欢迎登录材料期刊网

材料期刊网

高级检索

针对液-固挤压复合材料管、棒材成形时工艺参数难于选取、试验工作量大的问题,在正交试验的基础上,结合有限元模拟数据,构建200组样本集,将其中的150组作为训练样本用于网络的训练学习,其余的50组作为测试样本用于验证网络的精确性.通过对补偿模糊神经网络学习算法实现中的关键技术问题的处理,如输入、输出变量模糊集的划分、模糊规则的提取、学习速率的确定等,基于模糊神经网络建立了液-固挤压复合材料工艺系统模型,得到了浸渗时间与其它关键参数之间的映射关系及模糊规则,利用该模型,对关键工艺参数进行预测,预测值与试验值吻合较好.这为该工艺的实际应用和过程控制奠定了基础.

参考文献

[1] HU Lian-xi;LUO Shou-jing;HUO Wen-can.Development of the technique of extrusion directly following infiltration for the manufacturing of metal-matrix composites[J].Journal of Materials Processing Technology,1995:287-294.
[2] 齐乐华,孙乐民,侯俊杰,李贺军.高效成形复合材料管、棒材的新工艺研究[J].西北工业大学学报,1999(04):629-632.
[3] 齐乐华,李贺军,崔培玲,史忠科.Forming of tubes and bars of alumina/LY12 composites by liquid extrusion process[J].中国有色金属学会会刊(英文版),2003(04):803-808.
[4] 罗守靖;田文彤;李金平;压铸专刊.21世纪最具发展前景的近静成形技术-半固态加工[J].特种铸造及有色合金,2001(07):175-180.
[5] 张广安,罗守靖,田文彤.短碳纤维增强铝基复合材料的挤压浸渗工艺[J].中国有色金属学报,2002(03):525-528.
[6] 齐乐华,崔培玲,史忠科,李贺军.液-固挤压Al2O3sf /LY12复合材料管材成形过程的数值模拟[J].机械工程学报,2002(07):102-106.
[7] ZU Li-jun;LUO Shou-jing .Study on the power mixing and semi-solid extrusion forming process of SiCp/2024Al composites[J].Journal of Materials Processing Technology,2001,114:189-193.
[8] Hu Lianxi;Luo Shoujing;Huo Wencan .DETERMINATION OF THRESHOLD PRESSURE FOR INFILTRATION OF LIQUID ALUMINIUM INTO SHORT ALUMINA FIBER PREFORM[J].Transactions of Nonferrous Metals Society of China,1996(4):133-137.
[9] Park M W;Park B T;Kim S K et al.Incremental supervised learning of cutting conditions using the fuzzy ARTMAP neural network[Z].Annals of the CIRP,2000,49:375.
[10] Gurkan E.;Erkmen I.;Erkmen AM. .Two-way fuzzy adaptive identification and control of a flexible-joint robot arm[J].Information Sciences: An International Journal,2002(1/2):13-43.
[11] Chungchoo C;Saini D.On-line tool wear estimation in CNC turning operations using FNN model[J].Machine Tools and Manufacture,2002:29-40.
[12] Gwo-Ching Liao;Ta-Peng Tsao .Application of fuzzy neural networks and artificial intelligence for load forecasting[J].Electric Power Systems Research,2004(3):237-244.
[13] Yan-Qing Zhang;Kandel A. .Compensatory neurofuzzy systems with fast learning algorithms[J].IEEE Transactions on Neural Networks,1998(1):83-105.
[14] 齐乐华,史忠科,侯俊杰,李贺军.复合材料棒材半固态挤压工艺参数的神经网络预测方法[J].塑性工程学报,2003(01):20-24.
[15] Qi LH.;Shi ZK.;Li HJ.;Cui PL.;Han HM. .Simulation of liquid infiltration and semi-solid extrusion for composite tubes by quasi-coupling thermal-mechanical finite element method[J].Journal of Materials Science,2003(17):3669-3675.
上一张 下一张
上一张 下一张
计量
  • 下载量()
  • 访问量()
文章评分
  • 您的评分:
  • 1
    0%
  • 2
    0%
  • 3
    0%
  • 4
    0%
  • 5
    0%