长江口深水航道回淤量时间序列混沌特征分析

Chaotic characteristics of back-silting quantity of Yangtze River estuary’s deep-water channel

  • 摘要: 针对长江口深水航道回淤分布情况,以回淤最严重的H~N段为中间段P2段,H段以上为P1段,N段以下为P3段,将全部航道分为3段。采用混沌理论对深水航道全段及各分段回淤量时间序列的饱和关联维数以及K2熵进行混沌特征分析。各分段的饱和关联维数变化范围为1.80~2.15,K2熵变化范围为0.08~0.12;全段的分数维与K2熵的值大于各分段,分别为2.93和0.16。各分段的饱和关联维数研究表明,长江口深水航道回淤量的时间序列具有混沌特征,全段混沌特征的复杂性高于各分段。根据2011年,2012年和2013年长江口深水航道回淤量的时间序列,利用混沌方法对深水航道未来回淤量进行预测,各分段可预报时间尺度最多为1年,全段的可预报时间尺度为半年。给出了长江口深水航道全段及各分段回淤动力系统数学表达式的一般形式,全段需要3~6个状态变量,3个以上控制变量;各分段需要2~5个状态变量,3个以上控制变量。回淤动力系统数学表达式的一般形式可为建立回淤量预报模式提供参考。

     

    Abstract: According to the distribution of the back-silting quantity of the Yangtze River estuary’s deep-water channel, the whole channel is divided into three sections: the middle section P2 is section N~H, which is under the most severe back-silting condition while the other two sections are section P1 in the west of section H and section P3 in the east of section N. The chaotic characteristics of the whole and each section of the deep-water channel are analyzed by calculating the saturation correlation dimension and K2 entropy of back-silting quantity time series. The saturation correlation dimensions of sections P1, P2 and P3 are between 1.80 and 2.15 while the K2 entropies are between 0.08 and 0.12. The saturation correlation dimension and K2 entropy of the whole section are larger than each section, which are 2.93 and 0.16 respectively. The analysis results show that the back-silting quantity time series of the Yangtze River estuary’s deep-water channel has chaotic characterisics and that the chaotic characteristic of the whole channel is more complicated than that of each section. The longest time-scale for predicting each section is one year and the time-scale for predicting the whole channel is half a year, according to the time series of the back-silting quantity of the Yangtze River estuary’s deep-water channel in 2011, 2012 and 2013. A general form of the dynamic model of the whole and each section of the Yangtze River estuary’s deep-water channel is given in this paper. 3 to 6-state variables and more than 3 control variables are needed in the model of the whole channel while 2 to 5-state variables and more than 3 control variables are needed in the model of each section of the channel. The general form of the dynamic model of the Yangtze River estuary’s deep-water channel can provide references for the foundation of the back-silting quantity prediction model.

     

/

返回文章
返回