摘要
为了高效预测分析海量交通大数据,提高道路通行率和城市交通的智能化水平,提出一种高精度基于深度学习的并行卷积神经网络交通流量大数据预测模型.该模型首先对数据进行预处理以获得有效数据集,将具有规则时间间隔的一维时间序列样本和图像转换为时间一维、位置一维的二维像素网格,构建并行卷积神经网络模型用于对通过某路段的交通流量进行预测,并应用预测因子对交通量流数据进行建模.实验结果表明,与其他模型相比,本文提出的模型在平均绝对误差、平均相对误差和均方根误差方面均优于所对比的方法.
To predict and analyze the massive traffic big data efficiently to improve the road traffic rate and the intelligent level of urban traffic,a high-precision traffic flow big data prediction model based on deep learning parallel convolution neural network has been proposed.Firstly,the model preprocesses the data to obtain the effective data set,and transforms the one-dimensional time series samples and images with regular time intervals into two-dimensional pixel grids with one-dimensional time and one-dimensional position,and builds a parallel convolutional neural network model to predict the traffic flow through a certain road section and apply the prediction factors model the traffic flow data.Experimental results show that,compared with other models,the proposed model is better than the compared methods in terms of average absolute error,average relative error and root mean square error.
作者
周思吉
钱真坤
ZHOU Siji;QIAN Zhenkun(Informatization Construction and Service Center, Sichuan University of Arts and Science, Dazhou Sichuan 635000, China;Logistics Service, Sichuan University of Arts and Science, Dazhou Sichuan 635000, China)
出处
《西南师范大学学报（自然科学版）》
CAS
2022年第8期9-15,共7页
Journal of Southwest China Normal University(Natural Science Edition)
基金
四川省高校后勤协会2022-2023年度立项课题(20220602).
关键词
交通数据可视化
并行卷积神经网络
深度学习
交通流量预测
traffic data visualization
parallel convolutional neural network
deep learning
traffic flow prediction
