节点文献
基于遗传单纯形神经网络的大坝变形监控模型
Dam deformation monitoring model based on neural network with genetic algorithm simplex method
【摘要】 本文针对遗传算法局部搜索能力差的缺陷,把单纯形法嵌入到遗传算法中构成复合遗传算法,建立了基于遗传单纯形神经网络的大坝变形监控模型。实例研究表明,该模型较遗传神经网络模型、BP模型收敛性能好,具有较高的预报精度、较快的训练速度和较强的泛化能力,用于大坝变形预测有效可行,具有良好的应用前景。
【Abstract】 Considering the poor local search ability of genetic algorithm,simplex method is combined with genetic algorithm to form hybrid genetic algorithm,and then the dam deformation monitoring model based on neural network with genetic algorithm simplex method is established.The example shows that the neural network model with genetic algorithm simplex method owns good convergent rate,high prediction accuracy,fast training speed and superior generating ability compared with the genetic algorithm neural network model and BP neural network model.This method is feasible and effective for dam deformation prediction and has wide applied prospect as well.
【Key words】 management of water conservancy project; dam deformation monitoring; genetic algorithm; neural network; simplex method;
- 【文献出处】 水力发电学报 ,Journal of Hydroelectric Engineering , 编辑部邮箱 ,2007年04期
- 【分类号】TV698.1
- 【被引频次】10
- 【下载频次】237