Study on Short-Term Traffic Flow Velocity Prediction Based o
发布时间：2022-02-18 09:36
根据公安部交通管理局官方微博和公安部网站发布的数据,2018年全国新注册登记机动车3,172万辆,机动车保有量已达3.27亿辆,其中汽车2.4亿辆,小型载客汽车首次突破2亿辆;机动车驾驶人达4.09亿人,其中汽车驾驶人3.69亿人。截至2018年底全国汽车保有量达2.4亿辆,比2017年增加2285万辆,增长10.51%。面对如此多的汽车保有量,尤其是私家车的数量,虽然带给了人们诸多便利,但也加重了城市交通道路的负担,使得本来就高负荷运转的城市公路更加拥堵不堪,尤其是上下班高峰时期,大量因为堵车而停滞在道路上的车辆尾气的排放也加重了环境的污染。面对以上问题,目前的解决方法主要有以下几个方法:第一,控制汽车的数目,限制汽车的购买;第二,加强城市基础道路建设,扩建道路;第三,增加公共交通或修建地铁轻轨等;第四,搭建智能交通系统。前三种方法可以在一定程度上缓解交通的压力,但是他们不能从根本上解决城市交通拥堵的问题。随着互联网+热潮的兴起,近几年智能交通领域受到交通管理部门和相关企业越来越多的重视,交通智能领域迎来了新的发展机遇。交通预测是智能交通系统（ITS）的重要组成部分,在交通网络规划、...
【文章来源】：华中师范大学湖北省211工程院校教育部直属院校
【文章页数】：73 页
【学位级别】：硕士
【文章目录】：
Abstract
Nomenclature
1 Introduction
1.1 Research Background And Significance
1.2 Research Status at Home And Abroad
1.2.1 Prediction Methods Based on Statistical Theory
1.2.2 Prediction Methods Based on Neural Network
1.2.3 Prediction Methods Based on Hybrid Model
1.3 The Research Content And Innovation of This Thesis
1.3.1 The Main Research Content of This Thesis
1.3.2 The Main Innovation of This Thesis
1.3.3 Structure of This Thesis
2 Related Theoretical Research
2.1 Related Concepts of Traffic Flow
2.1.1 The Main Characteristic Parameters of Traffic Flow
2.1.2 Traffic Flow Characteristics And Influencing Factors
2.1.3 The Basic Flow of Traffic Flow Velocity Prediction
2.2 Time Series Theory
2.3 Support Vector Machine Regression (SVR)
2.4 A Brief Introduction to The Neural Network Correlation Model
2.4.1 Mathematical Model of Neuron
2.4.2 Recurrent Neural Network
2.4.3 LSTM Neural Network
3 Traffic Speed Prediction Algorithm Based on LSTM Neural Network And SVRCombination Model
3.1 Preamble
3.2 The Concept of Combined Forecasting Models
3.3 Principles of Combined Forecasting Models
3.4 Prediction Algorithm Based on LSTM-SVR Hybrid Model
3.5 Evaluation Index of The Model
3.6 Data Preprocessing
3.7 Experiment And Result Analysis
3.7.1 Model Training
3.7.2 Result Analysis
3.8 Summarizes of This Chapter
4 Fusion Model Based on New Information Source And Seq2Seq-LSTM-SVR
4.1 Preamble
4.2 A Discovery Algorithm For Extracting New Information Sources
4.3 LSTM Neural Network Based on Seq2Seq+Attention Model
4.3.1 Introduction of Seq2Seq Model With Attention Mechanism
4.3.2 Build The Seq2Seq-LSTM Neural Network With Integrate NewInformation Sources
4.4 Experimental Analysis
4.5 Summarizes of This Chapter
5 Summary And Future Work
5.1 Summarize
5.2 Future Work
References
Appendices
摘要
Acknowledgements
【参考文献】：
期刊论文
[1]改进人工蜂群算法优化RBF神经网络的短时交通流预测[J]. 黄文明,徐双双,邓珍荣,雷茜茜. 计算机工程与科学. 2016(04)
[2]一种基于非参数回归的交通速度预测方法[J]. 史殿习,丁涛杰,丁博,刘惠. 计算机科学. 2016(02)
[3]基于动态学习策略的群集蜘蛛优化算法[J]. 王艳娇,李晓杰,肖婧. 控制与决策. 2015(09)
[4]递推SOM神经网络在短时交通流预测中的应用[J]. 黄杰,李军,郭翔. 公路. 2015(04)
[5]DE优化T-S模糊神经网络的交通流量预测[J]. 侯越. 计算机工程与设计. 2013(09)
[6]基于Adaboost的BP神经网络改进算法在短期风速预测中的应用[J]. 吴俊利,张步涵,王魁. 电网技术. 2012(09)
[7]基于影响模型的短时交通流预测方法[J]. 丁栋,朱云龙,库涛,王亮. 计算机工程. 2012(10)
[8]粒子群优化RBF神经网络的短时交通流量预测[J]. 冯明发,卢锦川. 计算机仿真. 2010(12)
[9]非线性短时交通流的一种神经网络预测方法[J]. 华冬冬,陈森发. 现代交通技术. 2004(01)
本文编号：3630598
【文章来源】：华中师范大学湖北省211工程院校教育部直属院校
【文章页数】：73 页
【学位级别】：硕士
【文章目录】：
Abstract
Nomenclature
1 Introduction
1.1 Research Background And Significance
1.2 Research Status at Home And Abroad
1.2.1 Prediction Methods Based on Statistical Theory
1.2.2 Prediction Methods Based on Neural Network
1.2.3 Prediction Methods Based on Hybrid Model
1.3 The Research Content And Innovation of This Thesis
1.3.1 The Main Research Content of This Thesis
1.3.2 The Main Innovation of This Thesis
1.3.3 Structure of This Thesis
2 Related Theoretical Research
2.1 Related Concepts of Traffic Flow
2.1.1 The Main Characteristic Parameters of Traffic Flow
2.1.2 Traffic Flow Characteristics And Influencing Factors
2.1.3 The Basic Flow of Traffic Flow Velocity Prediction
2.2 Time Series Theory
2.3 Support Vector Machine Regression (SVR)
2.4 A Brief Introduction to The Neural Network Correlation Model
2.4.1 Mathematical Model of Neuron
2.4.2 Recurrent Neural Network
2.4.3 LSTM Neural Network
3 Traffic Speed Prediction Algorithm Based on LSTM Neural Network And SVRCombination Model
3.1 Preamble
3.2 The Concept of Combined Forecasting Models
3.3 Principles of Combined Forecasting Models
3.4 Prediction Algorithm Based on LSTM-SVR Hybrid Model
3.5 Evaluation Index of The Model
3.6 Data Preprocessing
3.7 Experiment And Result Analysis
3.7.1 Model Training
3.7.2 Result Analysis
3.8 Summarizes of This Chapter
4 Fusion Model Based on New Information Source And Seq2Seq-LSTM-SVR
4.1 Preamble
4.2 A Discovery Algorithm For Extracting New Information Sources
4.3 LSTM Neural Network Based on Seq2Seq+Attention Model
4.3.1 Introduction of Seq2Seq Model With Attention Mechanism
4.3.2 Build The Seq2Seq-LSTM Neural Network With Integrate NewInformation Sources
4.4 Experimental Analysis
4.5 Summarizes of This Chapter
5 Summary And Future Work
5.1 Summarize
5.2 Future Work
References
Appendices
摘要
Acknowledgements
【参考文献】：
期刊论文
[1]改进人工蜂群算法优化RBF神经网络的短时交通流预测[J]. 黄文明,徐双双,邓珍荣,雷茜茜. 计算机工程与科学. 2016(04)
[2]一种基于非参数回归的交通速度预测方法[J]. 史殿习,丁涛杰,丁博,刘惠. 计算机科学. 2016(02)
[3]基于动态学习策略的群集蜘蛛优化算法[J]. 王艳娇,李晓杰,肖婧. 控制与决策. 2015(09)
[4]递推SOM神经网络在短时交通流预测中的应用[J]. 黄杰,李军,郭翔. 公路. 2015(04)
[5]DE优化T-S模糊神经网络的交通流量预测[J]. 侯越. 计算机工程与设计. 2013(09)
[6]基于Adaboost的BP神经网络改进算法在短期风速预测中的应用[J]. 吴俊利,张步涵,王魁. 电网技术. 2012(09)
[7]基于影响模型的短时交通流预测方法[J]. 丁栋,朱云龙,库涛,王亮. 计算机工程. 2012(10)
[8]粒子群优化RBF神经网络的短时交通流量预测[J]. 冯明发,卢锦川. 计算机仿真. 2010(12)
[9]非线性短时交通流的一种神经网络预测方法[J]. 华冬冬,陈森发. 现代交通技术. 2004(01)
本文编号：3630598
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/3630598.html
