|
|||||
|
|
| 遗传算法改进的BP神经网络在混沌径流时间序列预测中的应用 | |
| 引用本文： | 刘媛媛,练继建,朱云.遗传算法改进的BP神经网络在混沌径流时间序列预测中的应用[J].水文,2007,27(2):45-48. |
| 作者姓名： | 刘媛媛 练继建 朱云 |
| 作者单位： | 1. 中国水利水电科学研究院防洪减灾所,北京,100038 2. 天津大学建工学院,天津,300072 3. 宁夏回族自治区水利厅防办,宁夏,银川,750001 |
| 摘 要： | 区别于传统的提取混沌时间序列饱和嵌入维数的方法,本文利用人工神经网络成功地对水库混沌径流时间序列的饱和嵌入维数进行了提取,计算了该时间序列里的最大Lyapunov指数,两种方法结果都证明了该时间序列的混沌性。并用遗传算法对BP神经网络进行了改进,利用该模型对三门峡水库混沌径流时间序列进行了预测。实例计算表明该方法解决了BP神经网络收敛速度慢和易于陷入极小值的问题,大大提高了BP神经网络的计算精度和收敛速度。无论在计算精度上还是在收敛次数上都优于没有改进的BP神经网络。 |
| 关 键 词： | 混沌径流时间序列 BP神经网络 遗传算法 Lyapunov指数 |
| 文章编号： | 1000-0852（2007）02-0045-04 |
| 修稿时间： | 2006-04-18 |
Application of BP Networks Improved by GA in Forecasting Chaotic Runoff Time Series |
|
| LIU Yuan-yuan,LIAN Ji-jian,ZHU Yun.Application of BP Networks Improved by GA in Forecasting Chaotic Runoff Time Series[J].Hydrology,2007,27(2):45-48. | |
| Authors: | LIU Yuan-yuan LIAN Ji-jian ZHU Yun |
| Institution: | 1. China Institute of Water Resources and Hydropower Research, Beijing 100038, China;2. College of Architecture Engineering, Tianjin University, Tianjin 300072, China;3. Water Resources Department of Ningxia, Yinchuan 750001, China |
| Abstract: | In this paper introduced Artificial Neural Networks(ANN),which is different from tradition way and has been used to take out the dimension of the chaotic runoff time series of reservoirs.The Genetic Algorithm(GA) and ANN were combined together to repair the shortcoming of each other.The new method has been applied to calculate and forecast the chaotic runoff of the Sanmenxia Reservoir.Contrast to the normal BP networks,the calculation error of the new model is reduced,and the convergence speed is improved. |
| Keywords: | chaotic runoff time series BP network GA Lyapunov exponent |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |
| 点击此处可从《水文》浏览原始摘要信息 | |
| 点击此处可从《水文》下载免费的PDF全文 | |