摘要
短时交通流预测是动态交通控制和诱导的前提。提出一种模糊C均值聚类和神经网络相结合的短时交通流预测方法。用同一组实测数据对比计算了该方法与BP神经网络预测方法、模糊神经网络预测方法分别得到的预测结果。计算结果表明:所提出的方法的预测准确性明显地高于其他两种方法。
Short-term traffic flow prediction is the basis of dynamic traffic control and guidance. In this article, a new fore- casting method, which combines fuzzy C-mean clustering and neural network techniques, is used to forecast short-term (traffic) flow. Based on the same real-time data, the forecasting results of this method are computed, and compared to other methods, e.g., Bp neural network and fuzzy neural network. The compute outcomes show that this method can produce (more) exact prediction results than other two methods.
出处
《系统工程》
CSCD
北大核心
2004年第8期83-86,共4页
Systems Engineering
关键词
交通控制与诱导
短时交通流预测
模糊C均值聚类
BP神经网络
Traffic Control and Guidance
Short-term Traffic Flow Forecasting
Fuzzy C-mean Clustering
Back-propagation(BP) Neural Network