摘要
研究短时交通流预测问题。传统的非参数回归预测模型中,输入状态向量的选取没有综合考虑预测点和领域点的情况。为了提高预测的精度,提出了用熵值组合的非参数回归短时交通流预测方法。首先用灰关联度和相关系数法确定邻域点和预测点的输入状态向量,用反馈机制动态调节输入变量与系统变量,然后采用带权重的预测算法分别计算这两组参数下的预测值,采用熵值法构造组合预测的权系数,可求解出新的预测结果。实测数据仿真结果表明,组合预测能够提高预测的准确性,为实际设计提供依据。
Study short - term traffic flow forecasting. In the traditional nonparametric regression prediction model, the input variables are selected from the prediction point or adjacent points independently. To improve the accuracy of short - term traffic flow forecasting, a combined forecasting method based on nonparametric regression was put forward. Specifically, two single forecasts used grey correlation degree and correlation coefficient respectively to dynamically select all input variables with error feedback to tune the system parameters and the input variables. Finally, the results of those two forecasting methods were combined together in which entropy theory was utilized to determine the weight of each single forecasting method. The simulation with highway traffic data demonstrates that the proposed combined forecasting method can effectively improve the forecasting accuracy.
出处
《计算机仿真》
CSCD
北大核心
2013年第2期140-143,共4页
Computer Simulation
基金
广西科技开发计划项目(桂科攻0992006-13)
广西科技大科学基金(0977207)
关键词
交通流预测
非参数回归
熵值法
Traffic flow forecast
Nonparametric regression
Entropy Approach
