摘要
实时准确的交通流量预测是实现智能交通诱导及控制的前提与关键,也是智能化交通管理的客观需要。把粗集理论引入神经网络的构造,应用粗神经元取代部分常规神经元,给出了一种交通流量的粗神经网络预测模型。实验结果表明,该模型在交通流量预测的精度和对交通路网的适应性方面明显优于常规神经网络,具有较高的实用价值。粗神经网络具有极强的鲁棒性,预测模型也可方便地处理季节、天气等随机因素对交通流量预测结果的影响。
Real-time and accurate traffic flow forecast is very important to the intelligent traffic guidance,control and management.Combining together rough sets and neural network formation and replacing some traditional neural cells with rough neural cells,a traffic flow forecast model based on rough neural network concept is given in this paper.The experiment results show that this model is superior to the models constructed with traditional neural cells in terms of forecast precision and adaptability to the traffic road networks.The rough neural network is robust to the uncertain factors such as seasons and weather in traffic flow forecast and the model is of academic and practical value in forecasting applications.
出处
《公路交通科技》
CAS
CSCD
北大核心
2004年第10期95-98,共4页
Journal of Highway and Transportation Research and Development
关键词
交通流预测
粗神经网络
智能交通系统
Traffic flow forecasting
Rough neural network
Intelligent Transportation Systems(ITS)