Laser cladding path planning for gear repair based on area division and trajectory correction

In addressing the issues of laser defocusing, laser beam interference, and low melting efficiency prevalent in current gear laser cladding path planning, a method for gear laser cladding path planning is formulated based on area segmentation and trajectory refinement. Firstly, the tooth surface model is reconstructed using three times NURBS surfaces. Subsequently, the tooth failure region is extracted through point cloud data alignment and Boolean operations, and the laser scanning region is preliminarily delineated using a graphical convolutional neural network. This is further refined by employing an ant colony algorithm. Secondly, by employing a geometrically constrained mathematical model of the gear, the compensation distance for laser focusing and the feasible domain range of the laser beam are determined to effectuate the trajectory refinement for the gear's laser cladding. Finally, completing the laser scanning area division and trajectory correction to perform the laser cladding gear repair experiment, the experiment relies on the HLC40 laser powder feeding additive manufacturing workstation, adopting the YLR-4000IPG laser for cladding operation. The experimental results demonstrate the absence of focus offset and laser beam interference during the cladding process. Moreover, the total travel distance of the planned path was reduced by 9-12%, the cladding time was reduced by 8-16%, and the morphological quality of the cladding layer was improved by 39-46%.