摘要
为解决卡尔曼滤波模型预测交通流量存在的时间滞后性问题,在已有卡尔曼滤波短时交通流量预测模型基础上,结合城市道路交通流日相似性特点,对已有卡尔曼滤波预测模型进行改进,并通过模糊逻辑方法对改进模型中的参数加以确定,设计出模糊卡尔曼滤波交通流量预测模型,从而对短时交通流量进行实时准确预测.数值分析及对比结果表明:相较于卡尔曼短时交通流量预测方法,模糊卡尔曼短时交通流量预测方法能够提升预测过程的实时性,并使平均相对误差降低0.27%,平均绝对相对误差降低7.26%,最大绝对相对误差降低32.43%,进一步提高了预测精度.
A new short-term traffic flow prediction model is proposed which is based on the Kalman filtering predictions. In order to avoid the time lag problem on the flow prediction, the Kalman filtering prediction model existed is improved and the parameters of the model are determined by the fuzzy logic method. This fuzzy Kalman filtering traffic flow prediction model can accurately predict short-term traffic flow in real time. Experimental results show that the new model has higher accuracy and better instantaneity compared to the traditional Kalman filtering model. The average relative error is reduced to 0. 27%, average absolute relative error is reduced to 7.26 % lower and absolute maximum relative error is reduced to 32.43 %. It further improves the prediction accuracy.
出处
《浙江工业大学学报》
CAS
2013年第2期218-221,共4页
Journal of Zhejiang University of Technology
基金
国家自然科学基金资助项目(50908213)
浙江省自然科学基金资助项目(Y1100891)
关键词
卡尔曼滤波
模糊逻辑
短时交通流
预测模型
Kalman filtering
fuzzy logic
short-term traffic flow
prediction model
