摘要
短时交通流量预测是智能交通系统中的重要环节,选用在短时交通流量预测方面表现出色的LSTM神经网络,并利用PSO算法优化LSTM神经网络模型。实验结果表明,与传统LSTM模型相比,所构建的PSO-LSTM模型对未来5分钟和10分钟两种短时交通流量预测,达到了更高的准确率。在此基础上,设计了一个交通流量预测网站更好地展示了预测结果,也方便用户随时查询。
Short term traffic flow prediction is an important part of intelligent transportation systems.This article selects the LSTM neural network that performs well in short-term traffic flow prediction,and uses the PSO algorithm to optimize the LSTM neural network model.Experiments have shown that the PSO-LSTM model constructed in this paper achieves higher accuracy in predicting short-term traffic flow for the next 5 minutes and the next 10 minutes compared with traditional LSTM models.On this basis,in order to better display the prediction results and facilitate users to query at any time,the author specifically designed a city traffic flow prediction website,which has certain practical application value.
作者
王宁
成利敏
甄景涛
段晓霞
Wang Ning;Cheng Limin;Zhen Jingtao;Duan Xiaoxia(Langfang Normal University,Langfang 065000,China)
出处
《廊坊师范学院学报(自然科学版)》
2024年第1期29-32,共4页
Journal of Langfang Normal University(Natural Science Edition)
基金
2020年廊坊市科学技术研究与发展计划(第一批)自筹经费项目“基于神经网络模型的廊坊市主干道交通流量预测网站设计”(2020011009)。
关键词
智能交通系统
短时交通流量预测
LSTM神经网络
PSO算法
交通流量预测网站
intelligent transportation system
short term traffic flow prediction
LSTM neural network
PSO algorithm
traffic flow prediction website