises. Support vector machine(SVM)is a machine learning method developed from statistics theories. It has many unique advantages in solving small sample problems,nonlinear and high-dimensional recognition problems. The prediction of the coke cold strength and thermal properties was based on parameters optimization of SVM method by the genetic algorithm. By comparing the deviations of the prediction results of the two factors and the five factors,it was known that a closer result can before casted when the SVM method building a model was used,considering the effect of the coal ash,the fineness and the average temperature of coke oven based on the effect of the volatile matter and the bond index of coal blend.
With the development of large-scale blast furnaces and oxygen-enriched injection technologies,the role of coke in the blast furnace has been increasingly important. Establishing a coke quality prediction method with good applicability and high accuracy may be important for iron and steel enterprises. Support vector machine(SVM)is a machine learning method developed from statistics theories. It has many unique advantages in solving small sample problems,nonlinear and high-dimensional recognition problems. The prediction of the coke cold strength and thermal properties was based on parameters optimization of SVM method by the genetic algorithm. By comparing the deviations of the prediction results of the two factors and the five factors,it was known that a closer result can before casted when the SVM method building a model was used,considering the effect of the coal ash,the fineness and the average temperature of coke oven based on the effect of the volatile matter and the bond index of coal blend.
引文
[1]张静,杨婷.焦炭市场现状分析及预测[J].煤炭加工与综合利用,2017(12):21.(CHEN Jing,YANG Ting.Analysis and prediction of the current situation of coke market[J].Coal Processing and Comprehensive Utilization,2017(12):21.)
[2]Diez M A,A lvarez R,Barrriocanal C.Coal for metallurgical coke production predicitions of coke quality and future requiremengts for coking making[J].International of Coal Geology,2002,50:389.
[3]邓乃扬,田英杰.数据挖掘中的新方法--支持向量机[M].北京:科学出版社,2004.(DENG Nai-yang,TIAN Ying-jie.New Methods in Data Mining-Support Vector Machine[M].Beijing:Science Press,2004.)
[4]Vapnik V N.The Nature of Statistical Learning Theory[M].New York:Springer Verlag,1995.
[5]李栋,王洪礼.城市生活用水量的支持向量机回归预测[J].天津大学学报,2006,8(1):64.(LI Dong,WANG Hong-li.Support vector machine regression prediction of urban domestic water consumption[J].Journal of Tianjin University,2006,8(1):64.)
[6]孔繁任,徐瑞华,姚胜勇.交通量的支持向量回归机预测及参数选择研究[J].计算机工程,2007,33(5):20.(KONG Fanren,XU Rui-hua,YAO Sheng-yong.Prediction and parameter selection of traffic volume based on support vector regression[J].Computer Engineering,2007,33(5):20.)
[7]田英杰.支持向量回归机及其应用研究[D].北京:中国农业大学,2005.(TIAN Ying-jie.Support Vector Regression Machine and its Application[D].Beijing:China Agricultural University,2005.)
[8]苏建龙,米鸿,王彦春,等.基于支持向量机的非线性弹性阻抗反演方法[J].石油地球物理勘测,2014,49(4):751.(SU Jianlong,MI Hong,WANG Yan-chun,et al.Nonlinear elastic impedance inversion method based on support vector machine[J].Petroleum Geophysical Survey,2014,49(4):751.)
[9]卢涛,陈德钊.径向基网络的研究进展与评述[J].计算机工程与应用,2005,41(4):60.(LU Tao,CHEN De-zhao.Progress and review of RBF network[J].Computer Engineering and Application,2005,41(4):60.)
[10]徐义田.支持向量回归算法的研究及其在食物安全中的应用[D].北京:中国农业大学,2004.(XU Yi-tian.Research on Support Vector Regression Algorithm and Its Application in Food Safety[D].Beijing:China Agricultural University,2004.)
[11]Cristianini N,Taylor J S.An Introduction to Support Vector Machines and Other Kernel-based Learning Methonds[M].England:Cambridge University Press,2000.
[12]Chih-Chung Chang,Chih-Jen Lin.LIBSVM:A library for support vector machines[J].ACM Transactions on Intelligent System and Technology,2011,27:1.
[13]孙涵,杨普容,成金华.基于Matlab支持向量回归机的能源需求预测模型[J].系统工程理论与实践,2011,31(10):2003.(SUN Han,YANG Pu-rong,CHENG Jin-hua.Energy demand forecasting model based on Matlab support vector regression[J].System Engineering Theory and Practice,2011,31(10):2003.)
[14]姚兆明,李郑飞,陈军浩.人工动土温度场的支持向量机法预测研究[J].煤炭工程,2012(9):89.(YAO Zhao-ming,LIZheng-fei,CHEN Jun-hao.Prediction of artificial soil temperature field based on support vector machine[J].Coal Engineering,2012(9):89.)
[15]王慧勤,雷刚.基于LIBSVM的风速预测方法研究[J].科学技术与工程,2011,11(22):5441.(WANG Hui-qin,LEI Gang.Research on wind speed prediction method based on LIBSVM[J].Science Technology and Engineering,2011,11(22):5441.)
[16]蒋伟,王运祥,张巍,等.基于LIBSVM的哈密瓜成熟度无损检测技术[J].江苏农业科学,2016,44(11):397.(JIANG Wei,WANG Yun-xiang,ZHANG Wei,et al.Nondestructive testing technology of Hami melon maturity based on LIBSVM[J].Jiangsu Agricultural Sciences,2016,44(11):397.)
[17]崔庆安,何桢,崔富新.基于支持向量机的焦炭质量预测模型[J].化工自动化及仪表,2006,33(1):28.(CUI Qing-an,HE Zhen,CUI Fu-xin.Prediction model of coke quality based on support vector machine[J].Chemical Automation and Instrument,2006,33(1):28.)
[18]谢海深,刘永新,吕庆,等.焦炭质量预测模型[J].东北大学学报:自然科学版,2007(3):373.(XIE Hai-shen,LIU Yong-xin,LU?Qing,et al.Prediction model of coke quality[J].Journal of Northeastern University:Natural Science,2007(3):373.)
[19]赵树果,单晓云,陈少敏,等.人工神经网络在炼焦配煤质量预测模型中的应用[J].中国煤炭,2006(7):51.(ZHAO Shuguo,SHAN Xiao-yun,CHEN Shao-min,et al.Application of artificial neural network in quality prediction model of coking coal blending[J].Chinese Coal,2006(7):51.)
[20]单晓云,高志方,赵树果,等.模糊聚类分析优化炼焦配煤的研究[J].煤炭科学技术,2005(6):68.(SHAN Xiao-yun,GAO Zhi-fang,ZHAO Shu-guo,et al.Study on optimization of coking coal blending by vague cluster analysis[J].Coal Science and Technology,2005(6):68.)
[21]张启峰,曹银平,刘全,等.焦炭灰分控制技术应用研究[C]//第十届中国钢铁年会暨第六届宝钢学术年会论文集Ⅲ.北京:冶金工业出版社,2015:1.(ZHANG Qi-feng,CAO Yinping,LIU Quan,et al.Research on the application of coke ash control technology[C]//The Tenth Annual China Annual Iron and Steel Meeting and the Symposium of the Sixth Annual on Baosteel Academic Conference III.Beijing:Metallurgical Industry Press,2015:1.)
[22]周勇,王光辉,田永胜,等.K和Na对炼焦煤工艺性质及捣固焦质量的影响[J].煤炭转化,2016,39(3):72.(ZHOU Yong,WANG Guang-hui,TIAN Yong-sheng,et al.Effects of K and Na on the technological properties of coking coal and the quality of tamping coke[J].Coal Conversion,2016,39(3):72.)
[23]Nomura S,Naito M,Yamaguchi K.Post-reaction strength of catalyst-added highly reactive coke[J].ISIJ International,2007,47(6):831.
[24]孔德文,陈永馨,闫宝忠,等.灰分对焦炭溶损反应起始温度的影响[J].燃料与化工,2015,46(3):11.(KONG De-wen,CHEN Yong-xin,YAN Bao-zhong,et al.Influence of ash on the initial solution reaction temperature of coke[J].Fuel and Chemical Processes,2015,46(3):11)
[25]郑朋超,张建良,刘征建,等.碱金属对焦炭热性能的影响[J].中国冶金,2017,27(5):19.(ZHENG Peng-chao,ZHANGJian-liang,LIU Zheng-jian,et al.Effect of alkali metals on thermal properties of coke[J].China Metallurgy,2017,27(5):19.)
[26]魏江红,周敏,韦小梅,等.应用线性回归法预测焦炭强度[J].中国煤炭,2011,37(6):90.(WEI Jiang-hong,ZHOU Min,WEI Xiao-mei,et al.Prediction of coke strength by linear regression[J].Chinese Coal,2011,37(6):90.)
[27]张进春,吴超.基于多重多元回归的焦炭质量预测模型[J].科技导报,2010,28(7):79.(ZHANG Jin-chun,WU Chao.Coke quality prediction model based on multiple regression analysis[J].Science and Technology Review,2010,28(7):79.)
[28]单晓云,赵树果,刘永新.基于神经网络的焦炭质量预测模型[J].选煤技术,2005,2:1.(SHAN Xiao-yun,ZHAO Shu-guo,LIU Yong-xin.Prediction model of coke quality based on neural network[J].Coal Preparation Technology,2005,2:1.)
[29]单晓云,陈黎明,吴政,等.共轭梯度法优化的BP神经网络焦炭质量预测模型[J].选煤技术,2010(4):22.(SHAN Xiao-yun,CHEN Li-ming,WU Zheng,et al.Prediction model of coke quality based on BP neural network optimized by conjugate gradient method[J].Coal Preparation Technology,2010(4):22.)
[30]张伟峰,孔金生.GA优化的BP神经网络焦炭质量预测模型[J].选煤技术,2011(3):56.(ZHANG Wei-feng,KONG Jinsheng.Coke quality prediction model of BP neural network optimized by GA[J].Coal Preparation Technology,2011(3):56.)
[31]陈晓舟.配合煤细度对JN60-6型焦炉生产的影响[J].燃料与化工,2014,45(3):21.(CHEN Xiao-zhou.Influence of coals fineness on the production of JN60-6 coke oven[J].Fuel and Chemical Industry,2014,45(3):21.)
[32]张艳丽.不同细度煤对焦炭质量的影响[J].四川冶金,2007,29(2):33.(ZHANG Yan-li.Influence of different fineness coal on coke quality[J].Sichuan Metallurgy,2007,29(2):33.)
[33]张代林,林慧薪,王晓婷,等.炼焦煤灰分对其结焦性的影响规律[J].钢铁,2017,52(8):10.(ZHANG Dai-lin,LIN Huixin,WANG Xiao-ting,et al.Influence of ash of coking coals on its coking property[J].Iron and Steel,2017,52(8):10.)
[34]吕庆,王岩,谢海深,等.灰成分及光学组织对焦炭热性能的影响[J].中国冶金,2016,26(8):8.(LüQing,WANG Yan,XIE Hai-shen,et al.Effect of ash composition and optical texture on coke thermal property[J].China Metallurgy,2016,26(8):8.)
[35]刘德军.冶金焦的本质特性与现代高炉冶炼的对应关系[J].钢铁,2016,51(10):78.(LIU De-jun.Relationship between essential characteristic of metallurgical coke and modern blast furnace smelting[J].Iron and Steel,2016,51(10):78.)
[36]王丰岩,张永强,王春霞,等.盛隆化工焦炭质量预测[J].山东冶金,2014,36(1):36.(WANG Feng-yan,ZHANG Yongqiang,WANG Chun-xia,et al.Quality prediction of coke in Shenglong chemical industry[J].Shandong Metallurgy,2014,36(1):36.)
[37]梁建华.武钢焦炭质量预测模型的研究[D].武汉:武汉科技大学,2011.(LIANG Jian-hua.Study on the Prediction Model of the Coke Quality in WISCO[D].Wuhan:Wuhan University of Science and Technology,2011.)
[38]张群,吴信慈,冯安祖,等.宝钢焦炭质量预测模型Ⅱ.焦炭质量预测模型的建立和应用[J].燃料化学学报,2002,30(4):302.(ZHANG Qun,WU Xin-ci,FENG An-zu,et al.Coke quality prediction model of BaosteelⅡ.The establishment and application of coke quality prediction model[J].Journal of Fuel Chemistry and Technology,2002,30(4):302.)
[39]张代林,赵梅梅,王培珍,等.影响焦炭热性质因素的研究[J].钢铁,2009,44(10):10.(ZHANG Dai-lin,ZHAO Mei-mei,WANG Pei-zhen,et al.Study on factors of influencing on coke thermal property[J].Iron and Steel,2009,44(10):10.)
[2]Diez M A,A lvarez R,Barrriocanal C.Coal for metallurgical coke production predicitions of coke quality and future requiremengts for coking making[J].International of Coal Geology,2002,50:389.
[3]邓乃扬,田英杰.数据挖掘中的新方法--支持向量机[M].北京:科学出版社,2004.(DENG Nai-yang,TIAN Ying-jie.New Methods in Data Mining-Support Vector Machine[M].Beijing:Science Press,2004.)
[4]Vapnik V N.The Nature of Statistical Learning Theory[M].New York:Springer Verlag,1995.
[5]李栋,王洪礼.城市生活用水量的支持向量机回归预测[J].天津大学学报,2006,8(1):64.(LI Dong,WANG Hong-li.Support vector machine regression prediction of urban domestic water consumption[J].Journal of Tianjin University,2006,8(1):64.)
[6]孔繁任,徐瑞华,姚胜勇.交通量的支持向量回归机预测及参数选择研究[J].计算机工程,2007,33(5):20.(KONG Fanren,XU Rui-hua,YAO Sheng-yong.Prediction and parameter selection of traffic volume based on support vector regression[J].Computer Engineering,2007,33(5):20.)
[7]田英杰.支持向量回归机及其应用研究[D].北京:中国农业大学,2005.(TIAN Ying-jie.Support Vector Regression Machine and its Application[D].Beijing:China Agricultural University,2005.)
[8]苏建龙,米鸿,王彦春,等.基于支持向量机的非线性弹性阻抗反演方法[J].石油地球物理勘测,2014,49(4):751.(SU Jianlong,MI Hong,WANG Yan-chun,et al.Nonlinear elastic impedance inversion method based on support vector machine[J].Petroleum Geophysical Survey,2014,49(4):751.)
[9]卢涛,陈德钊.径向基网络的研究进展与评述[J].计算机工程与应用,2005,41(4):60.(LU Tao,CHEN De-zhao.Progress and review of RBF network[J].Computer Engineering and Application,2005,41(4):60.)
[10]徐义田.支持向量回归算法的研究及其在食物安全中的应用[D].北京:中国农业大学,2004.(XU Yi-tian.Research on Support Vector Regression Algorithm and Its Application in Food Safety[D].Beijing:China Agricultural University,2004.)
[11]Cristianini N,Taylor J S.An Introduction to Support Vector Machines and Other Kernel-based Learning Methonds[M].England:Cambridge University Press,2000.
[12]Chih-Chung Chang,Chih-Jen Lin.LIBSVM:A library for support vector machines[J].ACM Transactions on Intelligent System and Technology,2011,27:1.
[13]孙涵,杨普容,成金华.基于Matlab支持向量回归机的能源需求预测模型[J].系统工程理论与实践,2011,31(10):2003.(SUN Han,YANG Pu-rong,CHENG Jin-hua.Energy demand forecasting model based on Matlab support vector regression[J].System Engineering Theory and Practice,2011,31(10):2003.)
[14]姚兆明,李郑飞,陈军浩.人工动土温度场的支持向量机法预测研究[J].煤炭工程,2012(9):89.(YAO Zhao-ming,LIZheng-fei,CHEN Jun-hao.Prediction of artificial soil temperature field based on support vector machine[J].Coal Engineering,2012(9):89.)
[15]王慧勤,雷刚.基于LIBSVM的风速预测方法研究[J].科学技术与工程,2011,11(22):5441.(WANG Hui-qin,LEI Gang.Research on wind speed prediction method based on LIBSVM[J].Science Technology and Engineering,2011,11(22):5441.)
[16]蒋伟,王运祥,张巍,等.基于LIBSVM的哈密瓜成熟度无损检测技术[J].江苏农业科学,2016,44(11):397.(JIANG Wei,WANG Yun-xiang,ZHANG Wei,et al.Nondestructive testing technology of Hami melon maturity based on LIBSVM[J].Jiangsu Agricultural Sciences,2016,44(11):397.)
[17]崔庆安,何桢,崔富新.基于支持向量机的焦炭质量预测模型[J].化工自动化及仪表,2006,33(1):28.(CUI Qing-an,HE Zhen,CUI Fu-xin.Prediction model of coke quality based on support vector machine[J].Chemical Automation and Instrument,2006,33(1):28.)
[18]谢海深,刘永新,吕庆,等.焦炭质量预测模型[J].东北大学学报:自然科学版,2007(3):373.(XIE Hai-shen,LIU Yong-xin,LU?Qing,et al.Prediction model of coke quality[J].Journal of Northeastern University:Natural Science,2007(3):373.)
[19]赵树果,单晓云,陈少敏,等.人工神经网络在炼焦配煤质量预测模型中的应用[J].中国煤炭,2006(7):51.(ZHAO Shuguo,SHAN Xiao-yun,CHEN Shao-min,et al.Application of artificial neural network in quality prediction model of coking coal blending[J].Chinese Coal,2006(7):51.)
[20]单晓云,高志方,赵树果,等.模糊聚类分析优化炼焦配煤的研究[J].煤炭科学技术,2005(6):68.(SHAN Xiao-yun,GAO Zhi-fang,ZHAO Shu-guo,et al.Study on optimization of coking coal blending by vague cluster analysis[J].Coal Science and Technology,2005(6):68.)
[21]张启峰,曹银平,刘全,等.焦炭灰分控制技术应用研究[C]//第十届中国钢铁年会暨第六届宝钢学术年会论文集Ⅲ.北京:冶金工业出版社,2015:1.(ZHANG Qi-feng,CAO Yinping,LIU Quan,et al.Research on the application of coke ash control technology[C]//The Tenth Annual China Annual Iron and Steel Meeting and the Symposium of the Sixth Annual on Baosteel Academic Conference III.Beijing:Metallurgical Industry Press,2015:1.)
[22]周勇,王光辉,田永胜,等.K和Na对炼焦煤工艺性质及捣固焦质量的影响[J].煤炭转化,2016,39(3):72.(ZHOU Yong,WANG Guang-hui,TIAN Yong-sheng,et al.Effects of K and Na on the technological properties of coking coal and the quality of tamping coke[J].Coal Conversion,2016,39(3):72.)
[23]Nomura S,Naito M,Yamaguchi K.Post-reaction strength of catalyst-added highly reactive coke[J].ISIJ International,2007,47(6):831.
[24]孔德文,陈永馨,闫宝忠,等.灰分对焦炭溶损反应起始温度的影响[J].燃料与化工,2015,46(3):11.(KONG De-wen,CHEN Yong-xin,YAN Bao-zhong,et al.Influence of ash on the initial solution reaction temperature of coke[J].Fuel and Chemical Processes,2015,46(3):11)
[25]郑朋超,张建良,刘征建,等.碱金属对焦炭热性能的影响[J].中国冶金,2017,27(5):19.(ZHENG Peng-chao,ZHANGJian-liang,LIU Zheng-jian,et al.Effect of alkali metals on thermal properties of coke[J].China Metallurgy,2017,27(5):19.)
[26]魏江红,周敏,韦小梅,等.应用线性回归法预测焦炭强度[J].中国煤炭,2011,37(6):90.(WEI Jiang-hong,ZHOU Min,WEI Xiao-mei,et al.Prediction of coke strength by linear regression[J].Chinese Coal,2011,37(6):90.)
[27]张进春,吴超.基于多重多元回归的焦炭质量预测模型[J].科技导报,2010,28(7):79.(ZHANG Jin-chun,WU Chao.Coke quality prediction model based on multiple regression analysis[J].Science and Technology Review,2010,28(7):79.)
[28]单晓云,赵树果,刘永新.基于神经网络的焦炭质量预测模型[J].选煤技术,2005,2:1.(SHAN Xiao-yun,ZHAO Shu-guo,LIU Yong-xin.Prediction model of coke quality based on neural network[J].Coal Preparation Technology,2005,2:1.)
[29]单晓云,陈黎明,吴政,等.共轭梯度法优化的BP神经网络焦炭质量预测模型[J].选煤技术,2010(4):22.(SHAN Xiao-yun,CHEN Li-ming,WU Zheng,et al.Prediction model of coke quality based on BP neural network optimized by conjugate gradient method[J].Coal Preparation Technology,2010(4):22.)
[30]张伟峰,孔金生.GA优化的BP神经网络焦炭质量预测模型[J].选煤技术,2011(3):56.(ZHANG Wei-feng,KONG Jinsheng.Coke quality prediction model of BP neural network optimized by GA[J].Coal Preparation Technology,2011(3):56.)
[31]陈晓舟.配合煤细度对JN60-6型焦炉生产的影响[J].燃料与化工,2014,45(3):21.(CHEN Xiao-zhou.Influence of coals fineness on the production of JN60-6 coke oven[J].Fuel and Chemical Industry,2014,45(3):21.)
[32]张艳丽.不同细度煤对焦炭质量的影响[J].四川冶金,2007,29(2):33.(ZHANG Yan-li.Influence of different fineness coal on coke quality[J].Sichuan Metallurgy,2007,29(2):33.)
[33]张代林,林慧薪,王晓婷,等.炼焦煤灰分对其结焦性的影响规律[J].钢铁,2017,52(8):10.(ZHANG Dai-lin,LIN Huixin,WANG Xiao-ting,et al.Influence of ash of coking coals on its coking property[J].Iron and Steel,2017,52(8):10.)
[34]吕庆,王岩,谢海深,等.灰成分及光学组织对焦炭热性能的影响[J].中国冶金,2016,26(8):8.(LüQing,WANG Yan,XIE Hai-shen,et al.Effect of ash composition and optical texture on coke thermal property[J].China Metallurgy,2016,26(8):8.)
[35]刘德军.冶金焦的本质特性与现代高炉冶炼的对应关系[J].钢铁,2016,51(10):78.(LIU De-jun.Relationship between essential characteristic of metallurgical coke and modern blast furnace smelting[J].Iron and Steel,2016,51(10):78.)
[36]王丰岩,张永强,王春霞,等.盛隆化工焦炭质量预测[J].山东冶金,2014,36(1):36.(WANG Feng-yan,ZHANG Yongqiang,WANG Chun-xia,et al.Quality prediction of coke in Shenglong chemical industry[J].Shandong Metallurgy,2014,36(1):36.)
[37]梁建华.武钢焦炭质量预测模型的研究[D].武汉:武汉科技大学,2011.(LIANG Jian-hua.Study on the Prediction Model of the Coke Quality in WISCO[D].Wuhan:Wuhan University of Science and Technology,2011.)
[38]张群,吴信慈,冯安祖,等.宝钢焦炭质量预测模型Ⅱ.焦炭质量预测模型的建立和应用[J].燃料化学学报,2002,30(4):302.(ZHANG Qun,WU Xin-ci,FENG An-zu,et al.Coke quality prediction model of BaosteelⅡ.The establishment and application of coke quality prediction model[J].Journal of Fuel Chemistry and Technology,2002,30(4):302.)
[39]张代林,赵梅梅,王培珍,等.影响焦炭热性质因素的研究[J].钢铁,2009,44(10):10.(ZHANG Dai-lin,ZHAO Mei-mei,WANG Pei-zhen,et al.Study on factors of influencing on coke thermal property[J].Iron and Steel,2009,44(10):10.)