燃机叶片型面三维光学扫描检测方法研究.pdfVIP

燃机叶片型面三维光学扫描检测方法研究 陈俊宇, 殷 鸣, 殷国富, 向召伟, 邓珍波 （四川大学制造科学与工程学院，四川 成都 610065） 摘 要：针对工业生产中提高燃气轮机叶片型面检测效率、实现叶片整体尺寸检测的实际需求,提出基于激光和光栅 两种三维光学扫描的叶片型面检测方法。 采集燃气轮机叶片三维点云数据,用统一采样和设定点云间最大偏差值的 方法进行点云精简。 根据 6点定位原理和最小二乘法将精简点云与设计模型进行配准,提取叶片型面轮廓度和检测 截面特征参数进行分析,并与三坐标检测结果对比。 结果表明：两者检测叶片型面轮廓度在相同偏差范围内点云所占 比例偏差都在 1%以内,叶片型面特征参数尺寸相差都在 0.02mm以内。 经试验证明,三维光学扫描方法检测速度快、 适应性强、可靠性高,可为高效准确地进行叶片型面检测提供有效的技术手段。 关键词：三维光学扫描；燃气轮机叶片；叶片型面检测；点云精简；检测截面 文献标志码：A 文章编号：1674-5124(2015)11-0007-05 Research of 3D optical scanning inspection method of gas turbine blade surface CHEN Junyu，YIN Ming，YIN Guofu，XIANG Zhaowei，DENG Zhenbo （School of Manufacturing Science and Engineering，Sichuan University，Chengdu 610065，China） Abstract: The paper has proposed a novel approach to examine the blade surface of steam turbines by laser scanning and raster scanning so as to improve the inspection efficiency and achieve 3D information detection in industrial production. The point cloud data of steam blade were collected and simplified through uniform sampling and by setting the maximum deviation for point cloud. The simplified point cloud is aligned with a design model in accordance with the six- point orientation principle and least square method to extract and analyze the profile of blade surface and characteristic parameters of detection section and then compare them with three - dimensional testing results. The results indicate that the proportions of point cloud differ by no more than 1% at the same deviation range in profile of blade surface， and the characteristic parameters of blade surface differ by no more than 0.02 mm in dimension. The experimental results demonstrate that the proposed method is fast and adaptable，reliable and accurate，it can be used to provide efficient method for blade surface inspection. Keywords: 3D optical scanning； steam turbine blade； blade surface inspection； point cloud simplification；detection section 收稿日期：2015-05-13；收到修改稿日期：2015-06-20 基金项目：四川省科技支撑计划项目(2015GZ0012, 2015GZ0114) 作者简介：陈俊宇(1990-),男,四川