發(fā)布時(shí)間：2018-03-05 22:22

  本文選題：交通流預(yù)測(cè)　切入點(diǎn)：機(jī)器學(xué)習(xí)　出處：《西南交通大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：經(jīng)濟(jì)的高速發(fā)展,城市化水平的不斷提高,在改善人民生活質(zhì)量的同時(shí),也隨之帶來(lái)了嚴(yán)重的交通擁堵問(wèn)題,如何利用城市的歷史交通流量,對(duì)未來(lái)的交通狀況進(jìn)行快速而精準(zhǔn)的預(yù)測(cè),是智能交通領(lǐng)域一大重要的研究課題。傳統(tǒng)的處理交通流預(yù)測(cè)問(wèn)題的方法可以分為基于數(shù)學(xué)模型的方法(如卡爾曼濾波模型、時(shí)間序列模型等)和無(wú)數(shù)學(xué)模型的方法(如神經(jīng)網(wǎng)絡(luò)模型、非參數(shù)回歸模型等)。然而,傳統(tǒng)的方法在應(yīng)對(duì)變化日益復(fù)雜的交通流數(shù)據(jù)上,已經(jīng)表現(xiàn)出了一定的局限性,這主要表現(xiàn)為:(1)在應(yīng)對(duì)非線性問(wèn)題上,許多算法存在局限性;(2)交通流的非平穩(wěn)特性,大大影響著模型的預(yù)測(cè)精度;(3)大量樣本所帶來(lái)的對(duì)于效率的挑戰(zhàn)。近年來(lái),隨著數(shù)據(jù)挖掘、機(jī)器學(xué)習(xí)等以數(shù)據(jù)為導(dǎo)向的技術(shù)的興起,對(duì)于交通流預(yù)測(cè)的研究越來(lái)越多地與以上算法結(jié)合,這帶來(lái)了預(yù)測(cè)精度的大大提升。論文以美國(guó)加州交通局Pems數(shù)據(jù)集作為實(shí)驗(yàn)數(shù)據(jù),首先,針對(duì)交通流的非平穩(wěn)特性,提出基于DBSCAN算法與最優(yōu)分割算法結(jié)合的雙階段有序聚類模型,實(shí)現(xiàn)了在缺少先驗(yàn)知識(shí)的條件下,以更小開銷對(duì)有序樣本的聚類,并在實(shí)驗(yàn)數(shù)據(jù)上證明了聚類結(jié)果的合理性;在有序聚類模型的基礎(chǔ)上,提出基于時(shí)間分段的支持向量機(jī)模型,以擬合優(yōu)度作為指標(biāo),證明了該模型能夠達(dá)到理想的回歸精度;論文還提出基于歷史數(shù)據(jù)加權(quán)的交通流序列生成模型,該模型利用基于時(shí)間分段的支持向量機(jī)模型來(lái)進(jìn)行參考值的生成,從而將生成的參考值與歷史數(shù)據(jù)進(jìn)行加權(quán),并通過(guò)迭代上述過(guò)程,生成交通流序列,并在與真實(shí)序列的比較中,證明了該模型所生成序列的精度;最后,論文引入標(biāo)簽傳播算法,將實(shí)驗(yàn)數(shù)據(jù)中的各個(gè)采樣時(shí)刻點(diǎn),根據(jù)其對(duì)應(yīng)特征分為上升點(diǎn)、下降點(diǎn)、平穩(wěn)點(diǎn)三類模式。在此分類結(jié)果的基礎(chǔ)上,引入隨機(jī)森林模型,以實(shí)時(shí)的交通流序列作為輸入,識(shí)別其對(duì)應(yīng)的交通變化模式。該模型在主要的性能指標(biāo)上,都達(dá)到了理想的效果。
[Abstract]:With the rapid development of economy and the continuous improvement of urbanization level, while improving the quality of life of the people, it also brings the serious traffic congestion problem, how to make use of the historical traffic flow of the city, Rapid and accurate prediction of future traffic conditions is an important research topic in the field of intelligent transportation. Traditional methods to deal with traffic flow forecasting problems can be divided into mathematical model-based methods (such as Kalman filter model). Time series models and methods without mathematical models (such as neural network models, non-parametric regression models, etc.). However, traditional methods have shown some limitations in dealing with increasingly complex traffic flow data. This is mainly shown as: 1) in dealing with nonlinear problems, many algorithms have limitations on the non-stationary characteristics of traffic flow, which greatly affect the prediction accuracy of the model and the efficiency challenge brought by a large number of samples. In recent years, with the data mining, With the rise of data-oriented technology such as machine learning, more and more research on traffic flow prediction is combined with the above algorithms, which brings a great improvement of prediction accuracy. This paper takes the Pems data set of California Transportation Bureau as experimental data. Firstly, aiming at the non-stationary characteristics of traffic flow, a two-stage ordered clustering model based on the combination of DBSCAN algorithm and optimal segmentation algorithm is proposed, which realizes the clustering of ordered samples with less cost under the condition of lack of prior knowledge. On the basis of the experimental data, the rationality of the clustering results is proved, and on the basis of the ordered clustering model, the support vector machine model based on time segmentation is proposed, which takes the goodness of fit as the index, and proves that the model can achieve the ideal regression accuracy. This paper also proposes a traffic flow sequence generation model based on historical data weighting. The model uses the support vector machine model based on time segmentation to generate reference value, thus weighting the generated reference value with historical data. By iterating the above process, the traffic flow sequence is generated, and the accuracy of the sequence generated by the model is proved in the comparison with the real sequence. Finally, the paper introduces the label propagation algorithm to sample each sampling time point in the experimental data. According to its corresponding characteristics, it can be divided into three models: ascending point, descending point and stationary point. On the basis of the classification results, the stochastic forest model is introduced, and the real-time traffic flow sequence is used as the input. The corresponding traffic change patterns are identified and the model achieves ideal results in terms of main performance indicators.
【學(xué)位授予單位】：西南交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP181;U491.14

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 張亮;寧芊;;CART決策樹的兩種改進(jìn)及應(yīng)用[J];計(jì)算機(jī)工程與設(shè)計(jì);2015年05期

2 郭釔宏;王博;劉勇;楊亦寧;;綜合優(yōu)度法和不一致性法的最優(yōu)分割參數(shù)選擇方法[J];遙感技術(shù)與應(yīng)用;2014年03期

3 ;The essential ability of sparse reconstruction of different compressive sensing strategies[J];Science China(Information Sciences);2012年11期

4 孔英會(huì);景美麗;;基于混淆矩陣和集成學(xué)習(xí)的分類方法研究[J];計(jì)算機(jī)工程與科學(xué);2012年06期

5 郭麗娟;孫世宇;段修生;;支持向量機(jī)及核函數(shù)研究[J];科學(xué)技術(shù)與工程;2008年02期

6 燕孝飛;葛洪偉;顏七笙;;RBF核SVM及其應(yīng)用研究[J];計(jì)算機(jī)工程與設(shè)計(jì);2006年11期

7 楊勝,李莉,胡福喬,施鵬飛;基于決策樹的城市短時(shí)交通流預(yù)測(cè)[J];計(jì)算機(jī)工程;2005年08期

8 韓超,宋蘇,王成紅;基于ARIMA模型的短時(shí)交通流實(shí)時(shí)自適應(yīng)預(yù)測(cè)[J];系統(tǒng)仿真學(xué)報(bào);2004年07期

，

本文編號(hào)：1572166