摘要
短时交通流预测是交通控制与交通诱导系统的关键问题之一。随着预测时间跨度的缩短,交通流量的变化 显示出越来越强的不确定性,使得一般的预测方法难以奏效。本文针对BP神经网络运行的特点,提出了用隔离小生 境遗传算法优化传统的BP网络。实例证明,该神经网络的进化建模方法设计简单,全局搜索效率较高,能有效的用 于短时交通流量的预测。
The short - term traffic flow forecasting for traffic flow is one of the key problems of traffic control and route guidance. With the shortening of the forecasting term, the uncertainty of traffic flow becomes more and more serious, so that the general approaches is incapable to forecast the short - term volume effectively. Aim at the specialties of BP, NN based on isolation niche technique Genetic Algorithm is proposed. Results show that the method has the superiority of computation complicacy, model performance evaluation and whole search efficiency.
出处
《现代交通技术》
2004年第1期60-62,共3页
Modern Transportation Technology
