轴承滚子表面疵病检测方法的研究
发布时间：2019-05-23 20:09
【摘要】：在圆锥滚子轴承中,滚子表面如果出现裂纹、斑点等表面疵病,将对轴承质量产生严重影响,直接导致轴承使用寿命的降低。因此在工业生产中,需对滚子表面进行100%的缺陷检测。就目前人工检测方法而言,已经满足不了工业发展越来越高效与高精度的要求,因此需要对滚子表面疵病的自动检测方法进行研究。 本文在总结轴承钢球自动检测的研究成果的基础上,结合CCD技术、计算机图像处理及模式识别技术,提出一种新的滚子表面图像缺陷自动检测方法。结合滚子检测的特点,介绍系统的自动检测原理及工作流程；提出滚子自动送料装置、自动展开装置、以及自动分拣装置等的设计方案,实现对滚子高速、完整的展开以及快速、准确的分拣；在选择适当的CCD相机及光学显微镜的同时,对光源进行优化设计,实现滚子表面图像的高清晰采集。利用数字图像处理及模式识别技术,完成对图像信息的相应处理与分析及对缺陷信息的对比与识别,最终完成对不合格品的分拣。 该方法实现了圆锥滚子表面疵病的实时检测,可以对直径6-15mm的滚子进行精度达5μm的表面疵病检测,检测准确率达99%以上,具有操作方便、自动化程度高、出错率低、检测周期短等优点,大大降低了检测成本。
[Abstract]:In the tapered roller bearing, if cracks, spots, etc. are present on the surface of the roller, the bearing quality can be seriously affected, and the service life of the bearing can be reduced directly. Therefore, in the industrial production, it is necessary to carry out 100% defect detection on the surface of the roller. As far as the current manual detection method is concerned, the requirement of more and more efficient and high-precision for industrial development has been satisfied, and therefore, the automatic detection method for the surface smoothness of the roller needs to be researched. On the basis of summarizing the research results of automatic detection of bearing steel balls, this paper presents a new automatic detection method for image defects of roller surface based on CCD technology, computer image processing and pattern recognition technology. According to the characteristics of roller detection, the automatic detection principle and working flow of the system are introduced. The design scheme of automatic roller feeding device, automatic unwinding device and automatic sorting device is put forward to realize the high-speed, complete deployment of the roller and the quick and accurate distribution of the roller. selecting the appropriate CCD camera and the optical microscope, and optimizing the light source so as to realize high-definition production of the roller surface image; The method of digital image processing and pattern recognition is used to complete the corresponding processing and analysis of the image information and the comparison and identification of the defect information. The method realizes the real-time detection of the surface smoothness of the conical roller, and can detect the surface smoothness of the roller with the diameter of 6-15 mm up to 5. m u.m, the detection accuracy is more than 99%, the method has the advantages of convenient operation, high degree of automation, low error rate and short detection period and the like, so that the detection is greatly reduced,
【学位授予单位】：长春理工大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TH133.3;TP391.41
本文编号：2484190
[Abstract]:In the tapered roller bearing, if cracks, spots, etc. are present on the surface of the roller, the bearing quality can be seriously affected, and the service life of the bearing can be reduced directly. Therefore, in the industrial production, it is necessary to carry out 100% defect detection on the surface of the roller. As far as the current manual detection method is concerned, the requirement of more and more efficient and high-precision for industrial development has been satisfied, and therefore, the automatic detection method for the surface smoothness of the roller needs to be researched. On the basis of summarizing the research results of automatic detection of bearing steel balls, this paper presents a new automatic detection method for image defects of roller surface based on CCD technology, computer image processing and pattern recognition technology. According to the characteristics of roller detection, the automatic detection principle and working flow of the system are introduced. The design scheme of automatic roller feeding device, automatic unwinding device and automatic sorting device is put forward to realize the high-speed, complete deployment of the roller and the quick and accurate distribution of the roller. selecting the appropriate CCD camera and the optical microscope, and optimizing the light source so as to realize high-definition production of the roller surface image; The method of digital image processing and pattern recognition is used to complete the corresponding processing and analysis of the image information and the comparison and identification of the defect information. The method realizes the real-time detection of the surface smoothness of the conical roller, and can detect the surface smoothness of the roller with the diameter of 6-15 mm up to 5. m u.m, the detection accuracy is more than 99%, the method has the advantages of convenient operation, high degree of automation, low error rate and short detection period and the like, so that the detection is greatly reduced,
【学位授予单位】：长春理工大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TH133.3;TP391.41
【参考文献】
相关期刊论文 前10条
1 陈禹伶;何臻祥;;模糊PID控制在缝纫机压框调速的踏板控制系统中的应用[J];兵工自动化;2010年02期
2 杨东林;于正林;;轴承钢球表面缺陷的快速检测方法[J];兵工学报;2009年06期
3 蒋沂萍;原材料缺陷对钢球质量的影响[J];轴承;2002年08期
4 米玉红;;圆锥滚子轴承的失效形式[J];轴承;2006年07期
5 蒋德军;;大锥角圆锥滚子轴承的计算机辅助优化设计[J];湖南工程学院学报(自然科学版);2007年02期
6 韩文;王艳;吕建英;孔达;;圆锥滚子磨加工表面粗糙度影响因素的分析[J];轴承;2005年12期
7 刘川;刘景林;;基于Simulink仿真的步进电机闭环控制系统分析[J];测控技术;2009年01期
8 周景亮;林志熙;黄耀志;;应用声振法对轴承滚子无损检测的研究[J];机电技术;2006年04期
9 陈晓鹏;;机器视觉技术[J];机电一体化;2008年05期
10 曹东海,卢泽生;静摩擦因数与表面粗糙度关系的实验分析[J];机械传动;2005年05期
相关博士学位论文 前3条
1 王义文;钢球表面缺陷检测关键技术研究及样机研制[D];哈尔滨理工大学;2010年
2 王鹏;基于运动视觉技术的钢球表面缺陷检测[D];哈尔滨理工大学;2008年
3 韦宏强;机器视觉技术在弹药测试设备中的应用研究[D];长春理工大学;2009年
，本文编号：2484190
本文链接：https://www.wllwen.com/kejilunwen/jixiegongcheng/2484190.html
