基于卡尔曼滤波的短时交通流量预测模型研究

发布时间：2018-07-04 16:58

本文选题：交通流量预测 + 智能交通　；参考：《沈阳工业大学》2014年硕士论文

【摘要】：随着社会的进步,交通问题己日益受到人们的关注。为了优化道路环境，保证交通畅通，减少空气污染、汽车噪声的危害，许多国家都在开展智能交通（ITS）的研究，作为ITS的重要研究领域——交通控制与诱导系统是智能交通系统建设的核心课题，而实现交通流诱导系统的关键问题是准确的短时交通流量预测，即如何有效地利用实时交通数据信息去滚动预测未来几分钟内的交通状况。早在20世纪六七十年代，国外就开始将预测模型用于短时交通流量预测领域。交通流预测的研究模型有很多种，如：神经网络模型、多元线性回归模型、时间序列模型、历史趋势模型、Kalman滤波模型等。而本文则着重研究Kalman滤波在交通流预测中的应用。本文研究了交通流的静态稳定性以及突变性，，对交通流的可预测性进行判别。结合灰色关联分析方法建立Kalman滤波交通流预测模型。本文对交通流在空间上分布的特点进行分析，利用灰色关联分析方法，分析被测路段会受到哪些参数的影响。此外,本文为了改善Kalman滤波模型预测效果，提出了利用相邻数周中相对应时间的交通流比值代替原始数据，建立基于历史数据的Kalman滤波交通流预测模型。本文将所建立预测模型与其他基于kalman滤波的交通流预测模型作对比，研究表明本文算法的计算模型性能指标要优于其他预测模型。本文利用模拟数据对上述预测模型及算法进行了验证。实验结果表明：灰色关联分析能够有效地分析出各项影响交通流的参数，提高预测模型的适应性；以历史数据、实时数据为基础的预测模型，其预测效果要优于只运用实时数据的交通流量预测模型，从而证明了该模型的适应性强，预测精度高。
[Abstract]:With the development of the society, people pay more and more attention to the traffic problem. In order to optimize the road environment, ensure the smooth flow of traffic, reduce air pollution and the harm of automobile noise, many countries are carrying out research on Intelligent Transportation (its). As an important research field of its, traffic control and guidance system is the core of its construction, and the key problem of realizing traffic flow guidance system is accurate short-term traffic flow forecasting. That is, how to use real-time traffic data effectively to predict traffic situation in the next few minutes. As early as 1960s and 1970s, foreign countries began to use forecasting models in the field of short-term traffic flow forecasting. There are many research models for traffic flow prediction, such as neural network model, multivariate linear regression model, time series model, historical trend model and Kalman filter model. This paper focuses on the application of Kalman filter in traffic flow prediction. In this paper, the static stability and catastrophe of traffic flow are studied, and the predictability of traffic flow is judged. The traffic flow prediction model of Kalman filter is established by using grey correlation analysis method. In this paper, the characteristics of traffic flow distribution in space are analyzed, and the influence of the parameters on the measured road sections is analyzed by using the grey correlation analysis method. In addition, in order to improve the prediction effect of Kalman filter model, a Kalman filtering traffic flow prediction model based on historical data is established by using the traffic flow ratio corresponding to the corresponding time in adjacent weeks instead of the original data. In this paper, the proposed prediction model is compared with other traffic flow prediction models based on kalman filter. The results show that the performance of the proposed algorithm is better than that of other models. In this paper, the simulation data are used to verify the above prediction model and algorithm. The experimental results show that the grey correlation analysis can effectively analyze the parameters that affect the traffic flow and improve the adaptability of the forecasting model, which is based on historical data and real-time data. The forecasting effect is better than the traffic flow forecasting model which only uses real time data, which proves that the model has strong adaptability and high prediction accuracy.
【学位授予单位】：沈阳工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495;U491.1

【参考文献】