基于神经网络遗传算法的CFRP拉挤工艺优化

陈幸开; 谢怀勤

基于神经网络遗传算法的CFRP拉挤工艺优化

哈尔滨工业大学材料科学与工程学院, 哈尔滨150001

基金项目: 黑龙江省自然科学基金资助项目(E01-10)

详细信息

作者简介:
陈幸开(1979-),男,广东深圳人,博士生,chenxingkai@126.com.

中图分类号: TB 332
计量
- 文章访问数: 3261
- HTML全文浏览量: 109
- PDF下载量: 1045
- 被引次数: 0
出版历程
- 收稿日期: 2008-08-10
- 网络出版日期: 2009-10-31

Optimization for CFRP pultrusion process based on genetic algorithm-neural network

School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, China

摘要

摘要: 根据固化动力学和传热学理论,建立了碳纤维增强聚合物基复合材料(CFRP,Carbon Fiber Reinforced Polymer)拉挤成型非稳态温度场与固化动力学数学模型.采用有限元与有限差分相结合的方法,结合ANSYS求解耦合场的间接耦合法,对CFRP拉挤过程非稳态温度场和固化度进行数值模拟.使用特殊设计制作的铝毛细管封装的布拉格光栅光纤(FBG,Fiber Bragg Gratings)传感器,对温度场进行实时检测;并采用索氏萃取实验测定CFRP制品固化度.模拟与实验结果基本吻合.以数值模拟结果为样本建立反向传播神经网络,训练得到固化炉温度与CFRP固化度之间的非线性相关关系.采用神经网络与遗传算法相结合的方法,优化得出拉挤固化炉三段最佳温度值,结果表明神经网络结合遗传算法优化拉挤工艺参数快捷有效.
- 碳纤维增强聚合物基复合材料 /
- 拉挤 /
- 数值模拟 /
- 布拉格光栅光纤 /
- 神经网络 /
- 遗传算法 /
- 优化
Abstract: According to the curing kinetics and heat transfer theory, the models of unsteady temperature field and resin curing for the pultrusion process of carbon fiber reinforced polymer (CFRP) were established. The finite element/finite different method associated with the indirect decoupling method based on ANSYS to simulate the temperature and degree of cure for CFRP pultrusion. The fiber Bragg gratings (FBG) sensor, encapsulated in an aluminum capillary, was utilized to real time monitor the temperature during pultrusion; and the final degree of cure was measured by Sorbitic extraction. The results show that the numerical model is reliable and correct. With the simulative results, the neural network was trained and predicted the degree of cure effectively. On the basis of the trained network, the genetic algorithm was used to optimize the temperature of die. It shows that the optimization method is effective and convenient.
- carbon fiber reinforced polymer /
- pultrusion /
- simulation /
- fiber Bragg gratings /
- neural network /
- genetic algorithm /
- optimization

HTML全文

参考文献(1)

[1] Jacob A. Globalisation of the pultrusion industry[J]. Reinforced Plastics,2006,50(5):38-41  [2] Martin J. Pultruded composites compete with traditional construction materials[J]. Reinforced Plastics,2006,50(5):20-27 [3] Stewart R. Pultrusion industry grows steadily in US[J]. Reinforced Plastics,2002,46(6):36-39 [4] 刘红梅,王少萍,欧阳平超.基于GRNN网络和遗传算法的旋翼动平衡调整[J].北京航空航天大学学报,2008,34(5):507-511 Liu Hongmei, Wang Shaoping, Ouyang Pingchao. Helicopter rotor smoothing based on GRNN neural network and genetic algorithm[J]. Journal of Beijing University of Aeronautics and Astronautics,2008,34(5):507-511(in Chinese) [5] Kurtaran H, Ozcelik B,Erzurumlu T. Warpage optimization of a bus ceiling lamp base using neural network model and genetic algorithm[J]. Journal of Materials Processing Technology,2005,169(2):314-319 [6] 汪苏,陈佳,李建明.复杂陶瓷件烧结过程温度场数值模拟及仿真[J].北京航空航天大学学报,2004,30(8):787-790 Wang Su, Chen Jia, Li Jianming. Study on the simulation of the temperature field of sintering complex ceramics parts[J]. Journal of Beijing University of Aeronautics and Astronautics,2004,30(8):787-790(in Chinese) [7] 李艳霞,张佐光,李敏,等.复合材料等厚层板热压成型中树脂流动过程数值模拟[J].复合材料学报,2008,25(2):47-50 Li Yanxia, Zhang Zuoguang, Li Min, et al. Numerical simulation on resin flow of laminated composites in hot pressing process[J]. Acta Materiae Composite Sinica,2008,25(2):47-50(in Chinese) [8] 沃丁柱,李顺林,王兴业,等.复合材料大全[M].北京:化学工业出版社, 2000: 161 Wo Dingzhu, Li Shunlin, Wang Xingye, et al. Cyclopedia of composite materials[M]. Beijing: Chemical Industry Press,2000:161(in Chinese) [9] 张银生,冀克俭,刘元俊, 等.用差示扫描量热法研究环氧树脂的固化特性[J].工程塑料应用,1997,25(6):22-24 Zhang Yinsheng, Ji Kejian, Liu Yuanjun, et al. Study on curing kinetics of epoxy resin using differential scanning calorimetry[J]. Engineering Plastics Application,1997,25(6):22-24(in Chinese) [10] 谢怀勤,卢少微,王武娟.埋入CFRP的FBG光纤传感器界面传递特性实验研究[J].材料科学与工艺,2006,14(6):605-607 Xie Huaiqin, Lu Shaowei, Wang Wujuan. Experomental study on interface-transferring properties of FBG embedded on CFRP[J]. Materials Science & Technology,2006,14(6):605-607(in Chinese) [11] Liu Lihui, Zhang Hao. Temperature-independent FBG pressure sensor with high sensitivity[J]. Optical Fiber Technology,2007,13(1):78-80 [12] Zhao Yong, Yu Chengbo, Liao Yanbiao. Differential FBG sensor for temperature-compensated high-pressure (or displacement) measurement[J]. Optics & Laser Technology,2004,36(1):39-42 [13] Suratno B R, Ye L, Mai Y W. Simulation of temperature and curing profiles in pultruded composite rods[J]. Composites Science and Technology,1998,58:191-197 [14] Liu Xiaolin, Grouch I G, Lam Y C. Simulation of heat transfer and cure in pultrusion with a general-purpose finite element package[J]. Composites Science and Technology,2000,60:857-864

施引文献

资源附件(0)

访问统计