基于改进回声状态网络的盖亚大数据短时交通状态预测研究
A Short-Term Traffic Forecasting Research Based on Didi Chuxing GAIA Open Dataset Using Echo State Network Optimized by PSO Algorithm
DOI: 10.12677/HJDM.2019.94019, PDF,    科研立项经费支持
作者: 张家顺：河北工业大学，交通运输系，天津
关键词: 回声状态网络；交通预测；粒子群算法；Echo State Network； Traffic Forecasting； PSO Algorithm
摘要: 短期交通预测是现代智能交通系统重要的组成部分，本文设计了改进的回声状态网络基于盖亚开放数据来对短时交通状态进行预测。对于数据集中的无效数据，采用随机森林算法进行数据预处理，剔除其中的无效数据。由于盖亚数据集中的实时交通数据量巨大，采用粒子群算法对回声状态网络参数进行优化。最后采用2016年10月~11月成都市二环局部区域轨迹数据作为样本集对所设计的方法进行了验证，结果表明了该方法可以有效的对短时交通状态进行预测。
Abstract: Short-term traffic forecasting is an important part of contemporary intelligent transportation systems. In this paper, based on echo state network, a procedure is proposed to provide a forecast of traffic state. For the invalid data in the Didi Chuxing GAIA Open Dataset, the random forest algo-rithm is used to preprocess the data and eliminate the invalid data. Because of the huge amount of real-time traffic data in GAIA data set, particle swarm optimization algorithm is employed to op-timize the parameters of echo state network. Finally, the method is validated by using the local area trajectory data of the second ring road in Chengdu from October to November 2016 as the sample set. The result illustrates the effectiveness of this method.
文章引用：张家顺. 基于改进回声状态网络的盖亚大数据短时交通状态预测研究[J]. 数据挖掘, 2019, 9(4): 153-158. https://doi.org/10.12677/HJDM.2019.94019

参考文献

[1] 蒲斌, 李浩, 卢晨阳, 王治辉, 刘华. 基于神经网络的海量GPS数据交通流量预测[J]. 云南大学学报(自然科学版), 2019, 41(1): 53-60.
[2] 张海鹏, 杨宏业, 邬鑫珏, 王葆元. 基于公交车GPS数据的短时交通流预测研究[J]. 内蒙古工业大学学报(自然科学版), 2018, 37(1): 75-80.
[3] 晏臻, 于重重, 韩璐, 苏维均, 刘平. 基于CNN+LSTM的短时交通流量预测方法[J]. 计算机工程与设计, 2019(9): 2620-2624+2659.
[4] 李志帅, 吕宜生, 熊刚. 基于图卷积神经网络和注意力机制的短时交通流量预测[J]. 交通工程, 2019, 19(4): 15-19+28.
[5] 闫杨, 孙丽珺, 朱兰婷. 一种基于时空相关性的短时交通流量预测方法[J/OL]. 计算机工程, 1-8.
[6] Lin, Y., Zhang, J.-W. and Liu, H. (2019) Deep Learning Based Short-Term Air Traffic Flow Prediction Considering Temporal—Spatial Correlation. Aerospace Science and Technology, 93, Article ID: 105113.
[7] Emami, A., Sarvi, M. and Bagloee, S.A. (2019) Using Kalman Filter Algorithm for Short-Term Traffic Flow Prediction in a Connected Vehicle Environment. Journal of Modern Transportation, 27, 222-232.
[Google Scholar] [CrossRef] 
[8] Jaeger, H. (2010) The “Echo State” Approach to Analysing and Training Recurrent Neural Networks—With an Erratum Note. German National Research Center for Information Technology GMD Technical Report Vol. 148, Bonn, 13.
[9] Jaeger, H. and Haas, H. (2004) Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Telecommunication. Science, 304, 78-80.
[Google Scholar] [CrossRef] [PubMed]
[10] Zhang, Q.Y., Qian, H., Chen, Y.P. and Lei, D.M. (2019) A Short-Term Traffic Forecasting Model Based on Echo State Network Optimized by Improved Fruit Fly Optimization Algorithm. Neurocomputing.
[Google Scholar] [CrossRef] 
[11] 王小洁. 基于回声状态网络的船舶交通事故预测[J]. 舰船科学技术, 2019, 41(16): 16-18.
[12] 李丁园. 回声状态网络结构设计及应用研究[D]: [博士学位论文]. 长春: 吉林大学, 2019.
[13] 张晋雁, 陶宏才. 回声状态网络研究[J]. 成都信息工程学院学报, 2015, 30(6): 546-550.
[14] Thiede, L.A. and Parlitz, U. (2019) Gradient Based Hyperparameter Optimization in Echo State Net-works. Neural Networks, 115, 23-29.
[15] 张家顺. 基于随机森林算法的盖亚大数据清洗的研究[J]. 计算机科学与应用, 2019, 9(9): 1747-1752.

为你推荐