基于支持向量回归的熔丝制造3D打印能效优化模型

An Energy Efficiency Optimization Model of Fused Filament Fabrication 3D Printing Based on Support Vector Regression

基于响应面法和支持向量回归模型对熔丝制造3D打印能效进行预测与优化。首先,利用田口方法设计六因素三水平正交试验,通过响应面法分析得出对加工能效影响较为显著的3个因素即层高、打印速度和热床温度;然后,通过支持向量回归方法建立加工能效预测模型,并与BP神经网络方法进行对比,结果表明支持向量回归方法建模预测性能更优;最后,建立以加工时间和能效为目标的优化模型,利用NSGA-Ⅱ、MOEA/D、SPEA2和MOPSO 4种算法分别对模型进行求解,分析比较4种算法的Pareto前沿,结果表明NSGA-Ⅱ算法在求解此问题时综合表现最佳,对比NGSA-Ⅱ算法求得的优化结果与试验结果可知,NSGA-Ⅱ算法具有有效性和合理性。

The fused filament fabrication 3D printing energy efficiency was predicted and optimized based on response surface method and support vector regression model.Firstly,the Taguchi method was used to design the six factor three level orthogonal test,based on the response surface method,three factors that had a significant impact on processing energy efficiency were obtained,namely layer height,printing speed and hot bed temperature.Then,the prediction model of processing energy efficiency was established by support vector regression method,and compared with BP neural network method,the results show that the modeling and prediction performance of support vector regression method is better.Finally,the optimization model aiming at processing time and energy efficiency was established,and NSGA-Ⅱ,MOEA/D,SPEA2 and MOPSO were used to solve the model respectively,the Pareto front of the four algorithms was analyzed and compared,the results show that NSGA-Ⅱ performs best in solving this problem,the optimization results obtained by NSGA-Ⅱ algorithm were compared with the experimental results,which reflects the effectiveness and rationality of the optimization results of NSGA-Ⅱ algorithm.