Study of Mechanical Properties Prediction of 3D Printing Parts by Artificial Neural Network Model
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/m2,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.
Three Dimensional PrintingFused Deposition ModelingArtificial Neural NetworkPredictionMechanical Properties
万方数据
维普
