Development and Applications of Extrusion Profile Prediction Model Based on BP Neural Network

doi:10.19491/j.issn.1001-9278.2019.10.010

Abstract

Abstract: In this work, the relationship between the process factors of a small-scale single-screw extruder for consumable filaments and the diameter of 3D printing consumable filaments was analyzed, and then a BP neural network-based prediction model was established by setting the die temperatures of three zones, barrel temperature, screw speed and tractor speed of single-screw extruder as input variables and the filament diameter as an output variable. The more accurate results could be obtained from this BP neural network prediction model, indicating a good prediction capability. Based on the network model developed by this work, the prediction software was designed and developed for application in the production line of consumable filaments, and this software could provide a high-efficient and rational guidance for determining the processing technology.

Key words: neural network, prediction model, single screw extrusion, extrusion process

CLC Number:

FANG Shengjie, BI Chao. Development and Applications of Extrusion Profile Prediction Model Based on BP Neural Network[J]. China Plastics, 2019, 33(10): 54-58.

References

[1] 徐学良. 人工神经网络的发展及现状[J]. 微电子学, 2017, 47(2): 239-242.
[1] Xu Xueliang. Development and Status of Artificial Neural Networks[J]. Microelectronics, 2017, 47(2): 239-242.
[2] X.kong, P.D.Hodgson. The application of constitute and artificial neural network models to predict the hot strength of steels [J], ISIJ International. 1999, 39(10):991-998
[3] Wu, Yamei, Ren, Yong. Prediction of mechanical properties of hot rolled strips by BP artificial neural network[C]. Proceedings-2011 International Conference of Information Technology, Computer Engineering and Management Sciences, ICM 2011, 2011, 1:15-17.
[4] 曾广胜, 孙刚. 工艺参数对淀粉/EVA生物质材料挤出发泡的影响及BP神经网络的预测[J]. 复合材料学报, 2014, 31(2):512-517.
[4] Zeng Guangsheng, Sun Gang. Effect of process parameters on extrusion foaming of starch/EVA biomass materials and prediction of BP neural network[J]. Journal of Composite Materials, 2014, 31(2): 512-517.
[5] 孙会波，闫宝瑞，信春玲，何亚东. 基于BP 神经网络的挤出温升预测[J]. 塑料, 2014,43(6):105-108.
[5] Sun Huibo, Yan Baorui, Xin Chunling, He Yadong. Prediction of Extrusion Temperature Rise Based on BP Neural Network[J]. Plastics, 2014,43(6):105-108.
[6] 张辉，曾广胜，陈华. 基于木塑专用同向双螺杆挤出机螺杆预测的神经网络模型及其应用[J]. 湖南工业大学学报, 2014, 28(5):50-54.
[6] Zhang Hui, Zeng Guangsheng, Chen Hua. Neural network model and its application based on screw prediction of co-rotating twin screw extruder for wood-plastic special purpose [J]. Journal of Hunan University of Technology, 2014, 28 (5): 50-54.
[7] 朱双东. 神经网络应用基础[M]．沈阳：东北大学出版社, 2000．
[7] Zhu Shuangdong. Application Basis of Neural Network [M]. Shenyang: Northeast University Press, 2000.
[8] YC Lin, DD Chen, MS Chen, XM Chen, J Li. A precise BP neural network-based online model predictive control strategy for die forging hydraulic press machine [J]. Neural Computing & Applications, 2018, 29(9):585-596.
[9] 刘振亮．基于BP 神经网络的机电设备温度监测预警管理系统研究[D]. 太原: 太原理工大学, 2012．
[9] Liu Zhenliang. Research on temperature monitoring and early warning management system of electromechanical equipment based on BP neural network [D]. Taiyuan: Taiyuan University of Technology, 2012.
[10] NV Bhat, TJ McAvoy. Determining model structure for neural models by network stripping[J]. Computers＆Chemical Engineering，1992, 16(4) : 271-281．
[11] Jing Li, Ji-hang Cheng, Jing-yuan Shi, and Fei Huang. Brief Introduction of Back Propagation (BP) Neural Network Algorithm and Its Improvement[J]. Advances in Computer Science and Information Engineering, 2012, 169:553-558.
[12] 于国庆. 3D打印耗材生产装备挤出牵引部分的联动模型及生产过程远程监控[D]. 北京化工大学, 2018.
[12] On National Day. The linkage model of extrusion traction part of 3D printing consumables production equipment and remote monitoring of production process [D]. Beijing University of Chemical Technology, 2018.
[13] 韩力群．人工神经网络理论、设计及应用[M], 北京：化学出版社, 2007.
[13] Han Liqun. Theory, Design and Application of Artificial Neural Networks [M], Beijing: Chemical Press, 2007.