摘要
随着机动车辆的不断增多,实现智能交通降低拥堵成为现代城市亟待解决的问题,而交通流量预测是智能交通的关键因素。在短时交通流量预测中,由于车流变化快,突发情况多,传统的预测方法不能很好地进行预测。极限学习机(ELM)的快速发展为解决这类问题提供理论依据。我们针对传统的ELM存在分类精度低、网络结构稳定性差等问题,提出一种自适应混沌粒子群算法(ACPSO)优化ELM参数的算法,以此来增强网络的稳定性,提高ELM对数据分类的精度,并将该方法应用到短时交通流量预测上。实验结果表明,该算法具有较好的稳定性和可靠性,在交通流量预测中具有实用性和推广性。
With the increasing number of vehicles, the realization of intelligent transportation and congestion reduction has become an urgent problem in modern cities, and traffic volume prediction is the key factor of intelligent transportation. In the short time traffic flow prediction, the traditional forecasting method cannot be well predicted because of the fast change of traffic flow and the sudden situation. The rapid development of the limit learning machine（ELM） provides a theoretical basis for solving these problems. To solve the low classification accuracy,network structure and poor stability of the traditional ELM, proposes an Adaptive Chaos Particle Swarm Optimization algorithm（ACPSO） to optimize the parameters of ELM algorithm, so as to enhance the stability of the network, improve the classification accuracy of ELM data,and the method is applied to the short-term traffic flow prediction. The experimental results show that the algorithm has good stability and reliability, and it is practical and popularized in traffic flow prediction.
作者
汤代佳
尚东方
杨露
TANG Dai-jia;SHANG Dong-fang;PAN Jiao(Henan Union Information TeehnologT Co., Ltd., Xuehang 461670;Shenzhen Greatehn Technology Co. Ltd., Shenzhen 518033;College of Electrical Engineering, Shanghai DianJi University, Shanghai 201306)
关键词
极限学习机
自适应混沌粒子群
交通流量
实时预测
Traffic Flow
Extreme Learning Machine（ELM）
Particle Swarm Optimization （ACPSO）
Real-Time Prediction
