發(fā)布時(shí)間：2018-03-16 12:41

  本文選題：短期交通流預(yù)測(cè)　切入點(diǎn)：深度學(xué)習(xí)　出處：《南京郵電大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：道路交通系統(tǒng)是一個(gè)國(guó)家國(guó)民經(jīng)濟(jì)發(fā)展的基礎(chǔ),建造一個(gè)合理高效的道路交通系統(tǒng)是至關(guān)重要的。隨著人們出行需求的增多,大眾和企業(yè)對(duì)道路交通系統(tǒng)的便捷度要求越來(lái)越高。為了解決道路擁堵的狀況,我們著力于研究交通流預(yù)測(cè)技術(shù),期望能獲取未來(lái)短期時(shí)間內(nèi)精準(zhǔn)的車流量數(shù)據(jù),以實(shí)現(xiàn)車輛分流、交通誘導(dǎo)、道路規(guī)劃和交通設(shè)施合理分布等目的。本文主要的研究對(duì)象是交通流數(shù)據(jù),研究目標(biāo)是精確的預(yù)測(cè)某選定路段未來(lái)一天以內(nèi)的交通流量,研究?jī)?nèi)容是交通流數(shù)據(jù)預(yù)處理、交通流預(yù)測(cè)模型搭建和預(yù)測(cè)效果檢驗(yàn),設(shè)計(jì)了預(yù)測(cè)精度較高的交通流預(yù)測(cè)算法。本文的主要研究?jī)?nèi)容和研究成果如下:(1)本文首先對(duì)交通流數(shù)據(jù)的參數(shù)、特征和影響因素進(jìn)行分析,選取時(shí)間和車流量作為本文的研究參數(shù)。使用EViews數(shù)據(jù)分析器獲取數(shù)據(jù)的季節(jié)性和趨勢(shì)性特征,為選取合適的非線性預(yù)測(cè)模型做鋪墊。然后對(duì)交通流數(shù)據(jù)進(jìn)行預(yù)處理。使用SPSS數(shù)據(jù)分析器調(diào)整數(shù)據(jù)順序、添加空缺值并改正非常規(guī)值;再對(duì)數(shù)據(jù)進(jìn)行小波軟閾值去噪,去噪過程包括小波分解、軟閾值去噪和小波重構(gòu),使用matlab代碼實(shí)現(xiàn)并獲得去噪后的數(shù)據(jù)表和數(shù)據(jù)圖。(2)建立交通流數(shù)據(jù)的LSTM模型和GM模型。LSTM模型使用keras框架和python代碼編寫。將預(yù)處理后的部分?jǐn)?shù)據(jù)輸入進(jìn)搭建好的LSTM網(wǎng)絡(luò),LSTM通過學(xué)習(xí)數(shù)據(jù)的特征確定網(wǎng)絡(luò)參數(shù)和權(quán)值,并輸出未來(lái)一天的交通流數(shù)據(jù)。LSTM模型的預(yù)測(cè)效果較好,但是模型訓(xùn)練需要的數(shù)據(jù)量較大。GM屬于灰色模型,我們采取10個(gè)數(shù)據(jù)一個(gè)模型,不斷改變模型參數(shù),構(gòu)造動(dòng)態(tài)灰色模型。GM模型的預(yù)測(cè)效果不如LSTM模型的預(yù)測(cè)效果好,但是預(yù)測(cè)所需的數(shù)據(jù)量較少且實(shí)時(shí)性強(qiáng)。(3)將LSTM模型和GM模型使用動(dòng)態(tài)權(quán)值w集成。針對(duì)單個(gè)模型預(yù)測(cè)法在應(yīng)對(duì)突發(fā)狀況時(shí)容易遺漏和忽視,導(dǎo)致預(yù)測(cè)精度降低,采用兩種預(yù)測(cè)模型集成的方式對(duì)交通流預(yù)測(cè)進(jìn)行研究。集成方式為加權(quán)組合,權(quán)值w利用關(guān)聯(lián)系數(shù)確定,權(quán)值的動(dòng)態(tài)步調(diào)與GM的建模步調(diào)保持一致。集成模型的預(yù)測(cè)結(jié)果顯示,其預(yù)測(cè)精確度比兩個(gè)模型單獨(dú)預(yù)測(cè)的精確度高。
[Abstract]:The road traffic system is the foundation of a country's national economic development. It is very important to build a reasonable and efficient road traffic system. In order to solve the problem of road congestion, we are working on traffic flow forecasting technology to get accurate traffic data in the short term in the future. In order to realize the purpose of vehicle shunt, traffic guidance, road planning and reasonable distribution of traffic facilities, the main research object of this paper is traffic flow data, the research goal is to accurately predict the traffic flow of a selected section of the road within one day in the future. The content of the research is traffic flow data preprocessing, traffic flow forecasting model building and forecasting effect testing. A traffic flow forecasting algorithm with high precision is designed. The main contents and results of this paper are as follows: (1) in this paper, the parameters, characteristics and influencing factors of traffic flow data are analyzed. Time and traffic flow are selected as the parameters of this paper. The seasonal and trend characteristics of the data are obtained by using the EViews data analyzer. In order to select the suitable nonlinear prediction model, the traffic flow data is preprocessed. The SPSS data analyzer is used to adjust the data order, add the vacant value and correct the unconventional value, and then the wavelet soft threshold is used to de-noise the data. The denoising process includes wavelet decomposition, soft threshold denoising and wavelet reconstruction. The LSTM model of traffic flow data and GM model. LSTM model are written using keras framework and python code. The pre-processed part of the data is input into the constructed LSTM. The LSTM determines the network parameters and weights by learning the characteristics of the data. And output the traffic flow data. LSTM model in the next day has good prediction effect, but the model training needs a large amount of data. GM belongs to the grey model. We take 10 data and one model, and constantly change the model parameters. The prediction effect of dynamic grey model. GM model is not as good as that of LSTM model. But the amount of data needed for prediction is less and real-time. 3) the LSTM model and GM model are integrated with dynamic weight w. The prediction method of single model is easy to be omitted and ignored when dealing with sudden situation, which leads to the decrease of prediction accuracy. Traffic flow forecasting is studied by two integrated forecasting models. The integration method is a weighted combination, the weight w is determined by the correlation coefficient, the dynamic step of the weight value is consistent with the GM modeling step, and the prediction results of the integrated model show that, Its prediction accuracy is higher than that of the two models alone.
【學(xué)位授予單位】：南京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：U491.14

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 孫占全;潘景山;張贊軍;張立東;丁青艷;;基于主成分分析與支持向量機(jī)結(jié)合的交通流預(yù)測(cè)[J];公路交通科技;2009年05期

2 張九躍;焦玉棟;;基于RBF神經(jīng)網(wǎng)絡(luò)的短時(shí)交通流量預(yù)測(cè)[J];山東交通學(xué)院學(xué)報(bào);2008年03期

3 唐鐵橋,黃海軍;用燕尾突變理論來(lái)討論交通流預(yù)測(cè)[J];數(shù)學(xué)研究;2005年01期

，

本文編號(hào)：1619958