基于红外反射法的金属薄板印涂湿膜厚度测量及曲面拟合研究
|
摘要
为了实现金属薄板印涂厚度的评估,设计了一种基于红外反射法的金属薄板印涂湿膜厚度在线检测系统,此系统主要包括机械结构和测控系统2部分。通过测量生产线上关键点厚度值拟合单板单线和多板全平面的印涂湿膜厚度,运用Matlab和上下四分位截断法对测得的厚度数据进行剔除粗大误差处理,最后利用BP神经网络法对处理后的数据进行曲面拟合和仿真。拟合的结果表明:设计的金属薄板印涂湿膜厚度测量系统可以对生产线上的一批金属薄板进行整体的厚度评估;BP神经网络曲面拟合方法精度高,曲面较平滑,为金属薄板印涂厚度的评估提供了指导作用。
In order to realize the evaluation of sheet metal printing and coating thickness,an online detection system based on infrared reflection method of sheet metal printing and coating wet film thickness was designed. The system mainly includes two parts,mechanical structure and control system. By measuring the key point on the production line to fitting single in single-line and multi-board plane printed wet film thickness,Matlab and upper and lower four bit truncation method was used to remove the gross error processing according to measured thickness data. Finally,the BP neural network method was used to fit and simulate the processed data. The fitting results show that the design of the metal sheet printing and coating wet film thickness measurement system can gauge the overall thickness of a number of metal sheet on the production line and the BP neural network curve fitting method has high precision and smooth surface,which provides a strong guidance for the assessment of the thickness of the metal sheet.
In order to realize the evaluation of sheet metal printing and coating thickness,an online detection system based on infrared reflection method of sheet metal printing and coating wet film thickness was designed. The system mainly includes two parts,mechanical structure and control system. By measuring the key point on the production line to fitting single in single-line and multi-board plane printed wet film thickness,Matlab and upper and lower four bit truncation method was used to remove the gross error processing according to measured thickness data. Finally,the BP neural network method was used to fit and simulate the processed data. The fitting results show that the design of the metal sheet printing and coating wet film thickness measurement system can gauge the overall thickness of a number of metal sheet on the production line and the BP neural network curve fitting method has high precision and smooth surface,which provides a strong guidance for the assessment of the thickness of the metal sheet.
引文
[1]LI X J,ZHANG C,ZHANG N,et al.Determination of coating thickness by crater grounding method[J].Advanced Materials Research,2014,833(11):237-240.
[2]任吉林,曾亮,张丽攀,等.碳纤维复合材料涂层厚度涡流法测量的研究[J].仪器仪表学报,2011,32(12):2662-2668.
[3]郑岗,刘丁,张震,等.基于提离点的脉冲涡流测厚研究[J].仪器仪表学报,2008,29(8):1745-1749.
[4]田世锋,桂卫华.几种测厚仪器的研究现状及其应用[J].中南工业大学学报(自然科学版),2003,34(z1):130-132.
[5]唐伟,屈国普,赵越.用于特征X射线测厚仪的数据处理器电路设计[J].湖南科技学院学报,2010,31(8):22-23.
[6]DILL S,ROBIGER V.Coating thickness measurement of thin gold and palladium coatings on printed circuit boards using X-ray fluorescence[J].Circuit World,2011,37(2):20-26.
[7]曹坚,钱苏翔,杨世锡,等.基于激光-CCD的高精度在线测厚系统研究[J].机械设计与研究,2006,22(6):91-98.
[8]赵斌.一种基于红外成像的薄膜测厚仪:中国,103175478A[P].2013-06-26.
[9]邱自学.金属薄板印刷涂层的湿膜厚度差分测量及均匀性评估方法:104613883A[P].2015-05-13.
[10]曾清红,卢德唐.基于移动最小二乘法的曲线曲面拟合[J].工程图学学报,2004(1):87-88.
[11]何兵朋,冯仁忠,余胜蛟.基于差分进化算法的B样条曲线曲面拟合[J].图学学报,2016,37(2):182-183.
[12]韩旭里,庄陈坚,刘新儒.基于径向基函数与B样条的散乱数据拟合方法[J].计算技术与自动化,2007,26(1):63-65.
[13]谢驰,吕育梅,王萍,等.基于神经网络的曲面拟合方法研究[J].四川大学学报,2006,40(12):47-48.
[14]樊振宇.BP神经网络模型与学习算法[J].软件导刊,2011.10(7):66.
[15]林瑞国,刘青,卢竞宪,等.IM-C型红外水份/薄膜厚度测试仪的研制[J].仪表技术与传感器,1998(10):17-18.
[16]乔运英,郭晓松,朱智,等.基于虚拟仪器的起竖实验系统设计与实现[J].计算机测量与控制,2010,18(9):2125-2129.
[17]刘彩红.BP神经网络学习算法的研究[D].重庆:重庆师范大学,2008.
[18]柳松青.Matlab神经网络BP网络研究与应用[J].计算机工程与设计,2003,24(11):82-83.
[19]罗成汉.基于MATLAB神经网络工具箱的BP网络实现[J].计算机仿真,2004,21(5):109-11.
[2]任吉林,曾亮,张丽攀,等.碳纤维复合材料涂层厚度涡流法测量的研究[J].仪器仪表学报,2011,32(12):2662-2668.
[3]郑岗,刘丁,张震,等.基于提离点的脉冲涡流测厚研究[J].仪器仪表学报,2008,29(8):1745-1749.
[4]田世锋,桂卫华.几种测厚仪器的研究现状及其应用[J].中南工业大学学报(自然科学版),2003,34(z1):130-132.
[5]唐伟,屈国普,赵越.用于特征X射线测厚仪的数据处理器电路设计[J].湖南科技学院学报,2010,31(8):22-23.
[6]DILL S,ROBIGER V.Coating thickness measurement of thin gold and palladium coatings on printed circuit boards using X-ray fluorescence[J].Circuit World,2011,37(2):20-26.
[7]曹坚,钱苏翔,杨世锡,等.基于激光-CCD的高精度在线测厚系统研究[J].机械设计与研究,2006,22(6):91-98.
[8]赵斌.一种基于红外成像的薄膜测厚仪:中国,103175478A[P].2013-06-26.
[9]邱自学.金属薄板印刷涂层的湿膜厚度差分测量及均匀性评估方法:104613883A[P].2015-05-13.
[10]曾清红,卢德唐.基于移动最小二乘法的曲线曲面拟合[J].工程图学学报,2004(1):87-88.
[11]何兵朋,冯仁忠,余胜蛟.基于差分进化算法的B样条曲线曲面拟合[J].图学学报,2016,37(2):182-183.
[12]韩旭里,庄陈坚,刘新儒.基于径向基函数与B样条的散乱数据拟合方法[J].计算技术与自动化,2007,26(1):63-65.
[13]谢驰,吕育梅,王萍,等.基于神经网络的曲面拟合方法研究[J].四川大学学报,2006,40(12):47-48.
[14]樊振宇.BP神经网络模型与学习算法[J].软件导刊,2011.10(7):66.
[15]林瑞国,刘青,卢竞宪,等.IM-C型红外水份/薄膜厚度测试仪的研制[J].仪表技术与传感器,1998(10):17-18.
[16]乔运英,郭晓松,朱智,等.基于虚拟仪器的起竖实验系统设计与实现[J].计算机测量与控制,2010,18(9):2125-2129.
[17]刘彩红.BP神经网络学习算法的研究[D].重庆:重庆师范大学,2008.
[18]柳松青.Matlab神经网络BP网络研究与应用[J].计算机工程与设计,2003,24(11):82-83.
[19]罗成汉.基于MATLAB神经网络工具箱的BP网络实现[J].计算机仿真,2004,21(5):109-11.