基于时间序列分析的短时交通流量实时自适应预测-控制理论与控制工程专业论文.docxVIP

AbstractReal。time Abstract Real。time and accurate short—term traffic flow forecasting has become a critical problem in intelligent transportation systems(ITS)．The strong random disturbance and strong uncertainty of model bring much difficulity for accnrate traffic flow forecasting。In addition，nonstationariPy of traffic flow data series is another problem to be solved． Based on time series analysis method adopting AR(p)model。a馘致d of real—time adaptive forecasting method for short·term traffic flow Was presented．In this method the recursive forgetting factor least square method(RFFLS)was adopted for parameter estimation．The Astrom forecasting algorithm WaS used for forecasting， which is based on linear minimum square error ofprediction．A lot ofreal observation data are used for simulation tests and results show that when forgetting faetor is decreased，the one—step forecasting performance caFl be improved．沁addition，when this method is respectively applied to the data at the weekday and the weekend，both simulation tests have good forecasting performance，which demonstrates that this method has good adaptability in different traffic flow circumstances， Then based on this method a kind of improving multi-step adaptive forecasting method was presented．The error compensation item Was added to this new method which could well meet the needs offorecasting for time．vailant models。A lot ofreal observation data are used for simulation tests and results show that when the improving method is applied to the strong time-variant multi。．step short—term traffic flow forecasting，it has good forecasting performance which is superior to the linear minimum square error forecasting method’S+ Another adaptive forecasting method based on GM(1，1)model is also applied to the short*term traffic flow forecasting and simulation results show it also has good forecasting performance．But its forecasting performance is not superior to the adaptive forecasting method’S based on AR(1 1)model． In order t