Answers are grounded in this paper and cite their sources.
Blade repair feature detection can determine the repair status of blades and is a prerequisite for the subsequent planning of machining parameters and the guarantee of blade repair accuracy. To address the difficulty of guaranteeing cladding-layer feature recognition accuracy caused by the irregular shape and uneven residual height of the cladding layer on repaired blade surfaces, this study establishes an online identification method for complex blade repair features based on robotic line-laser measurement technology. First, a 3D line laser measurement system was designed that integrates the motion of a rotary table with that of a robot, thereby expanding the system’s application range; accuracy calibration was then performed on key components such as the rotary table. Next, the blade point cloud data was filtered and simplified; specifically, a two-stage (coarse-to-fine) stitching method for multi-view point clouds was proposed by combining the mechanical rotary table with the ICP algorithm, yielding complete and accurate point cloud data of all blade features; Finally, for the blade cladding layer, normal reorientation and feature enhancement based on the minimum spanning tree were performed, and a feature recognition method for the blade cladding layer—based on regional growth factors and the rotating minimum bounding box—was established according to the blade’s regional division. When compared with coordinate measuring machine (CMM) results, the maximum error of the blade cladding layer in each region using the method described in this paper was less than 0.1 mm, demonstrating high accuracy and providing precise initial state input for subsequent machining of the blade cladding layer regions.
No takes yet. Share an insight, caveat, or question.
Qi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: