【作者】 李鸿雁； 姜珊； 王玉新；

【Author】 Li Hongyan Jiang Shan Wang Yunxin College of Environment and Resources,Jilin University,No 6 Ximinzhu Avenue,Changchun,Jilin Province,Post code 130026

【机构】 吉林大学环境与资源学院；

【摘要】 本文重点研究了水文序列的混沌特性分析中,延滞时间和嵌入维数对重构相空间及其混沌特性的影响。以嫩江流域上游阿彦浅站和中游江桥站月均径流时间序列为研究对象,得出如下规律性结论:①嫩江流域汛期为4个月,可以推断径流相关序列长度为4～5个月,延滞时间τ=3或4为合理选择。②借鉴月雨量时间序列可预报性实验结果,嫩江流域径流序列嵌入维数m的G-P法计算结果有其合理性,即嵌入维数取值不应大于7。③嫩江流域适合的重构相空间参数及混沌特征指标应为:延滞时间τ=3～4、嵌入维数m=6～7、最大Layapunov指数λ₁=0.24～0.32。更多还原

【Abstract】 In this paper,we study on the influence of delay time and embedding dimension on phase space reconstruction and its Chaotic characteristics by experimental analysis.Based on the monthly average of runoff time series of the typical station in Nenjiang River Basin,parameters of phase space reconstruction and maximum lyapunov index are analyzed comparatively by the methods of chaos theory and empirical analysis.By comparing the two methods,we draw the same conclusion:The delay timeτis 3 or 4,the embedding dimension m is 6 or 7,and the forecast time T_ f in medium and long-term runoff forecast is 3～4 months.更多还原