基于指數(shù)平滑法和ARIMA的交通量組合預(yù)測模型應(yīng)用研究
[Abstract]:Road traffic flow forecasting is an important task in modern traffic management and planning, so it is also very important to accurately predict the future trend of traffic flow, so it is necessary to study the methods of traffic volume prediction and improve the prediction accuracy. In order to ensure the scientific and rational decision-making of traffic projects to be built, the waste of resources in the process of road planning and design is reduced, and the social benefit is improved. Through the extensive collection of relevant literature, combined with the research results of wavelet analysis and exponential smoothing method, this paper analyzes the existing traffic forecasting methods, and discusses the development and application of traffic forecasting theory at home and abroad. This paper analyzes the advantages and disadvantages of grey prediction theory, genetic algorithm, neural network, etc. On this basis, combining with the randomness and non-linear change of traffic volume historical data, the abnormal data appearing in data signal collection are analyzed. In this paper, different methods for identifying and correcting abnormal data are discussed, and these methods are applied to correct abnormal data in traffic data series. In addition, the paper also carries on the empirical research, combines the Zheng-Xiao expressway road traffic volume and the related data, through the comparative analysis, further validates the feasibility of the forecast method and the reliability of the forecast result. According to the calculation principle of exponential smoothing method and combined with the traffic volume data of Zheng-Shao Expressway, the paper makes the regularity more obvious and easy to predict by separating the factors on different scales (the traffic volume is influenced by many factors). In this paper, wavelet method is introduced into traffic volume prediction. The wavelet multi-scale exponential smoothing method is proposed by using the function of wavelet multi-scale analysis and the prediction of cubic exponential smoothing method. The principle of wavelet multi-scale function analysis and exponential smoothing method are combined organically. At the same time, a series of ARIMA models are used to predict traffic volume through Spss software to grasp the nonlinear characteristics of traffic volume in essence. Through further research and application of combinatorial forecasting method, this paper combines the above two methods effectively through linear combination, and validates the prediction of traffic volume by wavelet multi-scale exponential smoothing prediction and ARIMA model combination forecasting model. The combination of the two methods can predict the long-term trend of the future traffic volume of Zheng-Xiao Expressway accurately, enrich and develop the forecasting method of the traffic volume, and make decision-making and operation management of the expressway project. It has certain important academic research value and engineering application value.
【學(xué)位授予單位】:鄭州大學(xué)
【學(xué)位級別】:碩士
【學(xué)位授予年份】:2014
【分類號】:U491.14
【相似文獻】
相關(guān)期刊論文 前10條
1 盧建昌,張世英,牛東曉;基于ARIMA的發(fā)電量預(yù)測方法[J];華北電力大學(xué)學(xué)報;2004年03期
2 李佳萌,曾濤,王偉;ARIMA模型在天津港交通量預(yù)測中的應(yīng)用[J];數(shù)理醫(yī)藥學(xué)雜志;2004年05期
3 姜慶華;趙麗萍;;基于ARIMA模型的我國電力生產(chǎn)預(yù)測研究[J];價值工程;2006年09期
4 王婷;;民航客運量的ARIMA模型與預(yù)測[J];五邑大學(xué)學(xué)報(自然科學(xué)版);2007年01期
5 蔣金良;林廣明;;基于ARIMA模型的自動站風(fēng)速預(yù)測[J];控制理論與應(yīng)用;2008年02期
6 惠軍;馬冬冬;;ARIMA模型在房屋售價中的預(yù)測[J];佳木斯大學(xué)學(xué)報(自然科學(xué)版);2009年01期
7 周葉;肖靈機;;基于ARIMA模型的我國航空貨運量預(yù)測分析[J];南昌航空大學(xué)學(xué)報(社會科學(xué)版);2010年03期
8 任歡;;基于ARIMA模型的河南省能源需求量預(yù)測[J];安陽師范學(xué)院學(xué)報;2010年05期
9 張士強;王雯;王健;;ARIMA模型在城市年用電量預(yù)測中的應(yīng)用[J];電力需求側(cè)管理;2010年06期
10 董小剛;李純凈;;基于ARIMA模型的上證指數(shù)預(yù)測實證分析[J];長春工業(yè)大學(xué)學(xué)報(自然科學(xué)版);2012年05期
相關(guān)會議論文 前10條
1 ;A hybrid ARIMA-ANN model and its Learning Algorithm on Short-term Load Forecasting[A];第二十三屆中國控制會議論文集(下冊)[C];2004年
2 王建鋒;高歌;陳立凌;李紅美;張明芝;王艾麗;;ARIMA模型及其在江蘇省衛(wèi)技人員數(shù)預(yù)測中的應(yīng)用[A];中國現(xiàn)場統(tǒng)計研究會第12屆學(xué)術(shù)年會論文集[C];2005年
3 李君華;王志堅;張立杰;陳雪;;基于小波理論及ARIMA模型的短期棉花價格預(yù)測[A];中國棉花學(xué)會2012年年會暨第八次代表大會論文匯編[C];2012年
4 陳興榮;;ARIMA模型和GM(1,1)在我國白銀消費需求預(yù)測應(yīng)用中的比較研究[A];第25屆全國灰色系統(tǒng)會議論文集[C];2014年
5 ;Double Trends Time Series Forecasting Using a Combined ARIMA and GMDH Model[A];Proceedings of 2010 Chinese Control and Decision Conference[C];2010年
6 劉軍;柴洪洲;陳軻;劉先冬;;ARIMA模型預(yù)報電離層VTEC研究[A];第一屆中國衛(wèi)星導(dǎo)航學(xué)術(shù)年會論文集(下)[C];2010年
7 ;Economic Design of Integrating SPC and APC with Quality Constraints[A];Proceedings of 2010 Chinese Control and Decision Conference[C];2010年
8 任家福;張f ;周宗放;;基于ARIMA和BP神經(jīng)網(wǎng)絡(luò)的組合預(yù)測模型及應(yīng)用研究[A];第三屆(2008)中國管理學(xué)年會——技術(shù)與創(chuàng)新管理分會場論文集[C];2008年
9 ;Modeling Chronobiologic Data: An Introduction to Time Series Analysis[A];2004全國時間生物醫(yī)學(xué)學(xué)術(shù)會議論文集[C];2004年
10 ;Traffic Flow Forecasting Based on Fuzzy-Neural[A];第二十六屆中國控制會議論文集[C];2007年
相關(guān)重要報紙文章 前1條
1 ;基于數(shù)量化方法對未來經(jīng)濟增長趨勢的預(yù)測[N];第一財經(jīng)日報;2009年
相關(guān)碩士學(xué)位論文 前10條
1 梁佳琦;ARIMA模型同MAXENT模型在自然保護區(qū)內(nèi)口蹄疫疫情風(fēng)險預(yù)警中的應(yīng)用研究[D];東北林業(yè)大學(xué);2015年
2 張海東;基于論壇的熱點話題識別與趨勢預(yù)測研究[D];上海師范大學(xué);2015年
3 雍永強;基于ARIMA和BPNN的組合預(yù)測模型在血糖預(yù)測中的應(yīng)用[D];鄭州大學(xué);2015年
4 張敏;基于ARIMA的組合模型問題研究[D];大連海事大學(xué);2015年
5 桂林;退耕還林與生態(tài)環(huán)境改善的關(guān)系研究與分析[D];西安建筑科技大學(xué);2015年
6 錢麗萍;基于ARIMA模型的兒童醫(yī)院門診量預(yù)測研究[D];蘇州大學(xué);2015年
7 于婷;基于ARIMA模型的股價的研究[D];大連海事大學(xué);2015年
8 程浩;武漢第三產(chǎn)業(yè)總量時間序列研究[D];華中師范大學(xué);2015年
9 陳天舒;基于ARIMA與GPR組合模型的人民幣匯率預(yù)測[D];山東大學(xué);2015年
10 袁磊;基于ARIMA-LSSVM混合模型的股指預(yù)測研究[D];哈爾濱工業(yè)大學(xué);2015年
,本文編號:2460611
本文鏈接:http://sikaile.net/kejilunwen/jiaotonggongchenglunwen/2460611.html