力学与实践 ›› 2016, Vol. 38 ›› Issue (3): 306-309,316.DOI: 10.6052/1000-0879-16-073

• 应用研究 • 上一篇    下一篇

基于级连相关神经网络的人工冻土本构模型

陈军浩1, 乔成2   

  1. 1. 福建工程学院土木工程学院, 福州 350118;
    2. 中国科学院水利部成都山地灾害与环境研究所, 成都 610041
  • 收稿日期:2016-03-16 修回日期:2016-04-10 出版日期:2016-06-15 发布日期:2016-06-20
  • 通讯作者: 陈军浩,博士,讲师,主要从事岩土与地下工程领域的教学与科研工作.E-mail:chjhtougao@163.com.
  • 基金资助:

    国家自然科学基金项目(51504070)、福建省教育厅中青年教师教育科研项目(JA15353)和福建工程学院科研启动基金(GY-Z15004)资助.

CONSTITUTIVE MODEL OF ARTIFICIAL FROZEN SOIL BASED ON CASCADE-CORRELATION NEURAL NETWORK

CHEN Junhao1, QIAO Cheng2   

  1. 1. College of Civil Engineering, Fujian University of Technology, Fuzhou 350118, China;
    2. Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Ministry of Water Resources, Chengdu 610041, China
  • Received:2016-03-16 Revised:2016-04-10 Online:2016-06-15 Published:2016-06-20

摘要:

为解决BP (back propagation) 神经网络收敛速度慢,网络结构需事先定义等缺点,采用了级连相关神经网络模型来建立人工冻土应力和应变之间的关系. 基于该模型推导了冻土的一致刚度矩阵形式,利用人工冻土三轴试验数据对神经网络模型进行训练,并用其替换有限元计算中的传统本构模型,将计算结果与性质及含水率相同的冻土的试验结果进行了对比,发现该神经网络本构模型很好地反应了材料的非线性,能够改善数值计算结果,与实测结果吻合地很好,比具有相同隐含层神经元个数的BP 模型更接近实测结果.

关键词:

神经网络|本构模型|人工冻土

Abstract:

In order to avoid the drawback of BP (back propagation) neural network of too slow velocity convergence, the network structure has to be defined in advance. A cascade-correlation artificial neural network model is adopted to create the relationship between stress and strain of the artificial frozen soil, the consistent stiffness matrix is derived based on this model for the frozen soil, the neural network model is trained by the triaxial test data to replace the traditional finite element constitutive model, and the calculated results of the properties and the moisture content of the frozen soil are compared with the experimental results. It is found that this neural network constitutive model can represent the nonlinear response of the material very well, can improve the numerical analysis results, with very good agreement with the measured results, and the results are closer to the measured results than the BP model with same number of neurons in the hidden layer.

Key words:

neural network|constitutive model|artificial frozen soil

中图分类号: