摘要
基于南桠河冶勒水库月平均径流资料,运用重标极差(R/S)分析方法揭示了径流复杂的非线性特性隐藏下的有序性,即持续性特性。由于持续性区间内径流变化的高度相关性,由此通过组织历史数据建立迭代函数系统,经分形插值方法求取吸引子并在吸引子基础上进行延拓,建立了基于分形插值的预测模型。通过实际验证表明,该预测模型取得了较好的结果,满足实际应用要求,并为径流预测提供了新的探索及进一步研究的铺垫。
According to the historical monthly runoff data of Yele reservoir on Nanya river, the durability property concealed behind the complex non-linear characteristics of runoff may be discovered by using the method of rescaled range analysis. Due to the close correlation among monthly runoff within the durability span, the iterated function system（IFS） is established by organizing the historical runoff data. Based on the IFS, the runoff forecasting model may be established with extending the attractor by using fractal interpolation method. A practical example proves that the model is valid and practical, it can lay a foundation for the runoff prediction in the future.
出处
《水力发电学报》
EI
CSCD
北大核心
2008年第4期20-25,共6页
Journal of Hydroelectric Engineering
基金
国家自然科学基金资助项目(50539140)
关键词
水文学
径流预测
分形插值
迭代函数系统
R/S分析
hydrology
runoff prediction
fractal interpolation
iterated function system
rescaled range analysis
