摘要
应用改进的多重分形消除趋势波动分析法对长江流域某水文站多年月径流资料进行分析计算,结果显示,月径流序列具有长程相关性和多重分形特征,并采用二项倍增串级模型,对其多重分形谱拟合,表明月径流序列具有较强的多重分形性。通过对其初始序列、重排序列和替代序列进行对比分析,揭示出月径流序列多重分形特征是序列本身的长程相关性和胖尾概率分布共同作用的结果,且胖尾概率分布起主要作用。
The improved multifractal detrended fluctuation analysis （MF-DFA） method is applied to analyze the long-term monthly runoffs of a hydrological station on Yangtze River, and finds the monthly runoff series has long-range correlation and multifraetal nature. The multifractal spectrum is fitted by a generalized expression of the muhiplieative cascade model, and the results show that the series has strong multifractal nature. Comparing the results for the original series to those for shuffled and surrogate series, it concludes that the multifractal nature of the monthly runoff series is due to both broad probability density function and long-range correlation, and the broad probability density function is dominant.
出处
《水力发电》
北大核心
2011年第9期21-24,共4页
Water Power
基金
国家自然科学基金资助项目(40971018)
湖北省自然科学基金资助项目(2009CDB200)
