文章提出一种基于多分辨率学习的正交基小波神经网络结构的设计方法,网络权值的学习采用阻尼递推最小二乘算法。
A designing method of wavelet neural network structure based on multiresolution learning is put forward, and the studies of network weights adopt damped least squares.
法的无条件稳定性的证明,可不要求阻尼矩阵满足振型正交条件。
The proof of its unconditionally stable requires no damping matrix to satisfying the condition of the orthogonality of modes.
本文将该方法应用于一种复合材料等间距正交网格加筋板的阻尼特性分析,考察了加筋对复合材料层合板阻尼特性的影响。
The modal loss factor of an orthogrid-stiffened composite panel is studied using the present method and particular attention is paid to the effect of ribs on damping.
应用推荐