| AFSA-BP神经网络在大坝变形预测中的应用 |
| |
| 引用本文： | 杨红,陈向阳,张飞,张付明.AFSA-BP神经网络在大坝变形预测中的应用[J].地理空间信息,2012,10(6):131-132,138,1. |
| |
| 作者姓名： | 杨红 陈向阳 张飞 张付明 |
| |
| 作者单位： | 长江三峡勘测研究院有限公司（武汉）,湖北武汉,430074 |
| |
| 摘 要： | 经典BP神经网络的初始权值和阈值随机给定,使得训练速度慢、网络易于陷入局部极值。引入具有强大全局搜索能力的人工鱼群算法(AFSA)优化BP网络的权值和阈值,建立了基于AFSA-BP神经网络的预测模型,并对大坝的实测资料进行了实证分析。与经典BP神经网络预测模型的预测结果比较表明:AFSA-BP神经网络模型不仅训练速度快,而且预测精度明显提高,是一种较好的大坝变形预测模型。
|
| 关 键 词： | 人工鱼群算法 BP神经网络 大坝变形 预测 |
Application of AFSA-BP Neural Network in Dam Displacement Prediction |
| |
| Abstract: | Initialized weights and thresholds of the BP neural network are random, which results in slow conver-gence and easily converging to local optima. According to these characteristics, Artificial Fish Swarm Algorithm (AFSA), which has strong global searching ability, was utilized to optimize the weights and thresholds of the BP neural network in this paper. It was established the model of dam displacement prediction based on AFSA-BP neutral network and the actual material data of a dam was used for evaluating the model. And it contrasted with ordinary BP neural network estimate result, the result indicated that the AFSABP neural network not only trains in a faster speed, but more accurate in prediction. And it is a better model of dam displacement prediction. |
| |
| Keywords: | Artificial Fish Swarm Algorithm BP neural network dam deformation prediction |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
