焦炭质量的DE-BP神经网络预测模型研究
|
摘要
焦炭的质量对高炉冶炼的生产有着重要的影响,为解决焦炭质量预测线性方法计算量大、检验难问题,在分析焦炭质量影响因素基础上,采用主元分析的方法确定配合煤参数的输入向量,以焦炭质量指标中灰分Ad、硫分Std、抗碎强度M40、耐磨强度M10作为输出向量,建立了焦炭质量的DE优化BP神经网络预测模型。预测效果表明该模型焦炭质量各项指标的预测值与真实值的相对误差均在4%以下,克服了BP神经网络预测精度低、易陷入局部极小值的缺点,可以满足生产工艺的要求,对焦炭生产具有一定的使用价值。
The quality of coke has a great effect on the furnace process. In order to solve the problem of large amount of calculation and inspection of coke quality prediction linear method, the coke quality prediction model based on DE-BP neural network is established on the basis of analyzing the factors affecting coke quality, which uses principal component analysis method to determine the parameters of the input vector with the coal and coke ash quality indicators Ad, sulfur Std, crushing strength M40, abrasion resistance M10 as the output vector prediction. The result of simulation shows that the relative errors of real and estimated values of the indicators are below 4%, overcoming the low prediction precision of BP neural network and the shortcoming of easily trapped in local minimum. It can meet the requirements of production process and has a certain value for coke production.
The quality of coke has a great effect on the furnace process. In order to solve the problem of large amount of calculation and inspection of coke quality prediction linear method, the coke quality prediction model based on DE-BP neural network is established on the basis of analyzing the factors affecting coke quality, which uses principal component analysis method to determine the parameters of the input vector with the coal and coke ash quality indicators Ad, sulfur Std, crushing strength M40, abrasion resistance M10 as the output vector prediction. The result of simulation shows that the relative errors of real and estimated values of the indicators are below 4%, overcoming the low prediction precision of BP neural network and the shortcoming of easily trapped in local minimum. It can meet the requirements of production process and has a certain value for coke production.
引文
[1]薛改凤,项茹,陈鹏,等.炼焦煤质量指标评价体系的研究[J].武汉科技大学学报,2009,32(1):36-40.XUE Gai-feng,XIANG Ru,CHEN Peng,et al.Index System for Evaluating the Quality of Coking Coal[J].Journal of Wuhan University of Science and Technology,2009,32(1):36-40.
[2]邓志平,张代林.优化配煤及焦炭质量的研究探讨[J].江西冶金,2014,34(4):1-3.DENG Zhi-ping,ZHANG Dai-lin.Study on Optimization of Coal Blending and Coke Quality[J].Jiangxi Metallurgy,2014,34(4):1-3.
[3]于青,王德全,吴子良,等.焦炭反应性和反应后强度的关系及其影响因素研究[J].中国冶金,2012,22(3):10-14.YU Qing,WANG De-quan,WU Zi-liang,et al.Relationship Between Coke Reactivity and Strength Affer Reaction and Research of Influence Factor[J].China Metallurgy,2012,22(3):10-14.
[4]曹密静,王胜春,贾万福.焦炭灰分与硫分的预测[J].燃料与化工,2012,43(5):14-16.CAO Mi-jing,WANG Sheng-chun,JIA Wan-fu.Prediction for Ash and Sulfur Content of Coke[J].Fuel&Chemical Processes,2012,43(5):14-16.
[5]刘颖义,穆红旺,张淑会.焦炭质量预测模型的建立[J].中国冶金,2012,22(1):28-31.LIU Ying-yi,MU Hong-wang,ZHANG Shu-hui.Establish of Prediction Mode of Coke Quality[J].China Metallurgy,2012,22(1):28-31.
[6]谢海深,刘永新,孟军波,等.焦炭质量预测模型的研究[J].煤炭转化,2006,29(3):54-57.XIE Hai-shen,LIU Yong-xin,MENG Jun-bo,et al.Study on The Predictive Model for Coke Quality[J].Coal Conversion,2006,29(3):54-57.
[7]刘俊,张学东,刘宏,等.基于BP神经网络的焦炭质量预测[J].燃料与化工,2006,37(6):12-15.LIU Jun,ZHANG Xue-dong,LIU Hong,et al.Prediction of Coke Quality Based on BP Neural Network[J].Fuel&Chemical Processes,2006,37(6):12-15.
[8]雷琪,刘君贤,何勇,等.基于PCA与RBF的焦炭质量预测模型[J].控制工程,2010,17(4):513-520.LEI Qi,LIU Jun-xian,HE Yong,et al.A Prediction Mode for Coke Quality Based on PCA and RBF[J].Control Engineering of China,2010,17(4):513-520.
[9]覃光华,李祚泳.BP网络过拟合问题研究及其应用[J].武汉科技大学学报,2006,39(6):55-58.QIN Guang-hua,LI Zuo-yong.Over-fitting of BP NN Research and Its Application[J].Engineering Journal of Wuhan University,2006,39(6):55-58.
[10]田英奇,张卫华,沈寓韬,等.镜质组反射率指导优化配煤炼焦方案的研究[J].煤炭科学技术,2016,44(4):162-168.TIAN Ying-qi,ZHANG Wei-hua,SHEN Yu-tao,et al.Research on optimization of coal blending coking guided by vitrinite reflectance[J].Coal Science and Technology,2016,44(4):162-168.
[11]周洪,闵礼书,邹祥林.基于神经网络的特大型焦炉焦炭质量预测研究[J].系统仿真学报,2009,21(6):1543-1552.ZHOU Hong,MIN Li-shu,ZOU Xiang-lin.Research on Prediction of Blended Coal for Coking Quality in Oversized Coke Furnace Based on Neural Network[J].Journal of System Simulation,2009,21(6):1543-1552.
[12]贺毅朝,王熙照,刘坤起.差分算法演义的收敛性分析与算法改进[J].软件学报,2010,21(5):875-885.HE Yi-chao,WANG Xi-zhao,LIU Kun-qi.Convergent Analysis and Algorithmic Improvement of Differential Evolution[J].Journal of Software,2010,21(5):875-885.
[13]汪慎文,丁立新,张文生,等.差分进化算法研究进展[J].武汉大学学报(理学版),2014,60(4):283-292.WANG Shen-wen,DING Li-xin,ZHANG Wen-sheng,et al.Survey of Differential Evolution[J].Journal of Wuhan University(Natural Science Edition),2014,60(4):283-291.
[14]Das S,Suganthan P N.Differentia Evolution:A Survey of the State-of-the-Art[J].IEEE Trans.Evolut.Comput.(S1089-778X),2011,15(1):4-31.
[15]Li Gaoyang,Liu Mingguang.The Summary of Differential Evolution Algorithm and its Improvements[C]//2010 3rd International Conference on Advanced Computer Theory and Engineering(ICACTE),Chengdu,China.2010:153-156.
[2]邓志平,张代林.优化配煤及焦炭质量的研究探讨[J].江西冶金,2014,34(4):1-3.DENG Zhi-ping,ZHANG Dai-lin.Study on Optimization of Coal Blending and Coke Quality[J].Jiangxi Metallurgy,2014,34(4):1-3.
[3]于青,王德全,吴子良,等.焦炭反应性和反应后强度的关系及其影响因素研究[J].中国冶金,2012,22(3):10-14.YU Qing,WANG De-quan,WU Zi-liang,et al.Relationship Between Coke Reactivity and Strength Affer Reaction and Research of Influence Factor[J].China Metallurgy,2012,22(3):10-14.
[4]曹密静,王胜春,贾万福.焦炭灰分与硫分的预测[J].燃料与化工,2012,43(5):14-16.CAO Mi-jing,WANG Sheng-chun,JIA Wan-fu.Prediction for Ash and Sulfur Content of Coke[J].Fuel&Chemical Processes,2012,43(5):14-16.
[5]刘颖义,穆红旺,张淑会.焦炭质量预测模型的建立[J].中国冶金,2012,22(1):28-31.LIU Ying-yi,MU Hong-wang,ZHANG Shu-hui.Establish of Prediction Mode of Coke Quality[J].China Metallurgy,2012,22(1):28-31.
[6]谢海深,刘永新,孟军波,等.焦炭质量预测模型的研究[J].煤炭转化,2006,29(3):54-57.XIE Hai-shen,LIU Yong-xin,MENG Jun-bo,et al.Study on The Predictive Model for Coke Quality[J].Coal Conversion,2006,29(3):54-57.
[7]刘俊,张学东,刘宏,等.基于BP神经网络的焦炭质量预测[J].燃料与化工,2006,37(6):12-15.LIU Jun,ZHANG Xue-dong,LIU Hong,et al.Prediction of Coke Quality Based on BP Neural Network[J].Fuel&Chemical Processes,2006,37(6):12-15.
[8]雷琪,刘君贤,何勇,等.基于PCA与RBF的焦炭质量预测模型[J].控制工程,2010,17(4):513-520.LEI Qi,LIU Jun-xian,HE Yong,et al.A Prediction Mode for Coke Quality Based on PCA and RBF[J].Control Engineering of China,2010,17(4):513-520.
[9]覃光华,李祚泳.BP网络过拟合问题研究及其应用[J].武汉科技大学学报,2006,39(6):55-58.QIN Guang-hua,LI Zuo-yong.Over-fitting of BP NN Research and Its Application[J].Engineering Journal of Wuhan University,2006,39(6):55-58.
[10]田英奇,张卫华,沈寓韬,等.镜质组反射率指导优化配煤炼焦方案的研究[J].煤炭科学技术,2016,44(4):162-168.TIAN Ying-qi,ZHANG Wei-hua,SHEN Yu-tao,et al.Research on optimization of coal blending coking guided by vitrinite reflectance[J].Coal Science and Technology,2016,44(4):162-168.
[11]周洪,闵礼书,邹祥林.基于神经网络的特大型焦炉焦炭质量预测研究[J].系统仿真学报,2009,21(6):1543-1552.ZHOU Hong,MIN Li-shu,ZOU Xiang-lin.Research on Prediction of Blended Coal for Coking Quality in Oversized Coke Furnace Based on Neural Network[J].Journal of System Simulation,2009,21(6):1543-1552.
[12]贺毅朝,王熙照,刘坤起.差分算法演义的收敛性分析与算法改进[J].软件学报,2010,21(5):875-885.HE Yi-chao,WANG Xi-zhao,LIU Kun-qi.Convergent Analysis and Algorithmic Improvement of Differential Evolution[J].Journal of Software,2010,21(5):875-885.
[13]汪慎文,丁立新,张文生,等.差分进化算法研究进展[J].武汉大学学报(理学版),2014,60(4):283-292.WANG Shen-wen,DING Li-xin,ZHANG Wen-sheng,et al.Survey of Differential Evolution[J].Journal of Wuhan University(Natural Science Edition),2014,60(4):283-291.
[14]Das S,Suganthan P N.Differentia Evolution:A Survey of the State-of-the-Art[J].IEEE Trans.Evolut.Comput.(S1089-778X),2011,15(1):4-31.
[15]Li Gaoyang,Liu Mingguang.The Summary of Differential Evolution Algorithm and its Improvements[C]//2010 3rd International Conference on Advanced Computer Theory and Engineering(ICACTE),Chengdu,China.2010:153-156.