基于能量平衡的矿热炉能量输入方法的研究
|
摘要
铁合金厂是耗能大户,其生产铁合金的主要设备是矿热炉。矿热炉冶炼过程中所需能量主要来源于电能,将电能转化为热能熔化炉料。不同的炉况对电能的要求是不同的,供电制度能否满足矿热炉的用电要求,直接影响着冶炼速度、产品质量、能量消耗以及能量损耗等。因此针对不同的炉况确定合理的供电制度,提供合理的二次电压、电流值,对于铁合金厂合理控制冶炼时间,提高生产率和合金品质,实现节能降耗的目的具有重要意义。
本文在查阅大量相关资料的基础上,概述了矿热炉的发展现状、发展趋势、神经网络在冶金工业中的应用以及矿热炉的机械设备、冶炼工艺和一些异常情况的处理。以中钢集团吉林铁合金股份有限公司八分厂801号炉为研究背景,对矿热炉的冶炼原理以及工艺进行分析,基于供应能量(电能和化学反应释放热能)与需求能量(消耗和损耗能量)的能量平衡建立了电能输入模型,确定了所需的总电量；基于电极位置判断矿热炉生产铁合金的三个冶炼阶段(引弧加料期、熔化期和精炼期),但是由于进入精炼期后电极位置基本不发生变化,无法准确判断精炼期的结束,所以在这里利用神经网络预测模型对精炼期的铁水温度和碳含量进行预测,当铁水温度和碳含量都满足工艺要求时,即可判断精炼期结束,即终点时刻,仿真结果验证了模型的准确性和有效性。根据各阶段不同的电能需求,分别给出电能输入方法,从理论上提出了一种新的电能输入方法：在满足单位时间供电损耗金额最小的约束条件时,确定矿热炉各个阶段的电压值,再结合各个阶段的用电量需求,得出电流值。最后利用JSP技术搭建了所提方法的能量输入优化平台。
通过仿真分析,验证了所提能量输入方法能够缩短矿热炉的冶炼时间,减少消耗。为矿热炉的生产过程实现节能降耗提供了理论依据。
Ferroalloy Plant is the energy-hungry factory, it's main equipment used for producing ferroalloy is submerged arc furnace. The energy smelting process needed mainly comes from electrical energy, electrical energy is converted into heat energy to melt furnace burden. The different furnace condition needs of different electrical energy, electricity requirements in different furnace condition can be met by power supply system, which directly affect the smelting speed, product quality, energy consumption and energy losses, etc. Therefore, determination of reasonable power supply system in view of the different furnace condition, providing reasonable secondary voltage and current value, for ferroalloy plant reasonable control smelting time, increase productivity and alloy quality, to achieve the purpose of saving energy and reducing consumption has important significance.
The paper on the basis of consulting a large number of relevant data and literatures, summarizes development present situation and development trend of the submerged arc furnace, applications of neural network in metallurgical industry as well as mechanical equipment, smelting process of submerged arc furnace and some treatment on the abnormal situation. With No.801 furnace of Sinosteel Jilin Ferroalloy Co., Ltd eighth branch as research background, the analysis of the smelting principle and process of the submerged arc furnace, based on energy balance on supply energy (electric energy and chemical heat release) and demand energy (consumption and wastage energy) establish the power input model, and determine the required total electric quantity; Based on the electrode position have judgment on three smelting stage of submerged arc furnace producing ferroalloy(including arc ignition、charging stage, melting stage, refining stage). But, because of the electrode position scarcely changed in the refining stage, can't have accurate judgement of the end of the refining period, so the paper forecasting molten iron temperature and carbon content of refining period by using neural network predictive model, when molten iron temperature and carbon content meet the technological requirements, refining period is over, namely the end time. The simulation results verify the validity and the accuracy of the models. Then put forward various power input method in accordance with the electricity demand of each stage. Theoretically propose a new power input method:In meeting the constraint condition that there is the smallest loss amount in power supply in unit of time, determine voltage values at all stages of submerged arc furnace, which are combined with electricity demand of all phases, it is concluded that the current value. Lastly, the energy input optimization platform is established to realize the proposed method by JSP technology.
Through the simulation analysis, validation proposed energy input method can shorten the time of smelting -arc furnace, reduce the consumption. And provide the theoretical basis for realizing that saving energy and reducing consumption in the production process of submerged arc furnace.
本文在查阅大量相关资料的基础上,概述了矿热炉的发展现状、发展趋势、神经网络在冶金工业中的应用以及矿热炉的机械设备、冶炼工艺和一些异常情况的处理。以中钢集团吉林铁合金股份有限公司八分厂801号炉为研究背景,对矿热炉的冶炼原理以及工艺进行分析,基于供应能量(电能和化学反应释放热能)与需求能量(消耗和损耗能量)的能量平衡建立了电能输入模型,确定了所需的总电量；基于电极位置判断矿热炉生产铁合金的三个冶炼阶段(引弧加料期、熔化期和精炼期),但是由于进入精炼期后电极位置基本不发生变化,无法准确判断精炼期的结束,所以在这里利用神经网络预测模型对精炼期的铁水温度和碳含量进行预测,当铁水温度和碳含量都满足工艺要求时,即可判断精炼期结束,即终点时刻,仿真结果验证了模型的准确性和有效性。根据各阶段不同的电能需求,分别给出电能输入方法,从理论上提出了一种新的电能输入方法：在满足单位时间供电损耗金额最小的约束条件时,确定矿热炉各个阶段的电压值,再结合各个阶段的用电量需求,得出电流值。最后利用JSP技术搭建了所提方法的能量输入优化平台。
通过仿真分析,验证了所提能量输入方法能够缩短矿热炉的冶炼时间,减少消耗。为矿热炉的生产过程实现节能降耗提供了理论依据。
Ferroalloy Plant is the energy-hungry factory, it's main equipment used for producing ferroalloy is submerged arc furnace. The energy smelting process needed mainly comes from electrical energy, electrical energy is converted into heat energy to melt furnace burden. The different furnace condition needs of different electrical energy, electricity requirements in different furnace condition can be met by power supply system, which directly affect the smelting speed, product quality, energy consumption and energy losses, etc. Therefore, determination of reasonable power supply system in view of the different furnace condition, providing reasonable secondary voltage and current value, for ferroalloy plant reasonable control smelting time, increase productivity and alloy quality, to achieve the purpose of saving energy and reducing consumption has important significance.
The paper on the basis of consulting a large number of relevant data and literatures, summarizes development present situation and development trend of the submerged arc furnace, applications of neural network in metallurgical industry as well as mechanical equipment, smelting process of submerged arc furnace and some treatment on the abnormal situation. With No.801 furnace of Sinosteel Jilin Ferroalloy Co., Ltd eighth branch as research background, the analysis of the smelting principle and process of the submerged arc furnace, based on energy balance on supply energy (electric energy and chemical heat release) and demand energy (consumption and wastage energy) establish the power input model, and determine the required total electric quantity; Based on the electrode position have judgment on three smelting stage of submerged arc furnace producing ferroalloy(including arc ignition、charging stage, melting stage, refining stage). But, because of the electrode position scarcely changed in the refining stage, can't have accurate judgement of the end of the refining period, so the paper forecasting molten iron temperature and carbon content of refining period by using neural network predictive model, when molten iron temperature and carbon content meet the technological requirements, refining period is over, namely the end time. The simulation results verify the validity and the accuracy of the models. Then put forward various power input method in accordance with the electricity demand of each stage. Theoretically propose a new power input method:In meeting the constraint condition that there is the smallest loss amount in power supply in unit of time, determine voltage values at all stages of submerged arc furnace, which are combined with electricity demand of all phases, it is concluded that the current value. Lastly, the energy input optimization platform is established to realize the proposed method by JSP technology.
Through the simulation analysis, validation proposed energy input method can shorten the time of smelting -arc furnace, reduce the consumption. And provide the theoretical basis for realizing that saving energy and reducing consumption in the production process of submerged arc furnace.
引文
[1]刘卫,王宏启.铁合金生产工艺与设备[M].北京：冶金工业出版社,2009-4.
[2]栾心汉等.铁合金生产节能及精炼技术[M].西安：西北工业大学出版社,2006-6.
[3]许传才.铁合金生产知识问答[M].北京：冶金工业出版社,2007-8.
[4]喻宗泉,喻晗.神经网络控制[M].西安：西安电子科技大学出版社,2009-1.
[5]姜长生等.智能控制与应用[M].北京：科学出版社,2007.
[6]邱东等.基于神经网络的高炉炉温预测模型的研究[J].人工智能,2009,25：5-2.
[7]袁平.电弧炉冶炼过程行进控制方法的研究与应用[D].沈阳：东北大学,2005-12.
[8]邓开楠.基于专家系统的电弧炉炉况判断方法研究[D].沈阳：东北大学,2005-2.
[9]韩颖.电弧炉能量输入模型的研究[D].沈阳：东北大学,2005-2.
[10]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计[M].北京：清华大学出版社,2005-7.
[11]楼顺天,姚若玉,沈俊霞.MATLAB7.X程序设计语言[M].西安：西安电子科技大学出版社,2007-8.
[12]李换琴,李晓华,万百五.用于大工业过程建模的新型小波神经网络结构[J].系统工程与电子技术2004,26(7)：941-944
[13]Chang-Xue Jack Feng; Abhirami C. Gowrisankar; Alice E. Smith; Zhi-Guang Samuel Yu; Practical Guidelines for Developing BP Neural Network Models of Measurement Uncertainty Data[J]. Journal of Manufacturing Systems,2006, vol.25 no.4, p.239-250
[14]许力.智能控制与智能控制系统[M].北京：机械工业出版社,2006-12.
[15]刘金琨.智能控制[M].北京：电子工业出版社,2005-5.
[16]刘祥官,刘芳.高炉炼铁过程优化与智能控制系统[M].北京：冶金工业出版社,2003-4.
[17]杨继进,闫立懿,吴振廷.直流电弧炉的供电与炉壁耐火材料消耗的关系[J].工业加热,1996(5)：21-24.
[18]叶蓉萱.电能需量模型及其控制策略研究[D].南京：南京理工大学,2007-6.
[19]高宪文.电弧炉炼钢过程建模与智能优化控制[M],沈阳：东北大学出版社,1999-2-16.
[20]杜俊红.在线超短期负荷预测的分析与应用[D].四川：四川大学,2004.
[21]张立明.人工神经网络的模型及其应用[M].上海：复旦大学出版社,1994.
[22]王白玲.电力负荷组合预测的理论方法及影响因素分析[D].北京：华北电力大学,2005.
[23]朱向阳.电力系统短期负荷预测及其应用系统[D].南京：东南大学,2004.
[24]陈兵,王华.基于神经网络的单步和多步电弧炉模型的研究与应用.现代电子技术,2003(2)：57-60.
[25]Hoffinan,A.J.Potchefstroom Univ.for CHE.Peak demand control in commercial buildings with target Peak adjustment based on load foreeasting.Proceedings of the 1998 IEEE Interational Conference on Control Applications Trieste,Italy,1998:1296-1292.
[26]武骏等.人工智能与交流电弧炉合理电气运行[J].包头钢铁学院学报,1999.
[27]顾根华等.考虑运行电抗与热平衡的电弧炉合理供电[J].加热设备,2001-1.
[28]谷玉合.电参数的发展趋势[C].世界电工会议论文,2002
[29]花皑,梁正敏.炼钢电弧炉的电气设备[M].北京：机械工业出版社,1987-8.
[30]王永骥,涂健.神经元网络控制[M],北京：机械工业出版社,1999.
[31]PBRAGA The high-impedance Arc Furnaee:A New High-efficiency for steel
making, Special Reprint Presented on the 4th European Electric Steel Congress[C], Madrid, 1992.
[32]G.GENSINI, New Developments in Electric Arc Furnace Technology[J].Metallurgical Plant And Technology,1991.
[33]S.Brown Electric arc furnace mast arm insulation old and new[J].Electric Furnace Conference Proceedings, Atinata, vol50,1992.
[34]冷劲松,宋福良,周继恩.铁合金矿热炉热平衡测试及能量损失分析[J].铁合金,2008(6)：10-13.
[35]文求实,罗隆福,陈光忠,童泽.矿热炉工业供电系统电能质量分析[J].铁合金,2009(1)：31-34.
[36]王威.及余热平衡的矿热炉控制特性研究[D].南京：东南大学,2009-4.
[37]唐恺,于荣祥,潘剑明.硅铁和工业硅的氧化精炼[J].浙江冶金,1998(1)：21-26.
[38]张得红,李红晓.75%硅铁杂质组成及精炼[J].铁合金,2004(1)：19-22.
[39]戴维,舒莉.铁合金冶金工程[M],北京：冶金出版社,1999.
[40]李春德.铁合金冶金学[M].北京：冶金工业出版社,1991.
[41]杨志忠,郭培民.中国铁合金工业的发展与思考[J].铁合金,2003(6)：40-46.
[42]王希廉.铁合金冶炼[M].北京：冶金工业出版社,1996.
[43]肖兴国,冶金反应工程学基础[M].北京：冶金工业出版社,1997.
[44]周进华.铁合金生产技术[M].北京：科学出版社,1991.
[45]B.M.斯特隆斯基.矿热熔炼炉[M].北京：冶金工业出版社,1984.
[46]郝俊宇,吕晓云等.矿热炉的计算机控制[J].冶金自动化,1992,11,16(6)：34-38.
[47]邹学柏,李江涛,吴祥宇.工业硅矿热炉自动控制系统的设计与应用[J].工业加热,2007,2：35-36.
[48]刘麟瑞,王玉珍.冶金炉料手册[M].北京：冶金工业出版社,2000.
[49]何波.低频矿热炉计算机控制系统及其控制算法研究[D].西安：西安建筑科技 大学,2000.
[50]彭小奇,梅炽等.多变量模糊控制模型辨识方法及其在矿热电炉决策支持系统中的应用[J].控制理论与应用,1994,11(5)：582-586.
[51]刘兆飞.基于熔池电阻实时计算的矿热炉自动控制系统研究[D].西安：西安建
筑科技大学,2007.
[52]Anna Huksdottir,et al. System Identification of a Three-phase Submerged-arc Ferrosilicon on Furnace[J]. IEEE Transaction Control Systems Technology,1995(12): 377-387.
[53]Anna Huksdottir,et al. Submerged-arc Ferrosilicon Furnace Simulator:Validation for Different Furnaces and Operating Ranges[J]. Control Engineering Practice.1998 (6): 1035-1042.
[54]Anna Huksdottir,et al. Current Control of a Three-phase Submerged Arc Ferrosilicon Furnace[J]. Control Engineering Practice.2002(10):1457-1463.
[55]S. A. Halvorsen, J. H. Downing,A. Schei. Developing a Unidimensinal Simimular Model for Producing Silicon in a Electric Furnace[J]. Proc. of the Fourth European Conference on Mathematics in Industry. B. G. Teubner, Stutgart and Kluwer Academic Publishers, Dordrecht,1991:293-298.
[56]Hilde Loken Larsen,Gudrun Sxvarsdottir,Jon Arne Bakken. Simulation of AC Arcs in the Silicon Metal Furnace[J]. Electric Furnace Conference Proceedings,1996,54: 157-168.
[57]G. A. Irons. Mathematical Modeling of Submerged Arc Smelting[J]. Electric Furnace Conference Proceedings,1998,56:311-317.
[58]张文怡,花皑.国外埋弧炉(矿热炉)的技术发展[J].工业加热,2006(6)：34-36.
[2]栾心汉等.铁合金生产节能及精炼技术[M].西安：西北工业大学出版社,2006-6.
[3]许传才.铁合金生产知识问答[M].北京：冶金工业出版社,2007-8.
[4]喻宗泉,喻晗.神经网络控制[M].西安：西安电子科技大学出版社,2009-1.
[5]姜长生等.智能控制与应用[M].北京：科学出版社,2007.
[6]邱东等.基于神经网络的高炉炉温预测模型的研究[J].人工智能,2009,25：5-2.
[7]袁平.电弧炉冶炼过程行进控制方法的研究与应用[D].沈阳：东北大学,2005-12.
[8]邓开楠.基于专家系统的电弧炉炉况判断方法研究[D].沈阳：东北大学,2005-2.
[9]韩颖.电弧炉能量输入模型的研究[D].沈阳：东北大学,2005-2.
[10]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计[M].北京：清华大学出版社,2005-7.
[11]楼顺天,姚若玉,沈俊霞.MATLAB7.X程序设计语言[M].西安：西安电子科技大学出版社,2007-8.
[12]李换琴,李晓华,万百五.用于大工业过程建模的新型小波神经网络结构[J].系统工程与电子技术2004,26(7)：941-944
[13]Chang-Xue Jack Feng; Abhirami C. Gowrisankar; Alice E. Smith; Zhi-Guang Samuel Yu; Practical Guidelines for Developing BP Neural Network Models of Measurement Uncertainty Data[J]. Journal of Manufacturing Systems,2006, vol.25 no.4, p.239-250
[14]许力.智能控制与智能控制系统[M].北京：机械工业出版社,2006-12.
[15]刘金琨.智能控制[M].北京：电子工业出版社,2005-5.
[16]刘祥官,刘芳.高炉炼铁过程优化与智能控制系统[M].北京：冶金工业出版社,2003-4.
[17]杨继进,闫立懿,吴振廷.直流电弧炉的供电与炉壁耐火材料消耗的关系[J].工业加热,1996(5)：21-24.
[18]叶蓉萱.电能需量模型及其控制策略研究[D].南京：南京理工大学,2007-6.
[19]高宪文.电弧炉炼钢过程建模与智能优化控制[M],沈阳：东北大学出版社,1999-2-16.
[20]杜俊红.在线超短期负荷预测的分析与应用[D].四川：四川大学,2004.
[21]张立明.人工神经网络的模型及其应用[M].上海：复旦大学出版社,1994.
[22]王白玲.电力负荷组合预测的理论方法及影响因素分析[D].北京：华北电力大学,2005.
[23]朱向阳.电力系统短期负荷预测及其应用系统[D].南京：东南大学,2004.
[24]陈兵,王华.基于神经网络的单步和多步电弧炉模型的研究与应用.现代电子技术,2003(2)：57-60.
[25]Hoffinan,A.J.Potchefstroom Univ.for CHE.Peak demand control in commercial buildings with target Peak adjustment based on load foreeasting.Proceedings of the 1998 IEEE Interational Conference on Control Applications Trieste,Italy,1998:1296-1292.
[26]武骏等.人工智能与交流电弧炉合理电气运行[J].包头钢铁学院学报,1999.
[27]顾根华等.考虑运行电抗与热平衡的电弧炉合理供电[J].加热设备,2001-1.
[28]谷玉合.电参数的发展趋势[C].世界电工会议论文,2002
[29]花皑,梁正敏.炼钢电弧炉的电气设备[M].北京：机械工业出版社,1987-8.
[30]王永骥,涂健.神经元网络控制[M],北京：机械工业出版社,1999.
[31]PBRAGA The high-impedance Arc Furnaee:A New High-efficiency for steel
making, Special Reprint Presented on the 4th European Electric Steel Congress[C], Madrid, 1992.
[32]G.GENSINI, New Developments in Electric Arc Furnace Technology[J].Metallurgical Plant And Technology,1991.
[33]S.Brown Electric arc furnace mast arm insulation old and new[J].Electric Furnace Conference Proceedings, Atinata, vol50,1992.
[34]冷劲松,宋福良,周继恩.铁合金矿热炉热平衡测试及能量损失分析[J].铁合金,2008(6)：10-13.
[35]文求实,罗隆福,陈光忠,童泽.矿热炉工业供电系统电能质量分析[J].铁合金,2009(1)：31-34.
[36]王威.及余热平衡的矿热炉控制特性研究[D].南京：东南大学,2009-4.
[37]唐恺,于荣祥,潘剑明.硅铁和工业硅的氧化精炼[J].浙江冶金,1998(1)：21-26.
[38]张得红,李红晓.75%硅铁杂质组成及精炼[J].铁合金,2004(1)：19-22.
[39]戴维,舒莉.铁合金冶金工程[M],北京：冶金出版社,1999.
[40]李春德.铁合金冶金学[M].北京：冶金工业出版社,1991.
[41]杨志忠,郭培民.中国铁合金工业的发展与思考[J].铁合金,2003(6)：40-46.
[42]王希廉.铁合金冶炼[M].北京：冶金工业出版社,1996.
[43]肖兴国,冶金反应工程学基础[M].北京：冶金工业出版社,1997.
[44]周进华.铁合金生产技术[M].北京：科学出版社,1991.
[45]B.M.斯特隆斯基.矿热熔炼炉[M].北京：冶金工业出版社,1984.
[46]郝俊宇,吕晓云等.矿热炉的计算机控制[J].冶金自动化,1992,11,16(6)：34-38.
[47]邹学柏,李江涛,吴祥宇.工业硅矿热炉自动控制系统的设计与应用[J].工业加热,2007,2：35-36.
[48]刘麟瑞,王玉珍.冶金炉料手册[M].北京：冶金工业出版社,2000.
[49]何波.低频矿热炉计算机控制系统及其控制算法研究[D].西安：西安建筑科技 大学,2000.
[50]彭小奇,梅炽等.多变量模糊控制模型辨识方法及其在矿热电炉决策支持系统中的应用[J].控制理论与应用,1994,11(5)：582-586.
[51]刘兆飞.基于熔池电阻实时计算的矿热炉自动控制系统研究[D].西安：西安建
筑科技大学,2007.
[52]Anna Huksdottir,et al. System Identification of a Three-phase Submerged-arc Ferrosilicon on Furnace[J]. IEEE Transaction Control Systems Technology,1995(12): 377-387.
[53]Anna Huksdottir,et al. Submerged-arc Ferrosilicon Furnace Simulator:Validation for Different Furnaces and Operating Ranges[J]. Control Engineering Practice.1998 (6): 1035-1042.
[54]Anna Huksdottir,et al. Current Control of a Three-phase Submerged Arc Ferrosilicon Furnace[J]. Control Engineering Practice.2002(10):1457-1463.
[55]S. A. Halvorsen, J. H. Downing,A. Schei. Developing a Unidimensinal Simimular Model for Producing Silicon in a Electric Furnace[J]. Proc. of the Fourth European Conference on Mathematics in Industry. B. G. Teubner, Stutgart and Kluwer Academic Publishers, Dordrecht,1991:293-298.
[56]Hilde Loken Larsen,Gudrun Sxvarsdottir,Jon Arne Bakken. Simulation of AC Arcs in the Silicon Metal Furnace[J]. Electric Furnace Conference Proceedings,1996,54: 157-168.
[57]G. A. Irons. Mathematical Modeling of Submerged Arc Smelting[J]. Electric Furnace Conference Proceedings,1998,56:311-317.
[58]张文怡,花皑.国外埋弧炉(矿热炉)的技术发展[J].工业加热,2006(6)：34-36.