| 用模糊聚类分析法建立神经网络岩土本构模型 |
| |
| 引用本文： | 魏星,黄茂松,虎旭林. 用模糊聚类分析法建立神经网络岩土本构模型[J]. 宁夏大学学报(自然科学版), 2004, 25(1): 54-56 |
| |
| 作者姓名： | 魏星 黄茂松 虎旭林 |
| |
| 作者单位： | 同济大学,地下建筑与工程系,上海,200092;西北第二民族学院,信息与计算科学系,宁夏,银川,750021 |
| |
| 摘 要： | 针对岩土类材料本构关系复杂的情况 ,采用BP网络建立了其本构模型 ,并从神经网络隐层的本质特征出发 ,对样本数据 (应力和应变信息 )采用模糊聚类分析法进行特征抽取 ,以确定隐层节点数目 .算例结果表明 ,该方法是可行的 ,并具有较直观的物理意义 ,可为样本的选取提供指导作用 .
|
| 关 键 词： | 模糊聚类 前向神经网络 岩土本构模型 |
| 文章编号： | 0253-2328(2004)01-0054-03 |
| 修稿时间： | 2003-10-13 |
Construction of Neural Network Constitutive Model of Rock and Soil by the Fuzzy Cluster Analysis Method |
| |
| Wei Xing,Huang Maosong,Hu Xulin. Construction of Neural Network Constitutive Model of Rock and Soil by the Fuzzy Cluster Analysis Method[J]. Journal of Ningxia University(Natural Science Edition), 2004, 25(1): 54-56 |
| |
| Authors: | Wei Xing Huang Maosong Hu Xulin |
| |
| Affiliation: | Wei Xing~1,Huang Maosong~1,Hu Xulin~2 |
| |
| Abstract: | The constitution of soil and rock is very complex, so we adopt BP network to construct it. According to the essence of hidden layer, a fuzzy cluster analysis method is put forward to exact the feature of sample data and both the strain and stress is comprehensively considered. Example has proved the feasibility of the fuzzy cluster method, which is more accessible and can be used to estimate whether the sample selection is rational. |
| |
| Keywords: | fuzzy cluster feed forward neural network constitutive model of rock and soil |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
