叶片复杂曲面原位测量关键技术研究

发布时间：2018-08-04 19:44

【摘要】：叶片作为发动机中的关键零部件已经广泛应用于航天、航海、汽车等重要领域,它的设计水平与制造质量直接影响发动机的可靠性,甚至整机的性能。叶片型面为复杂自由曲面,加工制造难度大,现有加工设备和技术存在加工环节脱节、测量与加工分离等问题,难以实现叶片的高效率和高质量制造。针对上述问题,研究叶片复杂曲面制造关键技术,研发叶片原位测量系统,对提升叶片的制造水平和生产效率具有重要的实践应用价值。依靠自主研发的叶片制造样机,搭建叶片原位测量系统,在原位测量系统的硬件设计研发、控制软件开发、测量轨迹优化、测量数据点处理等方面开展一系列创新性的研究。所搭建原位测量系统的特色在于:在不改变叶片装夹位置的前提下,完成叶片复杂曲面表面数据信息的采集,减少加工定位误差。提出新的叶片复杂曲面测量方式和特征点提取技术,提高测量效率,实现叶片加工过程中高效率和高精度的辩证统一。基于所搭建原位测量系统开展叶片快速测量的相关实验研究。分析叶片复杂曲面的传统测量方法,根据叶片截面线具有很强相似性的特点,提出截面线测量轨迹优化方法。对叶片的顶部截面线和根部截面线进行精确识别,规划叶片顶部、中部和根部截面线位置;通过分析截面线几何敏感点和最大挠度点位置,规划叶片轴流方向测量线。将Hausdorff距离作为误差评判标准,不断删除轴流方向测量线上的测量点个数,直到每条测量线上具有相同的测量点个数,由此确定出所需测量的截面线条数。对所规划的叶片截面线测量轨迹进行实验验证,对其有效性进行评估。提出两种不同的特征点提取算法—B样条曲线逼近算法和切线多边形逼近算法。B样条曲线逼近算法选取了能够体现截面线整体轮廓的基本特征点,并根据改进的Hausdorff距离计算方法实现迭代,提高算法效率。切线多边形逼近算法根据不同曲线段包含不同曲线信息的概念,选取了可变误差阈值进行特征点提取,有效的保留曲线的细节特征。通过不同算法对比分析、数值算例仿真分析和相关测量实验研究,验证算法的有效性。研究基于特征点的叶片曲面重构方法,从保留叶片原始设计信息的角度出发,遵循由线到面的重构思路,实现叶片截面线分段拟合与拼接,提出特征点配对算法实现曲面参数网格重构。利用模型对比和斑马映射分别分析重构模型的误差分布、光顺性和连续性。通过算法对比、算例分析和实验分析检验算法的有效性和实用性。理论及实验研究结果表明:所搭建的原位测量系统性能稳定,能够实现叶片的原位测量。所优化的叶片截面线测量轨迹分布合理,能够良好、快速地实现叶片整体表面扫略测量。所提出的叶片截面线特征点提取算法—B样条曲线逼近算法和切线多边形逼近算法,具有良好的适用性。切线多边形逼近算法所提取特征点能够准确表示任何加工阶段的叶片截面线轮廓,满足叶片从毛坯到成品的加工余量计算以及加工质量检测。B样条曲线逼近算法结果更适用于对叶片成品合格率的检测,较少的特征点数目能够大大提高检测效率。基于特征点的B样条曲面重构算法,最大程度的保留了叶片的设计信息,其重构结果在曲面连续性、光顺性和逼近精度上都有良好表现。所提出的原位快速测量与曲面重构方法为实现叶片的高效、精确加工提供了新的技术参考和解决办法。
[Abstract]:As the key parts of the engine, blade has been widely used in the important fields such as space, navigation and automobile. Its design level and manufacturing quality directly affect the reliability of the engine, even the performance of the whole machine. The blade surface is a complex free-form surface, it is difficult to process and manufacture, and the existing processing equipment and technology are disjointed and measured. It is difficult to realize the high efficiency and high quality of the blade. In view of the above problems, the key technology of the manufacturing of the blade complex surface is studied. The research and development of the blade in situ measurement system is of important practical value to the level and efficiency of the blade. A series of innovative research on the hardware design and development of the in-situ measurement system, the development of the control software, the optimization of the trajectory and the processing of the data points are carried out. The characteristics of the built in situ measurement system are that the data of the surface of the complex surface of the blade is completed without changing the position of the blade clamping. Collecting and reducing the error of machining location. A new method of measuring the complex surface of blade and feature point extraction is put forward to improve the measurement efficiency and realize the dialectical unity of high efficiency and high precision in the process of blade processing. Based on the experimental research on the rapid measurement of the blade in the in-situ measurement system, the traditional measurement of the complex surface of the blade is analyzed. According to the strong similarity of the cross section line of the blade, the optimization method of cross section line measurement trajectory is proposed. The top section line and the section line of the root section are accurately identified and the section line position of the top, middle and root section of the blade is planned. The axial flow square of the blade is planned by analyzing the geometric sensitivity point of the section line and the position of the maximum deflection point. To the measuring line, the Hausdorff distance is used as the standard of error evaluation, and the number of measurement points on the axial direction measurement line is continuously deleted until the number of the same measuring points on each measuring line. Thus the number of cross section lines for the required measurement is determined. The measured trajectory of the section line of the blade is tested and its effectiveness is evaluated. Two different feature point extraction algorithms, the B spline curve approximation algorithm and the tangent polygon approximation algorithm.B spline curve approximation algorithm, are proposed to select the basic feature points that can reflect the overall profile of the cross section line, and the algorithm efficiency is improved by the improved Hausdorff distance calculation method. The tangent polygon approximation algorithm is the root of the algorithm. According to the concept of different curve segments including different curve information, the variable error threshold is selected to extract the feature points, and the details of the curves are preserved effectively. The effectiveness of the algorithm is verified by comparison and analysis of different algorithms, numerical example simulation analysis and correlation measurement experiments. The method of reconstructing the blade surface based on feature points is studied. In the view of preserving the original design information of the blade, following the line to surface reconstruction, the section line segmentation and splicing are realized, and the feature point matching algorithm is proposed to reconstruct the surface parameter grid. The error distribution, the smoothness and continuity of the reconstructed model are analyzed by the model contrast and the zebra mapping. The effectiveness and practicability of the algorithm are tested by the case analysis and experimental analysis. The theoretical and experimental results show that the built in situ measurement system is stable and can realize the in-situ measurement of the blade. The optimized blade cross-section line measurement trajectory is reasonable, and the blade overall surface sweeping measurement can be achieved well and quickly. The proposed blade is proposed. The feature point extraction algorithm of slice section line, B spline curve approximation algorithm and tangent polygon approximation algorithm, has good applicability. The feature points extracted from the tangent polygon approximation algorithm can accurately express the profile of the blade section line in any processing stage, and meet the machining allowance calculation of the blade from the blank to the finished product and the processing quality detection. The B spline curve approximation algorithm is more suitable for the test of the qualified rate of the finished blade. The number of fewer feature points can greatly improve the detection efficiency. The B spline surface reconstruction algorithm based on the feature points has retained the design information of the blade to the maximum extent, and its reconstruction results have good tables in surface continuity, smoothness and approximation accuracy. The proposed in-situ rapid measurement and surface reconstruction method provides a new technical reference and solution for high efficiency and accurate machining of blades.
【学位授予单位】：吉林大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TK403

【参考文献】