基于三维激光扫描技术的复杂构件检测
发布时间：2019-05-05 09:20
【摘要】：在铁塔的一般生产过程中,为了保证现场顺利立塔,出厂前必须对铁塔进行“试组装”,除了通过试组装检验构件加工正确与否之外,更重要的是验证在实际装配过程中构件之间是否会出现干涉和孔位错误等现象,即不能正确组装的情况。传统的立式组装和卧式组装都有其局限性,费用高、周期长,很难适应如今大规模的检测任务。本文针对铁塔生产中的复杂构件开展产品几何质量检测理论和三维激光扫描技术应用于其检测的方法研究,从而取代铁塔生产中的“试组装”。本文主要进行了以下几方面工作： 1.针对具体构件,利用Pro/E软件对其图纸建立三维数字化模型作为检测的基准,并且制定检测流程。 2.研究三维坐标获取方法。结合本课题的需要,综合考虑仪器性能参数和数据质量,选择激光三角测量原理的扫描仪,通过分析采集数据的精度,说明了将测量数据用于建模和检测的可行性。 3.目前,三维检测方法只能检测轮廓,不能进行螺栓孔定位,本文提出了螺栓孔的空间位置检测方法。主要步骤包括提取螺栓孔的点云,点云数据预处理,非线性的最小二乘方法拟合螺栓孔中心,最后将测量坐标系中的螺栓孔中心坐标通过坐标系转换到模型坐标系中,计算对应位置的螺栓孔中心误差。坐标系转换时,结合公共点转换方法和ICP算法,先粗对齐然后精确对齐的方法求解坐标系转换矩阵,减小坐标系转换的误差,提高检测精度。通过实验可知需要检测的16个螺栓孔的空间位置误差在1mm左右最大误差不超过2mm,且波动不大,达到了较高的检测精度。 4.运用三维检测方法检测构件外形轮廓,并且给出了检测报告,直观的显示出外形轮廓的误差,但是检测结果给出的误差范围较大,对于物体表面的孔洞检测效果不理想,主要表现为孔洞周围数据的三维检测误差变化较大无法反映出孔洞的位置误差,因此,进一步运用本文提出的螺栓孔空间位置检测方法检测其空间位置误差。
[Abstract]:In the general production process of the tower, in order to ensure the smooth erection of the tower on the spot, it is necessary to "test assemble" the tower before leaving the factory. In addition to testing the correctness of the processing of the components through the trial assembly, More importantly, it is important to verify whether interference and hole position errors will occur between the components in the actual assembly process, that is, if the assembly cannot be properly assembled. Traditional vertical assembly and horizontal assembly have their limitations, high cost, long cycle, it is difficult to adapt to today's large-scale detection tasks. In this paper, the theory of product geometry quality inspection and the method of 3D laser scanning applied to the complex components in the production of iron tower are studied in order to replace the "trial assembly" in the production of iron tower. The main contents of this paper are as follows: 1. Aiming at the concrete component, the Pro/E software is used to establish the three-dimensional digital model of the drawing as the benchmark of the detection, and the detection process is established. 2. The method of obtaining 3D coordinates is studied. According to the need of this subject, considering the performance parameters and data quality of the instrument, the scanner with laser triangulation principle is selected. By analyzing the accuracy of the collected data, the feasibility of applying the measurement data to modeling and detection is explained. 3. At present, the three-dimensional detection method can only detect the contour, but can not locate the bolt hole. In this paper, the space position detection method of bolt hole is put forward. The main steps include extracting the point cloud of bolt hole, preprocessing the data of point cloud, fitting the center of bolt hole with nonlinear least square method, and finally converting the center coordinate of bolt hole in the measuring coordinate system into the model coordinate system through the coordinate system. The central error of bolt hole in the corresponding position is calculated. In order to reduce the error of coordinate transformation and improve the detection accuracy, the common point transformation method and ICP algorithm are combined to solve the coordinate transformation matrix by coarse alignment and then accurate alignment. The experimental results show that the spatial position error of the 16 bolt holes need to be detected is less than 2 mm in 1mm, and the fluctuation is not big, so the precision of detection is higher than that of the other 16 bolt holes. 4. The three-dimensional detection method is used to detect the contour of the component, and the detection report is given, which shows the error of the contour intuitively. However, the error range given by the detection result is large, and the detection effect for the holes on the surface of the object is not ideal. The main performance is that the three-dimensional measurement error of the data around the hole can not reflect the position error of the hole. Therefore, the space position error of the bolt hole is detected by the space position detection method of the bolt hole proposed in this paper.
【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P225.2
[Abstract]:In the general production process of the tower, in order to ensure the smooth erection of the tower on the spot, it is necessary to "test assemble" the tower before leaving the factory. In addition to testing the correctness of the processing of the components through the trial assembly, More importantly, it is important to verify whether interference and hole position errors will occur between the components in the actual assembly process, that is, if the assembly cannot be properly assembled. Traditional vertical assembly and horizontal assembly have their limitations, high cost, long cycle, it is difficult to adapt to today's large-scale detection tasks. In this paper, the theory of product geometry quality inspection and the method of 3D laser scanning applied to the complex components in the production of iron tower are studied in order to replace the "trial assembly" in the production of iron tower. The main contents of this paper are as follows: 1. Aiming at the concrete component, the Pro/E software is used to establish the three-dimensional digital model of the drawing as the benchmark of the detection, and the detection process is established. 2. The method of obtaining 3D coordinates is studied. According to the need of this subject, considering the performance parameters and data quality of the instrument, the scanner with laser triangulation principle is selected. By analyzing the accuracy of the collected data, the feasibility of applying the measurement data to modeling and detection is explained. 3. At present, the three-dimensional detection method can only detect the contour, but can not locate the bolt hole. In this paper, the space position detection method of bolt hole is put forward. The main steps include extracting the point cloud of bolt hole, preprocessing the data of point cloud, fitting the center of bolt hole with nonlinear least square method, and finally converting the center coordinate of bolt hole in the measuring coordinate system into the model coordinate system through the coordinate system. The central error of bolt hole in the corresponding position is calculated. In order to reduce the error of coordinate transformation and improve the detection accuracy, the common point transformation method and ICP algorithm are combined to solve the coordinate transformation matrix by coarse alignment and then accurate alignment. The experimental results show that the spatial position error of the 16 bolt holes need to be detected is less than 2 mm in 1mm, and the fluctuation is not big, so the precision of detection is higher than that of the other 16 bolt holes. 4. The three-dimensional detection method is used to detect the contour of the component, and the detection report is given, which shows the error of the contour intuitively. However, the error range given by the detection result is large, and the detection effect for the holes on the surface of the object is not ideal. The main performance is that the three-dimensional measurement error of the data around the hole can not reflect the position error of the hole. Therefore, the space position error of the bolt hole is detected by the space position detection method of the bolt hole proposed in this paper.
【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P225.2
【参考文献】
相关期刊论文 前10条
1 李广云;工业测量系统最新进展及应用[J];测绘工程;2001年02期
2 杨长强;叶泽田;;一种基于点对应的激光扫描仪外方位参数检校方法[J];测绘科学;2011年05期
3 岳建平;魏叶青;张永超;;船舶建造测量分析系统研究[J];测绘通报;2008年10期
4 S∮窭，
本文编号：2469451
本文链接：https://www.wllwen.com/kejilunwen/dizhicehuilunwen/2469451.html
