利用面片法向量保留模型特征的3D打印自适应分层算法

An adaptive slicing algorithm based on the normal vector of triangular facet in the STL file is proposed to solve the problem that traditional adaptive slicing algorithms are complicated to calculate and cannot effectively retain the model features. The algprithm can reduce the step effect, and recognize and preserve the model features. Firstly, the component of the facet normal vector in the delamination direction is calculated, and the larger component is identified as the place where the inclination angle of the model surface is small. Then a small layer thickness is used in this place to reduce the step effect. A method to recognize model features is proposed, and the normal vector of the facet is used to calculate the dihedral angle between the facets and the complexity index of the vertices, so as to identify the sharp edge and complex point features of the model surface. A smaller layer thickness is used near the features to retain the model features, while larger layer thicknesses are used elsewhere to reduce printing time. Finally, missing triangular facets are detected and the layered thickness is corrected to prevent the loss of model features. Experimental results show that for a dumbbell model, the proposed algorithm can identify the non-local highest or lowest feature points, which cann't be identified by the traditional algorithm; and for chinchilla model and cone model, the proposed algorithm can identify all model features, and the layer thickness can be adjusted according to the complexity of the model features. On the premise that the tip height is 0.400 mm and 0.355 mm respectively, the number of layers reduces by 2.5% and 1.2% respectively, compared with the traditional algorithm.