Optimization Method Of On-Machine Inspection Sampling Points Based On Surface Complexity

On-machine inspection (OMI) technology can improve the machining quality and detection efficiency of mechanical parts. The quantity of inspection points in the process of OMI is related to the inspection efficiency and data processing; it also affects the layout of inspection points. In this paper, the concept of surface complexity is proposed to replace the complex calculation of curvature, and the algorithm of surface complexity suitable for the mesh model is established to simplify the curvature calculation method in accordance with the changing trend of surface curvature. A multi-parameter radial basis function neural network including surface complexity is constructed, the advantage of the neural network in data processing is used to generate surface inspection points, and the changing trends of surface fitting accuracy under different numbers of inspection points are compared to verify the method in this paper. The experimental results show that the surface complexity can be used to quickly approximate the curvature of the surface and accords with the changing trend of the curvature. At the same time, the analysis results of the number of neural network inspection points meet the requirements of fitting accuracy.