Research On Intelligent Planning Method Of Laser Selective Melting Process Strategy Based On Feature Recognition

Selective Laser Melting(SLM)technology as one of the representative additive manufacturing technologies,has been widely concerned and developed rapidly in recent years,showing great application potential in aviation,aerospace,biomedical and other industries.However,in the existing SLM forming process,the forming layer of parts adopts a single and uniform general craft strategy,which is difficult to take into account the influence of structural features on the forming quality,to balance the stress distribution in the processing process,and to improve the forming quality more effectively.Therefore,it is necessary to realize the intelligent allocation of craft strategy according to the features of parts.This article mainly studies the craft data processing method of SLM based on feature recognition.By solving the key craft data processing problems,such as efficient and highprecision slicing contour calculation of large-scale complex model,scanning path hierarchical and regional planning,significant structure feature extraction,a multi-scale and multi-level craft strategy based on feature constraint is constructed to realize intelligent planning of craft strategy based on feature recognition.The main achievements are as follows:Aiming at the problem of low efficiency of slicing contour calculation of large-scale complex models,a parallel computing method of slicing contour calculation based on the copoint “welding” strategy was proposed.By constructing index information of STL model to assign label numbers to the endpoints of line segments,and based on the strategy of co-point "welding" combined with parallel computing mode,the calculation links such as the calculation of line segments,the construction of contour and the optimization of contour data are completed,and the fast calculation of the slicing contour is realized.Aiming at the limitation of hard to select and allocate the scanning strategy of the part entity area intelligently,a scanning strategy based on the geometric information discrimination of the connected entity area is proposed.Taking the projection span values in the direction of principal axis and secondary axis and area ratio of connected entity area as geometric information,the scanning strategy sucn as island scanning and parallel scanning can assign adaptively.This scanning strategy can reduce the number of scanning lines with extreme length and effectively balance the length of scanning lines.In view of the fact that the existing SLM forming process adopts a single and uniform general craft strategy,without fully considering the limitations of the structural features on the forming quality,a craft strategy based on structural feature recognition is proposed.Through the combination of 3D model salient feature extraction and 2D slicing contour area segmentation,the intelligent recognition of overhanging and thin-wall features is completed.The slicing contour is divided into different laser processing areas.The multi-scale and multilevel process strategy planning method based on feature constraints is established,and the regional planning of craft strategy based on feature recognition is realized.