摘要
针对高速公路愈加拥堵问题,提出了一种新的高速公路交通流预测方法,采用改进萤火虫算法优化RBF神经网络进行求解,结合历史节假日高速公路交通流数据进行训练和预测,与RBF神经网络预测结果进行对比,改进萤火虫算法优化RBF神经网络具有更好的预测结果,能够为高速公路管理中心提供稳定可靠的数据来源,为交管部门合理诱导疏散,缓解交通拥堵提供有效的技术支持。
With the free access of highway on holidays, the traffic flow of expressway is getting larger and larger. In order to solve the problem of highway congestion, this paper proposes a new highway traffic flow prediction method, with the improved firefly algorithm to optimize the RBF neural network in combination with the history of the holiday highway traffic flow data for training and prediction. Compared with RBF neural network prediction results, the improved firefly algorithm for optimizing RBF neural network has better prediction results. It can provide stable and reliable data source for expressway management center, and provide effective technical support for traffic management department to induce evacuation reasonably and alleviate traffic congestion.
作者
乐冰
蔡延光
阮嘉琨
LE Bing;CAI Yanguang;RUAN Jiakun(School of Automation,Guangdong University of Technology,Guangzhou 510006,China)
出处
《东莞理工学院学报》
2019年第3期7-11,共5页
Journal of Dongguan University of Technology
基金
国家自然科学基金(61074147)
广东省自然科学基金(S2011010005059)
广东省教育部产学研结合项目(2012B091000171,2011B090400460)
广东省科技计划项目(2012B050600028,2014B010118004,2016A050502060)
广州市花都区科技计划项目(HD14ZD001)
广州市科技计划项目(201604016055)
广州市天河区科技计划项目(2018CX005)