人工神经元网络模型预测3D打印部件力学性能的研究  

Study of Mechanical Properties Prediction of 3D Printing Parts by Artificial Neural Network Model

作　　者：吕志敏[1] 江豪 LYU Zhimin;JIANG Hao(Department of Mechanic Electrical and Automatic Engineering,Puyang Vocational and Technical College,Puyang 457000,China;Department of Electrical Information Engineering,Zhengzhou University of Light Industry,Zhengzhou 450002,China)

机构地区：[1]濮阳职业技术学院机电与汽车工程学院,河南濮阳457000 [2]郑州轻工业大学电气信息工程学院,河南郑州450002

出　　处：《塑料工业》2024年第1期59-66,100,共9页China Plastics Industry

基　　金：河南省2020年重点研发与推广专项(科技公关)项目(212102210081)。

摘　　要：熔融沉积成型(FDM)是一种高效的增材制造技术。将响应面模型与人工神经元网络(ANN)模型相结合,研究了FDM工艺的喷嘴温度、层高和层积角度对尼龙12(PA12)丝材制造部件力学性能的影响。当喷嘴温度、层高和层积角度分别在220~260℃、0.2~0.4 mm、0°~90°之间变化时,部件拉伸强度和缺口冲击强度分别在35.69~60.89 MPa和5.48~19.83 kJ/m^(2)之间。喷嘴温度、层高、层积角度以及层积角度的二阶效应是影响部件拉伸强度的显著因素;喷嘴温度、层积角度以及层积角度的二阶效应是影响缺口冲击强度的显著因素。ANN模型预测拉伸强度和缺口冲击强度的最优结构分别是3-10-5-1和3-25-24-1,预测的拉伸强度和缺口冲击强度均方误差函数(MSE)最低分别为2.54×10^(-4)和2.07×10^(-4),回归系数均在0.97以上。与响应面的二次回归模型相比,ANN模型预测的拉伸强度和缺口冲击强度与实验值的标准偏差分别为0.46和0.32,远低于二次回归模型的2.43和1.58,更适合于优化非线性的FDM工艺。Fused deposition modeling(FDM)was an efficient additive manufacturing technology.The response surface model was combined with an artificial neural network model to study the effects of nozzle temperature,layer height,and deposition angle on the mechanical properties of nylon 12(PA12)parts produced by FDM process.When the nozzle temperature,layer height,and deposition angle change between 220-260℃,0.2-0.4 mm,and 0°-90°,the tensile strength and notch impact strength of the parts are varied between 35.69-60.89 MPa and 5.48-19.83 kJ/m^(2),respectively.The nozzle temperature,layer height,stacking angle,and second-order effect of deposition angle are the most significant factors affecting the tensile strength of parts.The nozzle temperature,deposition angle,and second-order effect of deposition angle are the most significant factors affecting the notch impact strength.The optimal structures for predicting tensile strength and notch impact strength using the ANN model are 3-10-5-1 and 3-25-24-1,respectively.The minimum MSE for predicting tensile strength and notch impact strength are 2.54×10^(-4)and 2.07×10^(-4),respectively,with regression coefficients above 0.97.Comparing with the response surface quadratic regression model,the standard deviations of the predicted tensile strength and notch impact strength by the ANN model and the experimental values are 0.46 and 0.32,respectively,which are much lower than the 2.43 and 1.58 of the quadratic regression model,making ANN more suitable for optimizing nonlinear FDM processes.

关 键 词：3D打印 熔融沉积成型 人工神经元网络 预测 力学性能 

分 类 号：TQ320.669[化学工程—合成树脂塑料工业]