论文部分内容阅读
针对传统的应用数学模型方法在短时交通流预测精度和实时性方面存在的问题,提出了将Volterra滤波器自适应预测模型用于短时交通流的实时预测。为提高预测精度,在Volterra滤波系数计算过程中采用归一化最小均方自适应算法进行多次训练。应用该预测模型对几个典型的非线性系统进行预测,验证了算法的准确性。然后再用此方法对微观实测交通流的时间序列进行实证分析。仿真结果表明,该预测模型对实测交通流时间序列具有很好的预测效果,可以满足实时交通流预测的需要。
Aiming at the problems existing in the application of traditional mathematical model methods in short-term traffic flow prediction accuracy and real-time performance, the Volterra filter adaptive prediction model is proposed for real-time prediction of short-term traffic flow. In order to improve the prediction accuracy, a normalized least mean square adaptive algorithm was used for multiple trainings in Volterra filter coefficient calculation. The prediction model is used to predict several typical nonlinear systems, which verifies the accuracy of the algorithm. Then this method is used to analyze the time series of micro-measured traffic flow. The simulation results show that this prediction model has a good predictive effect on the measured time series of traffic flow and can meet the needs of real-time traffic flow prediction.