论文部分内容阅读
蚁群聚类径向基函数(ACC-RBF)神经网络是将蚁群聚类算法和径向基函数神经网络组合运用的一种新型神经网络模型,把该网络用于水布垭高面板坝堆石体的多参数反演问题,在室内试验参数的基础上用有限元计算获得学习样本,采用该网络对坝体堆石料的邓肯E-B模型参数进行反演分析,用反演所得参数结合三维非线性有限元计算坝体应力变形,并与实测资料比较,得出很接近的结论。
Ant Colony Clustering Radial Basis Function (ACC-RBF) neural network is a new type of neural network model which combines ant colony clustering algorithm and radial basis function neural network. The neural network model is used in Shuibuya high panel dam The multi-parameter inversion of rockfill body is studied. Based on the laboratory test parameters, the learning sample is obtained by finite element method. The Duncan EB model parameters of rockfill material are inverted by using the network. Combined with the three-dimensional Nonlinear Finite Element Calculation dam stress and deformation, and compared with the measured data, come to a very close conclusion.