基于上位机的旋转电弧信号处理及系统的智能化研究
|
摘要
焊缝跟踪是现代焊接技术中的一个主要的研究方向之一。本文以旋转电弧焊缝跟踪系统为研究对象,首先研究了基于特征谐波法的焊缝方向检测,为焊接小车的控制提供了有效的控制参数。在焊缝跟踪的整体控制中,提出了结合积分差值法和特征谐波法,分别控制十字滑架和焊接小车。系统的协调过程中,采用Agent及多Agent为智能主体的控制方案,用来协调各种方法与各个控制部分。控制系统以上位机为实现平台,上位机以个人电脑(PC机)为基础,以串口为通信渠道,以Visual C++为编程工具,以MATLAB为计算工具。本文以上位机为基础,设计了信号处理系统,对焊缝跟踪系统进行智能化研究。
本文发现在焊缝扫描过程中,扫描的时移特性放映了焊缝方向与焊接方向的偏差角,经过傅立叶变换后,其特征谐向量的相位角会发生改变,这个改变的角度就是焊缝方向与焊接方向的夹角。这个夹角可以用来作为焊接小车的控制量,用来保证焊接小车的方向与焊缝方向一致。本文以MATLAB为处理工具,通过离散傅立叶变换检测出了焊缝方向的夹角,并且进行了仿真,在理论上验证了可行性。
由于需要使用积分差值法与特征谐波法相结合的模式才能更加完整的提取特征量,因此提出建立焊缝跟踪系统的上位机系统,采取上位机与下位机分工合作的形式实现两种方法的结合。焊缝跟踪的控制中,在上位机的基础上,提出采用Agent与多Agent智能控制的思想,用于协调各部分的工作,在智能控制方面,作者认为智能Agent应该以智能主体的地位进行控制,真正实现Agent的自治性与主动性,并认为主动性是智能主体的一个基本的属性。
由于PC机的性能优异,而且软件成熟通用,因此上位机以PC机为基础。在Visual C++的基础上编写上位机程序,考虑到MATLAB在矩阵运算与复数运算方面的特长,在离散傅立叶变换时在Visual C++的基础上调用MATLAB引擎进行运算,解决了计算工具的问题。
上位机与下位机的通信与协调通过串口进行,串口的硬件简单,能够满足上位机与下位机的数据传递要求,而且连接的距离较长,稳定性好。串口的底层协议采用RS-232串口协议,本文还制定应用层的通信协议,以保证上位机与下位机的互相理解。
Seam tracking is a modern welding technology major research direction. By a rotary arc seam tracking system for the study object, the first research the weld direction detection method based on the characteristics harmonic, for welding car provides effective control parameters. In the overall control of seam tracking, the proposed method combining the characteristics harmonic method and integral D-value method respectively control the cross slide frame and welding car. In system’s coordination process, the use of Intelligent Agent and Multi Agent as intelligent main block to the control program, to coordinate the each method and control part. The control system has a host computer as a platform, has a PC as hardware base, has serial port for communications channels, makes Visual C++ as programming tools and makes MATLAB for the calculation tool. This paper is based on host computer, designs the signal processing system and research seam tracking system’s intelligent.
This dissertation found in the weld during the scan, the time-shift have shown the characteristics angle between welding seam direction and the direction of welding, through the Fourier transform, its characteristic vector’s the phase angle of harmonic change, this change of perspective is the angle between the welding seam direction and the direction of welding. This angle can be used as a welding carriage control value, to ensure the car's direction and weld seam in the same direction. In this paper, MATLAB as the processing tool, by discrete Fourier transform to detect the angle of the direction of the weld and simulation it, the feasibility is verified in theory.
As the need to use the Combination of the integral D-value method and the characteristic harmonic method to get a more complete extraction of characteristic quantities, the seam tracking system establishes the host computer system. Host computer and lower computer division of two parts to achieve combination. In seam tracking control, based on the host computer, intelligent control based on Agent and Multi-Agent are used, to coordinate the work of the various parts, in intelligent control, the Intelligent Agent is on the main block status, realize Agent’s autonomy and initiative, and think that the initiative is a basic agent properties.
As the PC-performance, and software maturity generic, so host computer is based on PC. Programming is based on Visual C++, taking into account the MATLAB can do matrix operations and complex operations, when in the discrete Fourier transform the Visual C++ call MATLAB engine operation to solve computing instrument.
Communication and coordination between host computer and lower computer is through the serial port, serial port hardware is simple, can satisfy the host computer and lower computer's data transmission requirements and long distance connections, and has good stability. The underlying protocol uses serial RS-232 serial protocol; the paper has developed the application layer communication protocol, to ensure the host computer and lower computer understanding each other.
本文发现在焊缝扫描过程中,扫描的时移特性放映了焊缝方向与焊接方向的偏差角,经过傅立叶变换后,其特征谐向量的相位角会发生改变,这个改变的角度就是焊缝方向与焊接方向的夹角。这个夹角可以用来作为焊接小车的控制量,用来保证焊接小车的方向与焊缝方向一致。本文以MATLAB为处理工具,通过离散傅立叶变换检测出了焊缝方向的夹角,并且进行了仿真,在理论上验证了可行性。
由于需要使用积分差值法与特征谐波法相结合的模式才能更加完整的提取特征量,因此提出建立焊缝跟踪系统的上位机系统,采取上位机与下位机分工合作的形式实现两种方法的结合。焊缝跟踪的控制中,在上位机的基础上,提出采用Agent与多Agent智能控制的思想,用于协调各部分的工作,在智能控制方面,作者认为智能Agent应该以智能主体的地位进行控制,真正实现Agent的自治性与主动性,并认为主动性是智能主体的一个基本的属性。
由于PC机的性能优异,而且软件成熟通用,因此上位机以PC机为基础。在Visual C++的基础上编写上位机程序,考虑到MATLAB在矩阵运算与复数运算方面的特长,在离散傅立叶变换时在Visual C++的基础上调用MATLAB引擎进行运算,解决了计算工具的问题。
上位机与下位机的通信与协调通过串口进行,串口的硬件简单,能够满足上位机与下位机的数据传递要求,而且连接的距离较长,稳定性好。串口的底层协议采用RS-232串口协议,本文还制定应用层的通信协议,以保证上位机与下位机的互相理解。
Seam tracking is a modern welding technology major research direction. By a rotary arc seam tracking system for the study object, the first research the weld direction detection method based on the characteristics harmonic, for welding car provides effective control parameters. In the overall control of seam tracking, the proposed method combining the characteristics harmonic method and integral D-value method respectively control the cross slide frame and welding car. In system’s coordination process, the use of Intelligent Agent and Multi Agent as intelligent main block to the control program, to coordinate the each method and control part. The control system has a host computer as a platform, has a PC as hardware base, has serial port for communications channels, makes Visual C++ as programming tools and makes MATLAB for the calculation tool. This paper is based on host computer, designs the signal processing system and research seam tracking system’s intelligent.
This dissertation found in the weld during the scan, the time-shift have shown the characteristics angle between welding seam direction and the direction of welding, through the Fourier transform, its characteristic vector’s the phase angle of harmonic change, this change of perspective is the angle between the welding seam direction and the direction of welding. This angle can be used as a welding carriage control value, to ensure the car's direction and weld seam in the same direction. In this paper, MATLAB as the processing tool, by discrete Fourier transform to detect the angle of the direction of the weld and simulation it, the feasibility is verified in theory.
As the need to use the Combination of the integral D-value method and the characteristic harmonic method to get a more complete extraction of characteristic quantities, the seam tracking system establishes the host computer system. Host computer and lower computer division of two parts to achieve combination. In seam tracking control, based on the host computer, intelligent control based on Agent and Multi-Agent are used, to coordinate the work of the various parts, in intelligent control, the Intelligent Agent is on the main block status, realize Agent’s autonomy and initiative, and think that the initiative is a basic agent properties.
As the PC-performance, and software maturity generic, so host computer is based on PC. Programming is based on Visual C++, taking into account the MATLAB can do matrix operations and complex operations, when in the discrete Fourier transform the Visual C++ call MATLAB engine operation to solve computing instrument.
Communication and coordination between host computer and lower computer is through the serial port, serial port hardware is simple, can satisfy the host computer and lower computer's data transmission requirements and long distance connections, and has good stability. The underlying protocol uses serial RS-232 serial protocol; the paper has developed the application layer communication protocol, to ensure the host computer and lower computer understanding each other.
引文
[1]潘际銮.现代弧焊控制[M].北京:机械工业出版社, 2000:171-188.
[2]杨春利,电弧焊基础[M].哈尔滨:哈尔滨工业大学出版社,2003.
[3]陈善本.焊接过程现代控制技术[M].哈尔滨:哈尔滨工业大学出版社, 2001.1.
[4]林尚扬.焊接过程的低成本自动化[C].焊接学会十周年纪念文集.焊接. 1993(1).
[5]潘际銮.论我国焊接结构生产自动化的方向[C].全国“机械热加工未来学”学术讨论会. 1988.5.
[6]毛鹏军,黄石生,薛家祥等.弧焊机器人焊缝跟踪系统研究现状及发展趋势[J].电焊机,2001, 31(10):9-12.
[7]陈武柱.图象处理在窄间隙MIG/MAG焊在线检测中的运用[J].焊接学报,1998,(4).
[8] C.Umeagukwn,B.Maqueira,J.Jarzynski,Application of ultrasonic sensors to robotic seam tracking[J],IEEE Trans,Robotics Automat,1989,5(3):337-344.
[9] E.L.Estochen,C.P.Neuman, F.B.Prinz. Application of acoustic sensors to robotic seam tracking[J]. IEEE trans,On industrial electronic,1984,31(3):219-224.
[10] R.Fenn,etc. Ultrasonic monitoring and control during arc welding[J].Welding Journal,1985,18-22.
[11]胡绳荪,张绍彬,侯文考.焊缝跟踪中的非接触式中的超声波传感器的研究[J].仪表技术与传感器,1999,4(2):5-6.
[12]张绍彬,等.用于弧焊焊缝跟踪的空气超声波传感系统[J].电焊机, 1994, (5):10-12.
[13]营泰雄,町田光.水中におけわ溶接线の自动追踪制御た对ヰゐ超音波セフシンング法の适用[C].日本机械学会论文集(C集),1995, 61(3):494-500(In Japanese).
[14]陈武柱,王勇,林青松.焊缝自动跟踪技术的现状及新发展[J].焊管,1992,15(5):1-10.
[15] E.Ide, etc.The Present Sensing Technique in Welding Process [J].Journal of JWS, 1991.60(2) (in Japanese).
[16]姜焕中等.焊接方法及设备(第一分册)[M].北京:机械工业出版社,1981.
[17]潘际銮,陈武柱等.焊缝棱边双向激光自动跟踪系统[J].焊接学报, 1984,5(2) .
[18]陈武柱,林青松等.红外焊缝跟踪传感器的光源调制及信号处理[C].X-Ⅻa-033-90,第六届全国焊接学术会议论文选集.
[19] Nomura H.et al. Automatic real-time bead height control with arc sensor in TIG welding, IIW Doc XII-884-85.
[20] Nishada Y.et al. Fully automatic control of vertical CO2 welding[R]. National Technical Report, 23(2).
[21]潘际銮.新型自动焊接跟踪系统[J].机械工程学报,1980,1(1):1-14.
[22] Murakarni S. Weld-line Tracking Control of Arc Welding Robot Using Fuzzy Logic Controller [J]. Fussy Sets and Systems, 1989.32(2):221-237.
[23] J W KIM, S J NA.A Self-Organizing Fuzzy Control Approach to Arc Sensor for Welding Joint Tracking in Gas Metal Arc Welding of Butt Joint[J]. Welding Journal,1993,72(20):60-66.
[24]费跃农.电弧传感器焊缝自动跟踪系统及电弧传感器基础理论研究[D].北京,清华大学,1990.
[25]吴世德,廖宝剑等.高速旋转电弧传感器[J].焊接学报,1997,18(1):61-66.
[26]廖宝剑.以电弧为传感器的多自由度智能焊接系统研究[D].北京,清华大学,1993.
[27]潘际銮等.空心轴电机驱动的旋转扫描焊炬[P].中国专利,No.922444765.
[28]廖宝剑等.实现电弧高速横向扫描坡口的焊炬机构[J].机械研究与应用,1995,(2):5-8.
[29]贾剑平,张华等.用于弧焊机器人的新型高速旋转电弧传感器的研制[J].南昌大学学报(工科版),2000,22(3):1-3, 18.
[30]熊震宇,张华等.旋转电弧传感器的研制[J].仪表技术与传感器,2003,(7):1-3.
[31]潘际銮,费跃农等.电弧传感器物理数学模型研究[C].第六届全国焊接学术会议文集,第六集,1990:106-163.
[32]廖宝剑,实现电弧高速横向扫描坡口的焊炬机构[J].机械研究与应用,1995,(2)5-8.
[33] Murakarni S. Weld-line Tracking Control of Arc Welding Robot Using Fuzzy Logic Controller [J]. Fussy Sets and Systems, 1989, 32(2) 221-237.
[34]屈岳波,洪波,潘际銮等.一种自动跟踪焊接小车的驱动方法[J].焊接技术,2004,33(6): 38-40.
[35]蔡自兴.智能控制原理与应用[M].北京:清华大学出版社,2007.
[36]李衍达.人工智能及其应用.第2版[M].北京:清华人学出版社,1996.
[37]史忠植.智能体及其应用[M].北京:科学出版社,2000.
[38]蔡自兴等.智能科学发展的若干问题[J].自动化学报,2002,(28):142-150.
[39]蔡自兴智能控制四元结构的研究[J].高技术通信,1996,19-23.
[40] Cai Zixing, Fu, K·S.Expert-System-based robot planning. Control Theory and Applications, 1988, 5(2):30-37.
[41]蔡自兴,肖晓明.动态模糊系统的鲁棒稳定性分析与控制器设计[J].中南工业大学学报,1999,30(4):418-421.
[42] Sugeno M, Kang GT. Structure identification of fuzzy model [J].Fuzzy Sets and Systems.1988.28(1):15-33.
[43]蔡自兴.模糊控制的典型结构[J].工业控制计算机,1997,(3):1-4.
[44] Kung S Y, Hwang J N.Neural network architecture for robotic applications[J]. IEEE Trans on Robotics and Automation, 1989, 5(4):641-657.
[45] Wang Y N.A neural network fuzzy self-organizing control for manipulator[J]. Int J Advances in Modeling &Analysis, 1995, 49(2):8-19.
[46]蔡自兴,王耀南.机器人逆模神经控制及其应用[J].中南工业大学学报,1998,29(1):80-84.
[47]宋一兵,杨永田,李忠池.基于命令的多智能体组织模型研究[J],哈尔滨工业大学学报,2009,41(1):219-222.
[48]曾伟,费奇.基于MAS理论的组织建模[J].华中理工大学学报,2000,8( 8): 24-26.
[49]徐晋晖,张伟,石纯一等.面向结构的Agent组织形成和演化机制[J].计算机研究与发展2001,38(8): 897-903.
[50]吴元斌.单agent强化学习与多agent强化学习比较研究[J].电脑与信息技术.2009,17(1):8-11.
[51]吴大正.信号与线性系统分析(第4版)[M].北京:高等教育出版社,2005.
[52]马家辰,孙玉德,张颖,等.M CS-51单片机原理及接日技术[M].哈尔滨:哈尔滨工业大学出版社,1997.
[53]任丽香,马淑芬,李方慧.TMS320C6000系列DSP的原理与应用.北京:电子工业出版社,2000.
[54]潘策,陈晓南.PC机与单片机串行通信的硬件设计[J].网络与通信,2003,158(15):4-6.
[55]求是科技,李现勇.Visual C++串口通信技术与工程实践[M].北京,人民邮电出版社,2002.
[56]万新光,王培东.MCS_51单片机开发系统结构及并行监控程序设计[J].哈尔滨科学技术大学学报.1987, 4:76-80.
[57]樊荣.实时嵌入式图像处理并行监控系统的研究与实现[D].华中科技大学硕士学位论文,2005.
[58]涂刚,阳富民,胡贯荣.嵌入式操作系统综述.计算机应用研究.2000(11):4-6.
[59]苏金明,黄国明,刘波.MATLAB与外部程序接口.电子工业出版社,北京,2004.
[60]张仁忠,常明志,许德新.利用MFC实现双缓存机制改善图形的显示效果[J].应用科技,2005,32,(1):47-49.
[61]韩丽娜,石昊苏.使用双缓存来解决GDI下的闪烁问题[J].计算机工程与设计,2006,27,(17):3258-3260.
[2]杨春利,电弧焊基础[M].哈尔滨:哈尔滨工业大学出版社,2003.
[3]陈善本.焊接过程现代控制技术[M].哈尔滨:哈尔滨工业大学出版社, 2001.1.
[4]林尚扬.焊接过程的低成本自动化[C].焊接学会十周年纪念文集.焊接. 1993(1).
[5]潘际銮.论我国焊接结构生产自动化的方向[C].全国“机械热加工未来学”学术讨论会. 1988.5.
[6]毛鹏军,黄石生,薛家祥等.弧焊机器人焊缝跟踪系统研究现状及发展趋势[J].电焊机,2001, 31(10):9-12.
[7]陈武柱.图象处理在窄间隙MIG/MAG焊在线检测中的运用[J].焊接学报,1998,(4).
[8] C.Umeagukwn,B.Maqueira,J.Jarzynski,Application of ultrasonic sensors to robotic seam tracking[J],IEEE Trans,Robotics Automat,1989,5(3):337-344.
[9] E.L.Estochen,C.P.Neuman, F.B.Prinz. Application of acoustic sensors to robotic seam tracking[J]. IEEE trans,On industrial electronic,1984,31(3):219-224.
[10] R.Fenn,etc. Ultrasonic monitoring and control during arc welding[J].Welding Journal,1985,18-22.
[11]胡绳荪,张绍彬,侯文考.焊缝跟踪中的非接触式中的超声波传感器的研究[J].仪表技术与传感器,1999,4(2):5-6.
[12]张绍彬,等.用于弧焊焊缝跟踪的空气超声波传感系统[J].电焊机, 1994, (5):10-12.
[13]营泰雄,町田光.水中におけわ溶接线の自动追踪制御た对ヰゐ超音波セフシンング法の适用[C].日本机械学会论文集(C集),1995, 61(3):494-500(In Japanese).
[14]陈武柱,王勇,林青松.焊缝自动跟踪技术的现状及新发展[J].焊管,1992,15(5):1-10.
[15] E.Ide, etc.The Present Sensing Technique in Welding Process [J].Journal of JWS, 1991.60(2) (in Japanese).
[16]姜焕中等.焊接方法及设备(第一分册)[M].北京:机械工业出版社,1981.
[17]潘际銮,陈武柱等.焊缝棱边双向激光自动跟踪系统[J].焊接学报, 1984,5(2) .
[18]陈武柱,林青松等.红外焊缝跟踪传感器的光源调制及信号处理[C].X-Ⅻa-033-90,第六届全国焊接学术会议论文选集.
[19] Nomura H.et al. Automatic real-time bead height control with arc sensor in TIG welding, IIW Doc XII-884-85.
[20] Nishada Y.et al. Fully automatic control of vertical CO2 welding[R]. National Technical Report, 23(2).
[21]潘际銮.新型自动焊接跟踪系统[J].机械工程学报,1980,1(1):1-14.
[22] Murakarni S. Weld-line Tracking Control of Arc Welding Robot Using Fuzzy Logic Controller [J]. Fussy Sets and Systems, 1989.32(2):221-237.
[23] J W KIM, S J NA.A Self-Organizing Fuzzy Control Approach to Arc Sensor for Welding Joint Tracking in Gas Metal Arc Welding of Butt Joint[J]. Welding Journal,1993,72(20):60-66.
[24]费跃农.电弧传感器焊缝自动跟踪系统及电弧传感器基础理论研究[D].北京,清华大学,1990.
[25]吴世德,廖宝剑等.高速旋转电弧传感器[J].焊接学报,1997,18(1):61-66.
[26]廖宝剑.以电弧为传感器的多自由度智能焊接系统研究[D].北京,清华大学,1993.
[27]潘际銮等.空心轴电机驱动的旋转扫描焊炬[P].中国专利,No.922444765.
[28]廖宝剑等.实现电弧高速横向扫描坡口的焊炬机构[J].机械研究与应用,1995,(2):5-8.
[29]贾剑平,张华等.用于弧焊机器人的新型高速旋转电弧传感器的研制[J].南昌大学学报(工科版),2000,22(3):1-3, 18.
[30]熊震宇,张华等.旋转电弧传感器的研制[J].仪表技术与传感器,2003,(7):1-3.
[31]潘际銮,费跃农等.电弧传感器物理数学模型研究[C].第六届全国焊接学术会议文集,第六集,1990:106-163.
[32]廖宝剑,实现电弧高速横向扫描坡口的焊炬机构[J].机械研究与应用,1995,(2)5-8.
[33] Murakarni S. Weld-line Tracking Control of Arc Welding Robot Using Fuzzy Logic Controller [J]. Fussy Sets and Systems, 1989, 32(2) 221-237.
[34]屈岳波,洪波,潘际銮等.一种自动跟踪焊接小车的驱动方法[J].焊接技术,2004,33(6): 38-40.
[35]蔡自兴.智能控制原理与应用[M].北京:清华大学出版社,2007.
[36]李衍达.人工智能及其应用.第2版[M].北京:清华人学出版社,1996.
[37]史忠植.智能体及其应用[M].北京:科学出版社,2000.
[38]蔡自兴等.智能科学发展的若干问题[J].自动化学报,2002,(28):142-150.
[39]蔡自兴智能控制四元结构的研究[J].高技术通信,1996,19-23.
[40] Cai Zixing, Fu, K·S.Expert-System-based robot planning. Control Theory and Applications, 1988, 5(2):30-37.
[41]蔡自兴,肖晓明.动态模糊系统的鲁棒稳定性分析与控制器设计[J].中南工业大学学报,1999,30(4):418-421.
[42] Sugeno M, Kang GT. Structure identification of fuzzy model [J].Fuzzy Sets and Systems.1988.28(1):15-33.
[43]蔡自兴.模糊控制的典型结构[J].工业控制计算机,1997,(3):1-4.
[44] Kung S Y, Hwang J N.Neural network architecture for robotic applications[J]. IEEE Trans on Robotics and Automation, 1989, 5(4):641-657.
[45] Wang Y N.A neural network fuzzy self-organizing control for manipulator[J]. Int J Advances in Modeling &Analysis, 1995, 49(2):8-19.
[46]蔡自兴,王耀南.机器人逆模神经控制及其应用[J].中南工业大学学报,1998,29(1):80-84.
[47]宋一兵,杨永田,李忠池.基于命令的多智能体组织模型研究[J],哈尔滨工业大学学报,2009,41(1):219-222.
[48]曾伟,费奇.基于MAS理论的组织建模[J].华中理工大学学报,2000,8( 8): 24-26.
[49]徐晋晖,张伟,石纯一等.面向结构的Agent组织形成和演化机制[J].计算机研究与发展2001,38(8): 897-903.
[50]吴元斌.单agent强化学习与多agent强化学习比较研究[J].电脑与信息技术.2009,17(1):8-11.
[51]吴大正.信号与线性系统分析(第4版)[M].北京:高等教育出版社,2005.
[52]马家辰,孙玉德,张颖,等.M CS-51单片机原理及接日技术[M].哈尔滨:哈尔滨工业大学出版社,1997.
[53]任丽香,马淑芬,李方慧.TMS320C6000系列DSP的原理与应用.北京:电子工业出版社,2000.
[54]潘策,陈晓南.PC机与单片机串行通信的硬件设计[J].网络与通信,2003,158(15):4-6.
[55]求是科技,李现勇.Visual C++串口通信技术与工程实践[M].北京,人民邮电出版社,2002.
[56]万新光,王培东.MCS_51单片机开发系统结构及并行监控程序设计[J].哈尔滨科学技术大学学报.1987, 4:76-80.
[57]樊荣.实时嵌入式图像处理并行监控系统的研究与实现[D].华中科技大学硕士学位论文,2005.
[58]涂刚,阳富民,胡贯荣.嵌入式操作系统综述.计算机应用研究.2000(11):4-6.
[59]苏金明,黄国明,刘波.MATLAB与外部程序接口.电子工业出版社,北京,2004.
[60]张仁忠,常明志,许德新.利用MFC实现双缓存机制改善图形的显示效果[J].应用科技,2005,32,(1):47-49.
[61]韩丽娜,石昊苏.使用双缓存来解决GDI下的闪烁问题[J].计算机工程与设计,2006,27,(17):3258-3260.