To realize high-precision machining of free-form blade surfaces, it is necessary to measure and reconstruct the blade morphology several times. Sampling measurements and data compression through free-curve approximation technology are of great benefit in shortening the processing time. A novel B-spline approximation method is proposed in this paper, which can determine a few dominant points based on curvature characteristics to approximate dense and noisy measured points with high precision. We adopt a unique adaptive knot placement strategy combined with a curvature dependent evaluation parameter and a conformal transformation method. This strategy is error-bounded in the initial knot placement, which greatly reduces the time consumed by subsequent iterative insertions. The proposed method shows advantages in approximation accuracy, compression rate, and especially in computational efficiency, in comparison with four traditional knot placement methods. Experiments were carried out and the results show that the dominant point selected by this method can replace the fine measurement results, and the compression rate of the measurement points can reach more than 90%. This method is suitable for the sampling measurement and compression of the data of blades, aircraft wings, and other similar free-form surfaces, and is especially suitable for solving the inefficiency problem caused by repeated measurements made in the process of iterative machining of aero-engine blades.

为了实现自由曲面叶片表面的高精度加工，需要多次测量和重建叶片形态。通过自由曲线逼近技术进行采样测量和数据压缩对于缩短处理时间有很大好处。小说B本文提出了-样条逼近方法，该方法可以根据曲率特征确定少数优势点，以高精度逼近密集和嘈杂的测量点。我们采用独特的自适应结放置策略，结合曲率相关评估参数和保形变换方法。这种策略在初始结点放置中是有错误限制的，这大大减少了后续迭代插入所消耗的时间。与四种传统的节点放置方法相比，所提出的方法在近似精度、压缩率，尤其是计算效率方面显示出优势。进行了实验，结果表明该方法选取的优势点可以替代精细的测量结果，测点压缩率可达90%以上。该方法适用于对叶片、飞机机翼等类似自由曲面的数据进行采样测量和压缩，特别适用于解决航空零件在迭代加工过程中重复测量造成的效率低下问题。发动机叶片。