论文部分内容阅读
根据混沌理论具有分析非线性动态系统混沌特性的特点,对货物发送量相关时间序列进行了分析和研究。本文在Takens相空间重构的基础上,利用C-C方法求嵌入时延与嵌入窗、G-P方法求嵌入维数;应用小数据量法计算铁路货物发送量相关时间序列的最大Lyapunov指数,并进行混沌特性分析,结果显示:货物发送增长量和增长率符合混沌特性,货物发送量不符合混沌特性;利用基于最大Lyapunov指数方法和BP神经网络方法对1999年1月到2013年4月共172个月的铁路货物发送增长量和增长率进行预测,预测结果表明基于最大Lyapunov指数预测值能够较好地与实际值相吻合,其预测的准确度明显好于BP神经网络预测值,因而混沌理论中的最大Lyapunov指数预测在货物发送量相关时间序列预测中有广泛的实用价值。
According to chaos theory has the characteristics of chaos analysis of nonlinear dynamic system, the related time series of cargo sending volume are analyzed and studied. Based on the reconstruction of Takens phase space, this paper uses the CC method to find the embedding delay and embedding window. The GP method is used to calculate the embedding dimension. The maximum Lyapunov exponent of the time series of the freight volume is calculated by the small data volume method, and the chaos The results show that the growth rate and growth rate of the goods are in accordance with the chaotic characteristics, and the quantity of goods sent does not conform to the chaotic characteristics. Based on the maximum Lyapunov exponent method and the BP neural network method, a total of 172 months from January 1999 to April 2013 The forecasting results show that the prediction based on the maximum Lyapunov exponent can be in good agreement with the actual value and the prediction accuracy is obviously better than the BP neural network prediction value. Therefore, in the chaos theory The maximum Lyapunov exponent forecast has a wide range of practical value in predicting the time series of cargo delivery.