-
针对去趋势波动分析方法中参数不重叠等长度子区间长度s的选取,基于信息论的基本原理,提出使用符号分析方法对原始数据进行符号编码,并使用不同的方式对符号序列进行分段、计算互信息函数. 细致描述了不同分段方式对原始混沌序列的信息编码能力,以此判断所采用的分段方式能否真实有效地还原原始序列所包含的全部信息. 给出了确定最优分段个数或各分段长度的具体方式,确定了不重叠等长度子区间长度s的选取算法,以及判断所研究序列是否适用于去趋势波动分析方法,避免了以往参数s选取中随机性和主观性给计算结果带来的错误信息. 进一步将该方法应用于实际温度资料,计算并分析中国1961—2000年逐日平均温度的去趋势波动分析指数分布状况.We develop a method to compute the segment size in the detrended fluctuation analysis (DFA), which is based on the basic concept of the information theory, and verify the method effectiveness by numerical experiment. This method is freed from the problem of subjectivity in the former process to choose the segment size which usually leads to false result. We Change the length of sequence with dynamics being the same, the results remain stable. The results indicate that when the length of sequence is too short, even the optimal selection of segment size is not enough for the portrait of the overall dynamic system, thus the DFA cannot be used in this circumstance. The method we developed in this paper can enhance the reliability of DFA results by judging whether the sequences analyzed meet the requirements of DFA. We also obtain the DFA index from 1961 to 2000 of China through DFA method and analyze its spatial characteristics of distribution.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] 501
[31] -
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] 501
[31]
计量
- 文章访问数:8634
- PDF下载量:839
- 被引次数:0
下载: