【摘要】 利用FARIMA模型,对铁路数据网中的真实数据流量进行建模并分析,提出一种新型的基于FARIMA模型的铁路数据网流量预测方法,该方法能够同时描述网络流量的长相关特性和短相关特性。将FARIMA过程转换为差分过程和ARMA过程进行趋势预测,并且根据平均绝对误差、绝对百分比误差等多项指标进行比较验证。通过对高速铁路数据网6个月的数据进行建模分析,利用不同的参数设置预测未来2个月的流量趋势,并与真实数据进行对比。实验结果表明该方法比传统的基于ARMA模型的预测方法更为精准,能够适用于铁路数据网流量趋势预测。更多还原

【Abstract】 This paper analyzed and modelled the actual traffic data of railway data network by analyzing the principle of FARIMA model,and proposed a new traffic forecasting method based on FARIMA.The method can describe the long-range dependence and short-range dependence characteristics of network traffic simultaneously.The FARIMA process was transformed into differential process and ARMA process to forecast tendency.The fitting results were verified with the average absolute error and the mean absolute percentage error.By modeling and analyzing the six-month data in the high-speed railway data network,forecasting the traffic trend in the next 2 months by different parameter settings,and comparing with the real data,the experiment shows that the method is more accurate than the traditional prediction method based on ARMA model,and is absolutely applicable to railway data network traffic trend forecast.更多还原