An adaptive sampling methodology for measuring blades with CMM based on dominant feature points

Abstract

Sampling strategy is one of the most critical components of the contact probing process with coordinate measuring machines (CMMs). The number and location of sampling points have direct impacts on the efficiency and reliability of the measurements. Blades of aviation engines have complex geometry and extremely high precision requirements, but the sampling strategy of blades is still a difficult problem which limits the use of CMMs in blade inspection. To balance the measuring efficiency and accuracy of CMM, an adaptive sampling algorithm based on dominant feature points is proposed. In this algorithm, the points are adaptively sampled through a modified B-spline least-square fitting process with dominant feature points and new dominant points if necessary, and the new dominant points are computed by a new method, namely weighted curvature mass moment method. The sections of the blade to be measured are determined with this adaptive sampling algorithm to be the leading and trailing edge curves, and the points on each section to be measured are determined using dominant feature points of the blade profile. The definition and extraction methods for dominant feature points of the blade profile are also researched. By comparing the reconstruction errors of sampling points for several different blades, this adaptive sampling algorithm is proved to be more efficient and accurate than the equal parameter sampling method, equal arc length sampling method, and chord deviation sampling method.

Smart Coordinate Measuring Machine (CMM) Inspection for Turbine Guide Vanes with Trend Line and Geometric Profile Tolerance

“…Knot points are the most important and informative points and after simplification they indicate crucial shape changes. The original CAD section contains 104 smaller Bezier curves (Figure 7a), which is more than needed, as proved by other authors [15,[17][18][19][20][21][22][23][24][25][26]. It was rebuilt using knot removal methods and it was concluded that for CMM measurement the number of control points can be slightly decreased, reducing it to 20-30.…”

“…Yun Zhang [20] developed a sampling method for twisted blades to improve measurement efficiency. A very similar method was developed by Feiyan Guo [21] for a turbine guide vane. Xuezhe Li [22] focused his attention on the twist itself, to solve high-precision measurement problems with this phenomenon.…”

Adaptive sampling strategy of five-axis inspection for free-form surface reconstruction

“…The special scanning pattern of five-axis sweep scanning requires the sampling path to be a sinusoidal curve [12]. The traditional sampling strategies in three-axis CMM inspection cannot fully manifest the dynamic performance of the five-axis sweep scanning [13][14][15][16]. Thus, a suitable sampling strategy for five-axis CMM inspection should be investigated.…”

Optimization method of on-machine inspection sampling points based on surface complexity

“…Guo et al proposed an adaptive detection algorithm based on an advantage feature point. The algorithm adaptively detects integral points by modifying the B-spline least-squares fitting process, determining the profile curve of the measured surface by this method and using the main feature points of the profile to determine points in the measuring area [20]. Zhao et al proposed a new detection method for free-form surface inspection and inspection points, by dividing the surface to be measured into several sections that make it easy to construct B-splines, and obtain the data points needed to construct B-splines as inspection points [21].…”

“…After that, the clustering center of each cluster Assume that {x i } N i=1 represents a sample set of size N. Based on the K-means method we divide the sample set into K clusters (K < N). The mapping is shown as equation (20), which divides the ith sample x i into k clusters:…”