大口径厚壁长管件挤压成型数值分析及优化
|
摘要
大口径厚壁无缝钢管是火电、核电设备使用的重要基础构件，目前超(超)临界火电第三代核电设备急需解决的耐高温大口径厚壁无缝钢管在国内无法满足需求，而且传统的生产工艺生产的无缝钢管无法满足质量要求，尤其是P91铁素体耐热钢等高端核电用管和高压锅炉管产品，国内能够批量生产且产品质量稳定、工艺稳定性可靠和有生产经验的厂家很少。本文针对这一亟待解决的问题，对P91大口径厚壁无缝钢管的垂直挤压工艺过程进行了数值模拟研究，并对其工艺进行了优化。
本文利用GLEEBLE-3500热模拟试验机对P91材料进行了热模拟实验，获得了P91材料的高温流动应力应变曲线，并对该实验结果进行了分析，该流动应力应变曲线作为数值模拟中材料的基础参数。
本文利用DEFORM-3D数值模拟软件对两种典型电站用钢管559mm(外径)×339mm(内径)×12000mm(管长)以及813 mm(外径)×679 mm(内径)×6500mm(管长)的垂直挤压过程进行了数值模拟。在数值模拟过程中分别针对挤压比、凹模锥角以及定径带长度三个参数，通过数值模拟结果分析了这三个参数对大口径厚壁无缝钢管垂直挤压过程中的成型壁厚、等效应力、等效应变、温度场以及凸模载荷的影响。
经过对两种不同管件的挤压成型过程数值模拟分析以后，利用神经网络结合遗传算法对两种管件的成型工艺进行了优化，最终确定Φ559mm×Φ339mm×12000mm管件的挤压比为5.8，凹模锥角15°，定径带长度为122mm；Φ813mm×Φ679mm×6500mm管件的挤压比为6.5，凹模锥角为20°，定径带长度为160mm时为最优工艺，该工艺为实际生产提供了一定的理论分析依据。
Large-diameter seamless steel tube with thick wall is one of the most important basiccomponents in the thermal and nuclear power equipments. At present the domestic can’t meetthe demand of high temperature resistant large-diameter seamless steel tube with thick wallthat in need of solution in the equipments of super (ultra) critical thermal power of the thirdgeneration. The traditional production technology of seamless steel tube can’t afford thequality requirements. The manufacturers in the domestic that can produce the pipes withstable quality, reliable process stability and are full of experience are seldom, especially forthe high level nuclear power and high-pressure boil tubes that are made of P91 heat resistancesteel of ferrite. In order to solve this problem, we simulated the vertical extrusion process ofP91 large-diameter seamless steel tube with thick wall, and optimized the technology.
During the research we had a thermal simulation experiment by GLEEBLE-3500thermal simulation test machine, got the high temperature flow strain-stress curves of P91material, and analyzed the experiment results. The high temperature flow strain-stress wasused as a basic parameter of the numerical simulation.
In this research, we used the finite element simulation software of DEFORM-3D tosimulate the vertical extrusion process of the two kinds of typical pipes whose sizes are
559mm×339mm×12000mm and 813 mm×679 mm×6500mm. During the research,we changed the numerical values of extrusion ratio, cone angle and sizing belt height.Through the simulation results we analyzed the effection of those parameters on the formingpipes’wall thickness, effective stress, effective strain, temperature and the load of top dieduring the extrusion process.
After the analysis of the vertical extrusion processes of the two pipes, we optimized thepipes’forming technologies using the method of neural network combined with geneticalgorithm, and finally determined the best technology to forming the aimed pipes. That arewhen the extrusion is 5.8, the cone angle is 15°and the sizing belt height is 122mm, the pipeof 559mm×339mm×12000mm can form best and when the extrusion is 6.5, the coneangle is 20°and the sizing belt height is 160mm, the pipe of 813 mm×679 mm×6500mmcan form best. All those research can provide certain theoretical analysis basis for practical.
本文利用GLEEBLE-3500热模拟试验机对P91材料进行了热模拟实验，获得了P91材料的高温流动应力应变曲线，并对该实验结果进行了分析，该流动应力应变曲线作为数值模拟中材料的基础参数。
本文利用DEFORM-3D数值模拟软件对两种典型电站用钢管559mm(外径)×339mm(内径)×12000mm(管长)以及813 mm(外径)×679 mm(内径)×6500mm(管长)的垂直挤压过程进行了数值模拟。在数值模拟过程中分别针对挤压比、凹模锥角以及定径带长度三个参数，通过数值模拟结果分析了这三个参数对大口径厚壁无缝钢管垂直挤压过程中的成型壁厚、等效应力、等效应变、温度场以及凸模载荷的影响。
经过对两种不同管件的挤压成型过程数值模拟分析以后，利用神经网络结合遗传算法对两种管件的成型工艺进行了优化，最终确定Φ559mm×Φ339mm×12000mm管件的挤压比为5.8，凹模锥角15°，定径带长度为122mm；Φ813mm×Φ679mm×6500mm管件的挤压比为6.5，凹模锥角为20°，定径带长度为160mm时为最优工艺，该工艺为实际生产提供了一定的理论分析依据。
Large-diameter seamless steel tube with thick wall is one of the most important basiccomponents in the thermal and nuclear power equipments. At present the domestic can’t meetthe demand of high temperature resistant large-diameter seamless steel tube with thick wallthat in need of solution in the equipments of super (ultra) critical thermal power of the thirdgeneration. The traditional production technology of seamless steel tube can’t afford thequality requirements. The manufacturers in the domestic that can produce the pipes withstable quality, reliable process stability and are full of experience are seldom, especially forthe high level nuclear power and high-pressure boil tubes that are made of P91 heat resistancesteel of ferrite. In order to solve this problem, we simulated the vertical extrusion process ofP91 large-diameter seamless steel tube with thick wall, and optimized the technology.
During the research we had a thermal simulation experiment by GLEEBLE-3500thermal simulation test machine, got the high temperature flow strain-stress curves of P91material, and analyzed the experiment results. The high temperature flow strain-stress wasused as a basic parameter of the numerical simulation.
In this research, we used the finite element simulation software of DEFORM-3D tosimulate the vertical extrusion process of the two kinds of typical pipes whose sizes are
559mm×339mm×12000mm and 813 mm×679 mm×6500mm. During the research,we changed the numerical values of extrusion ratio, cone angle and sizing belt height.Through the simulation results we analyzed the effection of those parameters on the formingpipes’wall thickness, effective stress, effective strain, temperature and the load of top dieduring the extrusion process.
After the analysis of the vertical extrusion processes of the two pipes, we optimized thepipes’forming technologies using the method of neural network combined with geneticalgorithm, and finally determined the best technology to forming the aimed pipes. That arewhen the extrusion is 5.8, the cone angle is 15°and the sizing belt height is 122mm, the pipeof 559mm×339mm×12000mm can form best and when the extrusion is 6.5, the coneangle is 20°and the sizing belt height is 160mm, the pipe of 813 mm×679 mm×6500mmcan form best. All those research can provide certain theoretical analysis basis for practical.
引文
[1]李晓红.国内大直径无缝钢管生产发展的装备选择[J].钢管. 2006,35(6):4~7.
[2]郭元蓉,吴红.P91无缝钢管国产化研究进展[J].钢管. 2008, 37(1):22~26.
[3]杨秀琴.我国钢管工业的现状、问题与发展前景(上)[J].钢管. 2008, 37(1):12~17.
[4]杨秀琴.我国钢管工业的现状、问题与发展前景(下)[J].钢管. 2008, 37(2):1~14.
[5]殷国茂.我国钢管飞速发展的10年概述[J].钢管. 2009, 38(2):1~14.6~13.
[6]陈明香,胡路卓,刘彦斌.大口径无缝钢管的工艺研究[J].天津冶金. 2002, 111(5):23~25.
[7]罗灿,罗涛,郑世建等.大直径无缝钢管反挤压冲孔和顶管拔伸+斜轧组合生产工艺的研究及实践.钢管.2010,39(4):22~24.
[8]刘春海,赵敬彬,王艳杰.宏润重工超(超)临界机组用大口径厚壁P91/P92钢管的研制与生产.600Mw／1000Mw超超临界机组新型钢国产化研讨会.中国扬州,2009.03.
[9]程世清,黄文.超(超)临界电站用P91大口径厚壁无缝钢管研制及其批量化生产[A].北京科技大学.第六届无缝钢管生产技术研讨会(国际)论文集[C].北京:北京科技大学,2009.
[10]何彪,肖功业,王国亮等.P91热轧无缝钢管的研制开发[J].天津冶金.2008,140(5):37~39.
[11]胡永平,王荣森,周仲成.采用锻造镗孔技术制造大口径厚壁9%Cr无缝钢管. [A].北京科技大学.第六届无缝钢管生产技术研讨会(国际)论文集[C].北京:北京科技大学,2009.
[12]周仲成,胡永平,崔玉军.锻造镗孔技术在大口径厚壁无缝钢管生产中的应用[J].大型铸锻件.2007,2:22~25.
[13]薛菲菲,隋健,李培力.金属挤压技工技术综述[J].科技资讯-工程技术.2007,04:79~80.
[14]光明日报.世界最大黑色金属垂直挤压机问世[J].有色设备.2009(4).
[15]谢建新,刘静安.金属挤压理论与技术[M].北京:冶金工业出版社,2001.
[16]张力.有限元法及ANSYS程序应用基础[M].北京:科学出版社,2008.
[17]刘建生,陈慧琴,郭晓霞.金属塑性加工有限元模拟技术[M].北京:冶金工业出版社,2003.
[18]张彦敏,张学宾,龚红英.有限元在金属塑性成型中的应用[M].北京:化学工业出版社,2010.
[19]李传民,王向丽,闫华军.DEFORM 5.03金属成形有限元分析实例指导教程[M].北京:机械工业出版社, 2009.
[20]史忠植.神经网络[M].高等教育出版社,2009.
[21]周景润,张丽娜.基于MATLAB与FUZZYTECH的模糊与神经网络设计[M].电子工业出版社.2010.
[22]葛哲学,孙志强.神经网络理论与MATLAB实现[M].北京:电子工业出版社, 2007.
[23]丁士圻,郭丽华.人工神经网络基础[M].哈尔滨:哈尔滨工程大学出版社, 2008.
[24] Bryson A E, Ho Y C. Applied optimal control[M]. New York:Blaisdell, 1969.
[25] Werbos P J. Beyond regression: New tools for prediction and analysis in the behavioral sciences [D].Cambridge, MA: Harvard University, 1974.
[26] Parker D B. Learning logic [R] // Center for computational research in economics and managementsciences. Technical report TR-47. Cambridge, MA:MIT,1985.
[27] Rumelhart D E, Hinton G E, Williams R J. Learning representation of back-propagation errors[J].Nature, 1986(323):533-536.
[28]罗晓曙.人工神经网络[M].广西师范大学出版社,2005.
[29]郑准备,张兵,王小迎.P91钢的性能与组织结构研究[J].热加工工艺, 2008,37(24):57~58.
[30]王雪凤,吴任东,邓晨曦等.新型耐热高强钢P91的高温力学性能[J].机械工程学报,2008,44(6):243~246.
[31]林肇杰,程世长,钟倩霞.钢9l无缝钢管的生产和应用[J].特殊钢, 1996,17(6):22~24．
[32]何彪,肖功业,王国亮等.P91热轧无缝钢管的研制开发[J].天津冶金, 2008,5:37.
[33]王如军,徐银庚,陈伟等.超(超)临界锅炉用耐热钢T/P91的性能及应用.现代冶金[J].2009,37(3):1~3.
[34]李志翔,李晓平,赵永强.火力发电厂新材料T91/P91钢性能综述.云南电力技术[J]. 2004,32:15~17.
[35] A.H. Yaghi, T.H. Hyde, A.A. Becker, J.A. Williams, W. Sun. Residual stress simulation in weldedsections of P91 pipes.Journal of Material Processing Technology. 2005:480~487.
[36] Feng Li, Li Li, Xiang Wang, et. Optimizing the Seamless Tube Extrusion Process using the FiniteElement Method.JOM, 2010:70~73.
[37] Chu-Ho Liu, A-cheng Wang, Yi-sian Chen, et a1. The coupled thermo-mechanical analysis in theupsetting process by the dynamic FEM[J]. Journal of Materials Processing Technology, 2008. 201:37-42.
[38] Ying Fu Qiang, Pan 130 Song. Analysis on temperature distribution in cross wedge rolling processwith finite element method[J]. Journal of Materials Processing Technology, 2007. 187-188: 392-396.
[39]李维特.热应力理论分析及应用[M].北京:中国电力出版社. 2004(6).
[40]郭学兵.圆柱形工件加热过程中热应力的数值模拟[J].能源技术与管理, 2010(1): 112-114.
[41]赵选民.试验设计方法[M].北京:科学出版社, 2006.
[42]雷英杰,张善文,李续武,等.MATLAB遗传算法工具箱及应用[M].西安:西安电子科技大学出版社,2005.
[2]郭元蓉,吴红.P91无缝钢管国产化研究进展[J].钢管. 2008, 37(1):22~26.
[3]杨秀琴.我国钢管工业的现状、问题与发展前景(上)[J].钢管. 2008, 37(1):12~17.
[4]杨秀琴.我国钢管工业的现状、问题与发展前景(下)[J].钢管. 2008, 37(2):1~14.
[5]殷国茂.我国钢管飞速发展的10年概述[J].钢管. 2009, 38(2):1~14.6~13.
[6]陈明香,胡路卓,刘彦斌.大口径无缝钢管的工艺研究[J].天津冶金. 2002, 111(5):23~25.
[7]罗灿,罗涛,郑世建等.大直径无缝钢管反挤压冲孔和顶管拔伸+斜轧组合生产工艺的研究及实践.钢管.2010,39(4):22~24.
[8]刘春海,赵敬彬,王艳杰.宏润重工超(超)临界机组用大口径厚壁P91/P92钢管的研制与生产.600Mw／1000Mw超超临界机组新型钢国产化研讨会.中国扬州,2009.03.
[9]程世清,黄文.超(超)临界电站用P91大口径厚壁无缝钢管研制及其批量化生产[A].北京科技大学.第六届无缝钢管生产技术研讨会(国际)论文集[C].北京:北京科技大学,2009.
[10]何彪,肖功业,王国亮等.P91热轧无缝钢管的研制开发[J].天津冶金.2008,140(5):37~39.
[11]胡永平,王荣森,周仲成.采用锻造镗孔技术制造大口径厚壁9%Cr无缝钢管. [A].北京科技大学.第六届无缝钢管生产技术研讨会(国际)论文集[C].北京:北京科技大学,2009.
[12]周仲成,胡永平,崔玉军.锻造镗孔技术在大口径厚壁无缝钢管生产中的应用[J].大型铸锻件.2007,2:22~25.
[13]薛菲菲,隋健,李培力.金属挤压技工技术综述[J].科技资讯-工程技术.2007,04:79~80.
[14]光明日报.世界最大黑色金属垂直挤压机问世[J].有色设备.2009(4).
[15]谢建新,刘静安.金属挤压理论与技术[M].北京:冶金工业出版社,2001.
[16]张力.有限元法及ANSYS程序应用基础[M].北京:科学出版社,2008.
[17]刘建生,陈慧琴,郭晓霞.金属塑性加工有限元模拟技术[M].北京:冶金工业出版社,2003.
[18]张彦敏,张学宾,龚红英.有限元在金属塑性成型中的应用[M].北京:化学工业出版社,2010.
[19]李传民,王向丽,闫华军.DEFORM 5.03金属成形有限元分析实例指导教程[M].北京:机械工业出版社, 2009.
[20]史忠植.神经网络[M].高等教育出版社,2009.
[21]周景润,张丽娜.基于MATLAB与FUZZYTECH的模糊与神经网络设计[M].电子工业出版社.2010.
[22]葛哲学,孙志强.神经网络理论与MATLAB实现[M].北京:电子工业出版社, 2007.
[23]丁士圻,郭丽华.人工神经网络基础[M].哈尔滨:哈尔滨工程大学出版社, 2008.
[24] Bryson A E, Ho Y C. Applied optimal control[M]. New York:Blaisdell, 1969.
[25] Werbos P J. Beyond regression: New tools for prediction and analysis in the behavioral sciences [D].Cambridge, MA: Harvard University, 1974.
[26] Parker D B. Learning logic [R] // Center for computational research in economics and managementsciences. Technical report TR-47. Cambridge, MA:MIT,1985.
[27] Rumelhart D E, Hinton G E, Williams R J. Learning representation of back-propagation errors[J].Nature, 1986(323):533-536.
[28]罗晓曙.人工神经网络[M].广西师范大学出版社,2005.
[29]郑准备,张兵,王小迎.P91钢的性能与组织结构研究[J].热加工工艺, 2008,37(24):57~58.
[30]王雪凤,吴任东,邓晨曦等.新型耐热高强钢P91的高温力学性能[J].机械工程学报,2008,44(6):243~246.
[31]林肇杰,程世长,钟倩霞.钢9l无缝钢管的生产和应用[J].特殊钢, 1996,17(6):22~24．
[32]何彪,肖功业,王国亮等.P91热轧无缝钢管的研制开发[J].天津冶金, 2008,5:37.
[33]王如军,徐银庚,陈伟等.超(超)临界锅炉用耐热钢T/P91的性能及应用.现代冶金[J].2009,37(3):1~3.
[34]李志翔,李晓平,赵永强.火力发电厂新材料T91/P91钢性能综述.云南电力技术[J]. 2004,32:15~17.
[35] A.H. Yaghi, T.H. Hyde, A.A. Becker, J.A. Williams, W. Sun. Residual stress simulation in weldedsections of P91 pipes.Journal of Material Processing Technology. 2005:480~487.
[36] Feng Li, Li Li, Xiang Wang, et. Optimizing the Seamless Tube Extrusion Process using the FiniteElement Method.JOM, 2010:70~73.
[37] Chu-Ho Liu, A-cheng Wang, Yi-sian Chen, et a1. The coupled thermo-mechanical analysis in theupsetting process by the dynamic FEM[J]. Journal of Materials Processing Technology, 2008. 201:37-42.
[38] Ying Fu Qiang, Pan 130 Song. Analysis on temperature distribution in cross wedge rolling processwith finite element method[J]. Journal of Materials Processing Technology, 2007. 187-188: 392-396.
[39]李维特.热应力理论分析及应用[M].北京:中国电力出版社. 2004(6).
[40]郭学兵.圆柱形工件加热过程中热应力的数值模拟[J].能源技术与管理, 2010(1): 112-114.
[41]赵选民.试验设计方法[M].北京:科学出版社, 2006.
[42]雷英杰,张善文,李续武,等.MATLAB遗传算法工具箱及应用[M].西安:西安电子科技大学出版社,2005.