Research On The Key Technology Of In-situ Measurement For Blade Complex Surface

Blades as the key components of engine are widely used in aerospace, automotive, vessel and other fields. The quality of blade design and manufacture directly affects the performance of engine and even the whole machine. Compared with the other work-pieces, the blade is more difficult to process because of its complex surface. There are drawbacks of disjoints associated with the current process equipment and technology in processing, and the separation of measurement and manufacture, which make it difficult to realize the high efficiency and high precision of blade manufacturing. Aiming at these problems, the key manufacturing technology of blade surface, the development of blade in-situ measurement system are being studied to improve the manufacturing quality and the production efficiency of blade, which have an important engineering significance.The in-situ measurement system is developed based on the built manufacturing prototype of blade surface. The design of hardware, the development of control soft ware, the optimization of measuring trajectory, the processing of measured points are subjected to a series of innovative research. The features of the built in-situ measurement system are that the collection of blade complex surface data information need not to change the clamping position which reduces positioning error and the proposed measurement mode and measured points compression technology of blade complex surface which improves the measurement efficiency. The features further achieve the dialectical unity of high efficiency and high precision of blade during processing. A series of rapid measurement experiments are carried out based on the built in-situ measurement system.According to analysis the traditional measurement methods of blade complex surface, a measurement trajectory optimization strategy of blade section curve is proposed based on the feature of strong similarities between the section curves. The tip, middle and hub section curves of blade are planned based on the accurate identification of tip and hub section curves. Then according to the section curves geometry sensitive points and the maximum deflection points position, the measurement trajectory of blade axial direction is planned. The Hausdorff distance is taken as an error criterion, and the measured points on measurement trajectory of blade axial direction are constantly removed until each trajectory has the same number of measured points, which determines the number of section curves needed in blade measuring. The optimized measurement trajectory is experimentally verified and its effectiveness is evaluated.Two different feature points extraction algorithms, B spline approximation algorithm and tangent polygon approximation algorithm, are presented to extract the feature points. B-spline approximation algorithm selects the basic feature points that can reflect the overall profile of section curve, and an iterative algorithm based on the improved Hausdorff distance is used to improve the efficiency of the algorithm. Tangent approximation algorithm based on the concept that different curve segments represent different information of the curve, selects variable error threshold to extract feature point, which effectively preserves the detail features of the section curve. The effectiveness of the algorithm is proved by the comparison with other algorithms, numerical examples and experiments.The blade surface reconstruction method based on feature points is proposed that follows a reconstruction idea of curve to surface and a concept of retaining the original design information. The algorithm achieves the fitting and stitching of blade section curve segments and surface parameter mesh reconstruction by using feature point matching strategy. The error distribution, smoothness and curvature continuityof reconstruction surface are analyzed by the model comparison and Zebra mapping respectively. The result of comparison, numerical examples and experiments show the effectiveness and practicality of the algorithm.The theoretical and experimental results show that: the built prototype has a stable performance and is able to achieve the in-situ measurement; the optimization measurement trajectory has a reasonable distribution and measures blade surface well and quickly; the B-spline approximation algorithm and tangent polygonal approximation algorithm have good applicability; the feature points obtained by tangent polygonal approximation algorithm adapt to any stage of processing, and satisfy the requirement of allowance calculation from the rough to the finished machining and the inspection of surface quality. B-spline curve approximation algorithm result is more suitable for the detection of finished blade Pass-Rate, the less number of feature points can greatly improve the detection efficiency; B-spline surface reconstruction algorithm based on the feature point preserves maximal design information and has a good performance in continuity, smoothness and approximation accuracy. The built in-situ measurement technology and fast surface reconstruction theory provide a new technical reference and solution for achieving high efficiency and accurate processing of blades.