叶片轮廓快速扫描系统开发与研究

发布时间：2018-03-12 18:30

本文选题：叶片检测　切入点：线激光扫描测量　出处：《天津大学》2016年硕士论文　论文类型：学位论文

【摘要】：目前航空发动机叶片轮廓检测应用最多的是采用三坐标测量机检测,随着计算机技术的发展,三坐标测量机在叶片上采集特定基准点,通过上位机软件即可获得叶片轮廓的点云,且具有较高的精度。但由于叶片的数量大,单个叶片完成检测需要5~10分钟,使得完成检测的周期过长,加大了叶片生产制造成本。针对以上缺点,有人提出可用三坐标测量机先对标准叶片进行检测,使用非接触的三维视觉测量方法对其余叶片进行检测,将两种方法获取的点云进行比较,比较后的误差小于一定阈值即可满足要求,由于线激光结构光测量系统能进行扫描测量,测量速度快,便于批量流水操作,在这方面的应用与开发越来越必要。本课题是天津大学精密仪器与光电子工程学院与海克斯康(青岛)测量技术有限公司的合作项目,课题借助了海克斯康公司强大的硬件平台,依托了天津大学在该领域的技术优势,具有较高的工程应用价值和指导意义。本文的主要工作和如下:(1)由于单个线激光结构光扫描测量系统测量范围有限,因此再增加一个测量系统形成多传感器扫描测量系统,可有效的避免测量盲区,提高测量效率。(2)建立线激光结构光扫描测量系统数学模型,结合系统的性能指标和总体结构,完成硬件系统与软件系统的开发工作。(3)深入研究多传感测量系统全局标定问题,结合实际实验情况,针对系统标定误差主要集中在坐标轴垂直度误差上,推导垂直度误差补偿数学模型,建立坐标系转换矩阵,通过补偿测量空间整体示值误差的方法,完成全局标定工作。(4)系统全局标定完成后,由于测量数据为点云数据,研究了在逆向工程中点云拼接与精简方法,利用Geomagic软件完成点云数据处理,通过与标准叶片CAD数模比较验证系统精度,验证后系统满足性能参数要求。
[Abstract]:At present, the most popular application of aero-engine blade profile detection is the use of CMM. With the development of computer technology, CMM collects specific reference points on the blade. The point cloud of blade contour can be obtained by the software of upper computer, and it has high precision. However, because of the large number of blades, it takes 5 minutes and 10 minutes to complete the detection of a single blade, which makes the detection period too long. In view of the above disadvantages, it is suggested that the standard blade can be detected by the CMM first, and the remaining blades be detected by the non-contact 3D visual measurement method. Comparing the point clouds obtained by the two methods, the error after comparison can be satisfied when the error is less than a certain threshold. Because the line laser structured light measurement system can carry out scanning measurement, the measurement speed is fast, and it is convenient for income to operate in batches. This project is a joint project between the School of Precision instrument and Optoelectronics Engineering of Tianjin University and Hexkang (Qingdao) surveying Technology Co., Ltd. With the help of Hexcom's powerful hardware platform, the project relies on the technological advantages of Tianjin University in this field. The main work of this paper and the following are as follows: (1) since the measurement range of single line laser structured light scanning measurement system is limited, so an additional measurement system is added to form a multi-sensor scanning and measuring system. It can effectively avoid the blind area of measurement and improve the measuring efficiency. (2) the mathematical model of the linear laser structured light scanning measurement system is established, which combines the performance index and the overall structure of the system. Finish the development of hardware system and software system. (3) deeply study the global calibration problem of multi-sensor measurement system. According to the actual experiment, the calibration error of the system is mainly focused on the coordinate axis verticality error. The mathematical model of verticality error compensation is derived, and the coordinate transformation matrix is established. By compensating the error of the whole indication value in the measurement space, the global calibration of the system is completed. After the global calibration of the system is completed, because the measured data are point cloud data, The method of point cloud splicing and reduction in reverse engineering is studied. The point cloud data processing is completed by using Geomagic software. The accuracy of the system is verified by comparing it with standard blade CAD. The system meets the requirements of performance parameters after verification.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP391.41

【相似文献】