摘要
将混沌理论应用于水文变量尺度分析,对无定河流域不同时间尺度(80 d,40 d,20 d,10 d和5 d)降雨量进行降尺度转换.根据关联维数法对不同时间尺度降雨量间的分解系数序列进行混沌性识别,结果表明:不同时间尺度降雨量的分解系数存在低维混沌现象,用混沌方法进行降雨量降尺度分析是可行的.在此基础上,建立加权一阶局域法多步预测模型对不同时间尺度降雨量间分解系数进行了预测,继而实现了降雨量的降尺度分析,良好的预测结果及可操作性使得混沌理论为水文序列的降尺度分析提供了一种方法.
Rainfall data of successively doubled resolutions （i. e, 80 d, 40 d, 20 d, 10 d and 5 d） observed at Wuding river valley located at northwest China were studied with chaos theory. The correlation dimension method was employed to investigate the presence of chaos in the rainfall transformation. The results of investigating the presence of chaos in the distributions of weights between rainfall data of different scales indicate the existence of chaos in the rainfall transformation, suggesting the applicability of a chaotic model. The local-region multi-steps forecasting model was applied to the distributions of weighs prediction, the encouraging results indicated that the chaos theory provided a new method for hydrological time series disaggregation.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2008年第6期129-132,共4页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家重点基础研究发展计划资助项目(2006CB403400)
国家自然科学基金资助项目(50579079,50528910)
关键词
水文现象
时间序列
降雨
混沌理论
降尺度
hydrological phenomenon
time series
rainfall
chaos theory
downscaling
