基于劣化模型的伺服系统误差补偿方法研究
|
摘要
数控机床是现代化工业生产中重要的生产设备,是实现高效率、高精度加工的关键。数控机床水平和性能的好坏直接影响一个国家的工业制造水平。目前,数控技术正朝着高速、高精度方向发展,这不仅对数控机床有较高的性能以及功能要求,也期望数控机床在整个使用期限内都能保持良好的工作性能以及工作状态。提高数控机床性能的关键之一是提高机床数控系统的性能。而伺服系统是数控系统的重要组成部分,因此也是影响数控机床性能指标的重要环节。
鉴于目前半闭环伺服系统在我国广泛应用的现状,本文研究半闭环伺服系统在使用过程中的性能劣化过程,并针对位置精度降低、位置误差增大以及动态响应特性下降的现象,研究提高半闭环伺服系统性能的方法和措施。
首先,对半闭环进给伺服系统进行动力学分析并建立其仿真模型,分析数控机床位置误差产生原因以及伺服系统在其寿命周期内的劣化失效过程,阐明对伺服系统进行误差补偿的重要意义。
接着,分析半闭环伺服系统位置误差的小样本特性,引入统计学习理论中的支持向量机方法(SVM),建立位置误差劣化模型,实现对位置误差的预测。同时,针对SVM的参数选择特点,对SVM的参数优化过程进行改进,提高SVM的预测精度。
然后,以位置误差反馈补偿为第一次位置误差补偿,以模糊PID位置误差补偿为第二次位置误差补偿,建立半闭环伺服系统双层位置误差补偿结构。位置误差反馈补偿将位置误差劣化模型预测得到的位置误差反馈至伺服系统输入端,进行位置误差补偿,提高位置精度。为提高伺服系统的动态响应特性,在位置误差反馈补偿的基础上引入模糊PID控制,构成模糊PID位置误差补偿方法。同时,针对模糊PID位置误差补偿对位置精度的影响,对模糊PID控制的控制规则进行改进,使伺服系统同时具有较高的位置精度以及较优的动态响应特性。
最后,对进给伺服系统的主轴进行双层位置误差补偿仿真,对补偿效果进行验证,并介绍自主设计并开发的基于SVM的劣化预测软件。
The CNC machine plays an important role in modern industry production, and the performance of CNC machine reflects the manufacturing level of a country. A high manufacturing performance is required during the whole life of CNC machine in order to guarantee the manufacturing quality. Servo system is an essential part of numerical control system and has a strong influence on the performance of CNC machine.
The semi-closed loop servo system has been widely used in our country. Taking the degradation of position precision and dynamic response characteristics into consideration, this thesis analysis the performance degradation process of semi-closed loop servo system in its whole life cycle and then discusses the method of improving the performance of semi-closed loop servo system.
Firstly, this thesis performs the dynamic analysis on semi-closed loop feed servo system and builds the simulation model of servo system. And then, the thesis analysis the cause reason of position error and the process of degradation failure.
Secondly, support vector machine (SVM) is used to model and predict position error with small position error sample. The parameter optimization method of SVM is modified in order to improve the prediction precision.
Thirdly, the method of double position error compensation is proposed. The primary compensation is completed by feeding back the position error which is predicted by the position error degradation model to the input entry of servo system. In order to improve the dynamic property of servo system, the secondary compensation which combines the fuzzy PID control with the primary compensation is proposed. And the fuzzy control rules are modified to improve position precision further.
Finally, the double position error compensation is carried out on the simulation model of servo system and verifies the compensation results. A degradation prediction software based on SVM is designed and development.
鉴于目前半闭环伺服系统在我国广泛应用的现状,本文研究半闭环伺服系统在使用过程中的性能劣化过程,并针对位置精度降低、位置误差增大以及动态响应特性下降的现象,研究提高半闭环伺服系统性能的方法和措施。
首先,对半闭环进给伺服系统进行动力学分析并建立其仿真模型,分析数控机床位置误差产生原因以及伺服系统在其寿命周期内的劣化失效过程,阐明对伺服系统进行误差补偿的重要意义。
接着,分析半闭环伺服系统位置误差的小样本特性,引入统计学习理论中的支持向量机方法(SVM),建立位置误差劣化模型,实现对位置误差的预测。同时,针对SVM的参数选择特点,对SVM的参数优化过程进行改进,提高SVM的预测精度。
然后,以位置误差反馈补偿为第一次位置误差补偿,以模糊PID位置误差补偿为第二次位置误差补偿,建立半闭环伺服系统双层位置误差补偿结构。位置误差反馈补偿将位置误差劣化模型预测得到的位置误差反馈至伺服系统输入端,进行位置误差补偿,提高位置精度。为提高伺服系统的动态响应特性,在位置误差反馈补偿的基础上引入模糊PID控制,构成模糊PID位置误差补偿方法。同时,针对模糊PID位置误差补偿对位置精度的影响,对模糊PID控制的控制规则进行改进,使伺服系统同时具有较高的位置精度以及较优的动态响应特性。
最后,对进给伺服系统的主轴进行双层位置误差补偿仿真,对补偿效果进行验证,并介绍自主设计并开发的基于SVM的劣化预测软件。
The CNC machine plays an important role in modern industry production, and the performance of CNC machine reflects the manufacturing level of a country. A high manufacturing performance is required during the whole life of CNC machine in order to guarantee the manufacturing quality. Servo system is an essential part of numerical control system and has a strong influence on the performance of CNC machine.
The semi-closed loop servo system has been widely used in our country. Taking the degradation of position precision and dynamic response characteristics into consideration, this thesis analysis the performance degradation process of semi-closed loop servo system in its whole life cycle and then discusses the method of improving the performance of semi-closed loop servo system.
Firstly, this thesis performs the dynamic analysis on semi-closed loop feed servo system and builds the simulation model of servo system. And then, the thesis analysis the cause reason of position error and the process of degradation failure.
Secondly, support vector machine (SVM) is used to model and predict position error with small position error sample. The parameter optimization method of SVM is modified in order to improve the prediction precision.
Thirdly, the method of double position error compensation is proposed. The primary compensation is completed by feeding back the position error which is predicted by the position error degradation model to the input entry of servo system. In order to improve the dynamic property of servo system, the secondary compensation which combines the fuzzy PID control with the primary compensation is proposed. And the fuzzy control rules are modified to improve position precision further.
Finally, the double position error compensation is carried out on the simulation model of servo system and verifies the compensation results. A degradation prediction software based on SVM is designed and development.
引文
[1]刘玉荣.高速进给单元刚度与定位精度的研究.广东工业大学硕士学位论文. 2000
[2] P.S.Huang,J.Ni. Online error compensation of coordinate measuring machines [J]. International Journal of Machine Tools & Manufacture,1995,5:725-738.
[3] W.G.Weekers,P.H.J.Sehellekens. Assessment of Dynamic Error of CMMS for Fast Probing [J]. Annals of CIRP,1995(44),1:469-474.
[4]项伟红,郑力,刘大成,赵大泉.机床主轴热误差建模.制造技术与机床,2000,11:12-14.
[5]卢碧红,王启义,施志辉,葛研军,江渡.数控车床加工精度预测系统研究.组合机床与自动化加工技术,2001,5:7-9,12.
[6] Okuyama S,et al. Effect of floating capacity on the measurement error of the CBP method [M]. Advanced in Abrasive technology (II). YomanoPress Ltd. 1998.
[7] S.C.Veldhuis and M. A. Elbestawi,A Strategy for the Compensation of Error in Five-Axis Machining. Annals of the CIRP,1995(44),1:373-377.
[8] W.J.Love,A.J.Scarr. Determination of the volumetric accuracy of multi-axes machine [J]. M.T.D.R.Conf. Proc. 1973,14:307-315.
[9] D.French,S.H.Humphries. Compensation for backlash and alignment errors in a numerically controlled machine-tool by a digital computer program [J]. M.T.D.R. Conf. Proc. 1967,8:707-726.
[10] Kapoor S G,DeVor R E,Zhu R. Development of mechanistic models for the prediction of machining performance:model building methodology. Machining Science and Technology. 1998,2(2):2132388
[11] P.M.Ferreira,C.R.Liu. An Analytical Quadratic Model for the Geometric Errors of a Machine tool [J]. Journal of Manufacturing System,1986(5) 1:51-62.
[12] Donmez A. A General Methodology for Machine Tool Accuracy Enhancement by Error Compensation. Precision Engineering,1986(8),4:187-196.
[13] Z.J.Han , K.Zhou. Improvement of positioning accuracy of rotating table by microcomputer control compensation [J]. M.T.D.R. Conf. Proc.,1986,26:115-120
[14] M.Anjanappa ,D.K.Anand ,J.A.Kirk. Error correction methodologies and control strategies for numerical control machines [J]. Control Method for ManufacturingProcess,1988,7:41-49.
[15] http://mtamri.me.uiue.edtJ/Lastrevisedg/16/96
[16]李建广,姚英学.虚拟制造中的工件建模与描述.计算机集成制造系统,2000,6(l):54-59
[17]陈子辰.热模态理论和机床热态精度控制策略研究.浙江大学博士学位论文,1989.
[18]武建勇.数控机床几何误差识别方法的研究[M].河南科技大学,2003
[19]杨建国.数控双主轴车床几何和热误差综合数学模型及实时补偿[J].机械设计与研究,1998(1):44-46
[20]廖平兰.机床加工过程综合误差实时补偿技术.机械工程学报,1992,2:65-68
[21]金键,甘锡英.机床主轴温升和热变形在线检测及显示系统.机械与电子,1993,3:3-4
[22]叶佩青,汪劲松,杨开明,段广洪,张辉.数控机床误差补偿方法及其系统[P].中国专利:200410003487.5,2004-3-31.
[23]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.
[24] Graepel T,et al. Classification on proximity data with LP-machines[C],Ninth International Conference on Artificial Neural Networks IEEE. London:Confernce Publications,1999,470:304-309.
[25]刘江华,程君实,陈佳品.支持向量机训练算法综述[J].信息与控制,2002,31(1):45-49.
[26] Keerthi S S,et al. A Fast Iterative Nearest Point Algorithm for Support Vector Machine Classifier Design[C]. Indian Institute of Science Bangatolore,1999.
[27] Vapnik V,Golowich S,Smola A. Support Vector Method Approximation, Regression Estimation, and Signal Processing. In:Neural Information Processing Systems 9. Cambridge,MA,MIT Press,1997:281-187.
[28] Wei G,Sethi I K. Face detection for image annotation. Pattern Recognition,1999,20(11):1313-1321.
[29] Karlekar J,Desai U B. Finding faces in color images using wavelet transform. Proc IEEE Conference on Image Analysis and Processing. Venice,Italy,1999:1085-1088.
[30] Meng Shi,Yoshiharu Fujisawa,et al. Handwritten numeral recognition using gradient and curvature of gray scale image. Pattern Recognition,2002,35(10):2051-2059.
[31] Cheng-Lin Liu,Masaki Nakagawa. Evaluation of prototype learning algorithms for nearest-neighbor classifier in application to handwritten character recognition. PatternRecognition,2001,34(3):601-615.
[32]王一军,游思坤,李爱社,邓宏光,吴青凤.基于图像识别技术的马氏体识别模型研究.河南科技大学学报(自然科学版),2007,(04):4-6.
[33] Liu B M. Multiple resolution segrnentation of textured images. IEEE Trans on Pattern Analysis Machine Intelli-gence,1991,13(2):99-113.
[34]刘俊泉.交流伺服控制系统的设计仿真与智能控制.南京工业大学硕士学位论文,2003.
[35]郭敬枢,庄继东,孔峰.微机控制技术.重庆:重庆大学出版社,1994.
[36]张彦海,李睿敏,姜培刚.基于模糊自适应PID的数控机床进给控制器设计.青岛理工大学学报,2006,27(3):82-85.
[37]潘洪亮,姚建均.专家PID控制器在电液伺服系统中的应用研究.自动化技术与应用,2008(11):29-32.
[38]曾向昌.基于专家PID控制的双闭环调速系统仿真分析.吉林工程技术师范学院学报,2005(03):29-33.
[39]金鑫,刘吉臻,谭文.基于遗传算法的鲁棒PID控制器设计.华北电力大学学报,2004,131(3):39-42.
[40]韩华,罗安,杨勇.一种基于遗传算法的非线性PID控制器.控制与决策,2005,20(4):448-450.
[41]舒怀林. PID神经元网络多变量控制系统分析.自动化学报,1999,25(1):105-111.
[42]舒怀林. PID神经元网络对强耦合带时延多变量系统的解耦控制.控制理论与应用,1998,15(6):920-924.
[43]张远深,徐正华,张文涛,董仲斌.基于灰色PID的摆动气缸伺服系统位置跟踪研究.液压气动与密封,2007(4):27-29.
[44]李楠.灰色PID控制在磁盘驱动控制系统中的应用研究.现代电子技术,2007(18):137-139.
[45]吴祖育,秦鹏飞.数控机床[M].上海:上海科学技术出版社,1990.
[46]周勇,陈吉红,彭芳瑜.高速高精度数控进给驱动的机电联合系统仿真.机械科学与技术,2007,26(2):135-139.
[47]周勇.高速进给伺服系统动态特性分析及其运动控制研究.华中科技大学博士学位论文,2008.
[48]虞文华,吴昭同.伺服系统动特性对数控机床圆轨迹加工精度影响的机理.中国机械工程,1995,6(1):21-23.
[49]侯媛彬,汪梅,王立琦等.系统辨识及其MATLAB仿真.北京:科学出版社,2004.
[50]关丹阳.经济型数控机床位置误差的补偿及实验研究.吉林大学硕士学位论文,2007.
[51]董建国.数控机床伺服系统刚度对定位精度影响的研究.兰州理工大学硕士学位论文,2007.
[52]李皖.数控机床定位精度分析和误差补偿程序设计.安徽冶金科技职业学院学报,2001,11(3):17-24
[53]刘曙光,魏俊民,竺志超.模糊控制技术.北京:中国纺织出版社,2001.
[54]廉小亲.模糊控制技术.北京:中国电力出版社,2003.
[55]张俊芳.智能PID控制算法在闭环经济型数控机床中应用的研究.辽宁工程技术大学硕士学位论文,2003.
[56]史文浩,刘鸿雁.模糊自校正PID控制器在数控机床交流伺服系统中的应用.组合机床与自动化加工技术,2003(1):7-9
[57]敖荣庆.伺服系统.北京:航空工业出版社,2006.
[2] P.S.Huang,J.Ni. Online error compensation of coordinate measuring machines [J]. International Journal of Machine Tools & Manufacture,1995,5:725-738.
[3] W.G.Weekers,P.H.J.Sehellekens. Assessment of Dynamic Error of CMMS for Fast Probing [J]. Annals of CIRP,1995(44),1:469-474.
[4]项伟红,郑力,刘大成,赵大泉.机床主轴热误差建模.制造技术与机床,2000,11:12-14.
[5]卢碧红,王启义,施志辉,葛研军,江渡.数控车床加工精度预测系统研究.组合机床与自动化加工技术,2001,5:7-9,12.
[6] Okuyama S,et al. Effect of floating capacity on the measurement error of the CBP method [M]. Advanced in Abrasive technology (II). YomanoPress Ltd. 1998.
[7] S.C.Veldhuis and M. A. Elbestawi,A Strategy for the Compensation of Error in Five-Axis Machining. Annals of the CIRP,1995(44),1:373-377.
[8] W.J.Love,A.J.Scarr. Determination of the volumetric accuracy of multi-axes machine [J]. M.T.D.R.Conf. Proc. 1973,14:307-315.
[9] D.French,S.H.Humphries. Compensation for backlash and alignment errors in a numerically controlled machine-tool by a digital computer program [J]. M.T.D.R. Conf. Proc. 1967,8:707-726.
[10] Kapoor S G,DeVor R E,Zhu R. Development of mechanistic models for the prediction of machining performance:model building methodology. Machining Science and Technology. 1998,2(2):2132388
[11] P.M.Ferreira,C.R.Liu. An Analytical Quadratic Model for the Geometric Errors of a Machine tool [J]. Journal of Manufacturing System,1986(5) 1:51-62.
[12] Donmez A. A General Methodology for Machine Tool Accuracy Enhancement by Error Compensation. Precision Engineering,1986(8),4:187-196.
[13] Z.J.Han , K.Zhou. Improvement of positioning accuracy of rotating table by microcomputer control compensation [J]. M.T.D.R. Conf. Proc.,1986,26:115-120
[14] M.Anjanappa ,D.K.Anand ,J.A.Kirk. Error correction methodologies and control strategies for numerical control machines [J]. Control Method for ManufacturingProcess,1988,7:41-49.
[15] http://mtamri.me.uiue.edtJ/Lastrevisedg/16/96
[16]李建广,姚英学.虚拟制造中的工件建模与描述.计算机集成制造系统,2000,6(l):54-59
[17]陈子辰.热模态理论和机床热态精度控制策略研究.浙江大学博士学位论文,1989.
[18]武建勇.数控机床几何误差识别方法的研究[M].河南科技大学,2003
[19]杨建国.数控双主轴车床几何和热误差综合数学模型及实时补偿[J].机械设计与研究,1998(1):44-46
[20]廖平兰.机床加工过程综合误差实时补偿技术.机械工程学报,1992,2:65-68
[21]金键,甘锡英.机床主轴温升和热变形在线检测及显示系统.机械与电子,1993,3:3-4
[22]叶佩青,汪劲松,杨开明,段广洪,张辉.数控机床误差补偿方法及其系统[P].中国专利:200410003487.5,2004-3-31.
[23]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.
[24] Graepel T,et al. Classification on proximity data with LP-machines[C],Ninth International Conference on Artificial Neural Networks IEEE. London:Confernce Publications,1999,470:304-309.
[25]刘江华,程君实,陈佳品.支持向量机训练算法综述[J].信息与控制,2002,31(1):45-49.
[26] Keerthi S S,et al. A Fast Iterative Nearest Point Algorithm for Support Vector Machine Classifier Design[C]. Indian Institute of Science Bangatolore,1999.
[27] Vapnik V,Golowich S,Smola A. Support Vector Method Approximation, Regression Estimation, and Signal Processing. In:Neural Information Processing Systems 9. Cambridge,MA,MIT Press,1997:281-187.
[28] Wei G,Sethi I K. Face detection for image annotation. Pattern Recognition,1999,20(11):1313-1321.
[29] Karlekar J,Desai U B. Finding faces in color images using wavelet transform. Proc IEEE Conference on Image Analysis and Processing. Venice,Italy,1999:1085-1088.
[30] Meng Shi,Yoshiharu Fujisawa,et al. Handwritten numeral recognition using gradient and curvature of gray scale image. Pattern Recognition,2002,35(10):2051-2059.
[31] Cheng-Lin Liu,Masaki Nakagawa. Evaluation of prototype learning algorithms for nearest-neighbor classifier in application to handwritten character recognition. PatternRecognition,2001,34(3):601-615.
[32]王一军,游思坤,李爱社,邓宏光,吴青凤.基于图像识别技术的马氏体识别模型研究.河南科技大学学报(自然科学版),2007,(04):4-6.
[33] Liu B M. Multiple resolution segrnentation of textured images. IEEE Trans on Pattern Analysis Machine Intelli-gence,1991,13(2):99-113.
[34]刘俊泉.交流伺服控制系统的设计仿真与智能控制.南京工业大学硕士学位论文,2003.
[35]郭敬枢,庄继东,孔峰.微机控制技术.重庆:重庆大学出版社,1994.
[36]张彦海,李睿敏,姜培刚.基于模糊自适应PID的数控机床进给控制器设计.青岛理工大学学报,2006,27(3):82-85.
[37]潘洪亮,姚建均.专家PID控制器在电液伺服系统中的应用研究.自动化技术与应用,2008(11):29-32.
[38]曾向昌.基于专家PID控制的双闭环调速系统仿真分析.吉林工程技术师范学院学报,2005(03):29-33.
[39]金鑫,刘吉臻,谭文.基于遗传算法的鲁棒PID控制器设计.华北电力大学学报,2004,131(3):39-42.
[40]韩华,罗安,杨勇.一种基于遗传算法的非线性PID控制器.控制与决策,2005,20(4):448-450.
[41]舒怀林. PID神经元网络多变量控制系统分析.自动化学报,1999,25(1):105-111.
[42]舒怀林. PID神经元网络对强耦合带时延多变量系统的解耦控制.控制理论与应用,1998,15(6):920-924.
[43]张远深,徐正华,张文涛,董仲斌.基于灰色PID的摆动气缸伺服系统位置跟踪研究.液压气动与密封,2007(4):27-29.
[44]李楠.灰色PID控制在磁盘驱动控制系统中的应用研究.现代电子技术,2007(18):137-139.
[45]吴祖育,秦鹏飞.数控机床[M].上海:上海科学技术出版社,1990.
[46]周勇,陈吉红,彭芳瑜.高速高精度数控进给驱动的机电联合系统仿真.机械科学与技术,2007,26(2):135-139.
[47]周勇.高速进给伺服系统动态特性分析及其运动控制研究.华中科技大学博士学位论文,2008.
[48]虞文华,吴昭同.伺服系统动特性对数控机床圆轨迹加工精度影响的机理.中国机械工程,1995,6(1):21-23.
[49]侯媛彬,汪梅,王立琦等.系统辨识及其MATLAB仿真.北京:科学出版社,2004.
[50]关丹阳.经济型数控机床位置误差的补偿及实验研究.吉林大学硕士学位论文,2007.
[51]董建国.数控机床伺服系统刚度对定位精度影响的研究.兰州理工大学硕士学位论文,2007.
[52]李皖.数控机床定位精度分析和误差补偿程序设计.安徽冶金科技职业学院学报,2001,11(3):17-24
[53]刘曙光,魏俊民,竺志超.模糊控制技术.北京:中国纺织出版社,2001.
[54]廉小亲.模糊控制技术.北京:中国电力出版社,2003.
[55]张俊芳.智能PID控制算法在闭环经济型数控机床中应用的研究.辽宁工程技术大学硕士学位论文,2003.
[56]史文浩,刘鸿雁.模糊自校正PID控制器在数控机床交流伺服系统中的应用.组合机床与自动化加工技术,2003(1):7-9
[57]敖荣庆.伺服系统.北京:航空工业出版社,2006.