基于神经网络的快速成型工艺

针对选择性激光烧结成型件变形大、精度较低的问题,将神经网络方法应用于选择性激光烧结(SLS)加工工艺的研究.根据SLS加工工艺的特点,研究的工艺参数包括:层厚、扫描间距、激光功率、扫描速度、环境温度、层与层之间的加工时间间隔和扫描方式.建立了SLS加工工艺参数与加工变形、收缩率之间的神经网络预测模型.实验结果与神经网络模型计算结果十分吻合,说明该神经网络模型能定量地反映出工艺参数与加工材料变形、收缩率之间的关系.

Rapid prototyping process by neural network

To solve the problem of large deforming and poor accuracy of SLS product ,the method of the artificial neural network is applied in the study of selective laser sintering (SLS) process. According to the feature of SLS, the parameters of interest are layer thickness, hatch spacing, laser power, scan speed, work surroundings temperature, interval time and scanning mode. The neural network model on the relationship between the processing parameter and distortion, shrinkage ratio of the job was built. The calculation results of neural network model are in good agreement with the experimental results, which indicating that the neural network model can analysis the relationship quantitatively.