信息扩散及神经网络在建立注塑预测模型中的应用
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
注塑工艺参数和塑料件质量之间的关联模型常采用神经网络建立,然而训练样本不足将直接影响模型的质量。为了在适量训练样本条件下得到较好的预测模型,对原始样本数据进行信息扩散处理,然后使用神经网络建模。笔者以汽车音响面板为例建立预测模型,确定了最佳工艺参数,证明了该方法的有效性。
Neural Network was widely used to uncover the relationship between injection molding process parameters and part quality.However,the number of training samples was a non-negligible factor which seriously affected the accuracy of the prediction model.To generate better model with certain samples,the information diffusion theory to derive new patterns for training the network was applied.Using the automobile sounder faceplate as the example,the best process parameters was achieved and the effectiveness of this method was proved.
引文
[1]刘斌,许建文,等.基于CAE和神经网络的注射成型工艺参数优化[J].工程塑料应用,2007,35(11):31-34.
    [2]贾相武.基于BP网络的注塑件翘曲变形工艺参数优化研究[J].塑料工业,2007,35(6):235-237.
    [3]黄娜斌,江波,张志莲,等.基于Taguchi实验设计法CAE技术优化注塑成型工艺参数[J].上海塑料,2007(3):28-31.
    [4]申长雨,王利霞,张勤星.神经网络与混合遗传算法结合的注塑成型工艺优化[J].高分子材料科学与工程,2005,21(5):23-27.
    [5]叶世伟,王海娟,译.神经计算原理[M].北京:机械工业出版社,2007.
    [6]黄崇福.信息扩散原理与计算思维及其在地震工程中的应用[D].北京:北京师范大学,1993.
    [7]刘涛,邵东国,刘武艺.基于信息扩散和BP神经网络的水资源预测[J].武汉大学学报,2007,40(3):31-33,75.
    [8]何书,王家鼎,王欢,等.基于信息扩散和BP网络的黄土边坡稳定性分析[J].西北大学学报,2008,38(6):983-988.
    [9]Huang Chongfu.Information diffusion techniques and small sample problem[J].International Journal of Information Technology and Decision Making,2002,1(2):229-249.
    [10]Huang Chongfu,Claudio Moraga.A diffusion-neural-network for learning from small samples[J].International Journal of Approxi-mate Reasoning,2004,35(2):137-161.

版权所有:© 2023 中国地质图书馆 中国地质调查局地学文献中心