發(fā)布時(shí)間：2018-05-28 06:45

  本文選題：BP神經(jīng)網(wǎng)絡(luò) + K近鄰　； 參考：《北京交通大學(xué)》2017年碩士論文

【摘要】：隨著經(jīng)濟(jì)的快速發(fā)展,交通運(yùn)輸已經(jīng)迅速成為國(guó)民經(jīng)濟(jì)發(fā)展命脈。雖然,交通行業(yè)的快速發(fā)展給人們帶來(lái)了巨大便利,但隨之而來(lái)的就是嚴(yán)重的交通擁堵問(wèn)題。實(shí)時(shí)準(zhǔn)確的交通流預(yù)測(cè)是交通引導(dǎo)系統(tǒng)中的關(guān)鍵技術(shù),而交通路線系統(tǒng)是智能交通系統(tǒng)的重要組成部分。由于交通系統(tǒng)是一個(gè)有人為因素參與的非平穩(wěn)的隨機(jī)系統(tǒng),傳統(tǒng)的線性模型預(yù)測(cè)越來(lái)越不適應(yīng)于非線性的交通預(yù)測(cè)了,智能預(yù)測(cè)和組合優(yōu)化模型越來(lái)越受到人們的關(guān)注。本文深入研究符合交通流數(shù)據(jù)特性的代表性預(yù)測(cè)方法,經(jīng)分析選取人工智能中的的經(jīng)典方法BP神經(jīng)網(wǎng)絡(luò)作為交通流預(yù)測(cè)的基本算法。傳統(tǒng)的BP神經(jīng)網(wǎng)絡(luò)在進(jìn)行交通流預(yù)測(cè)時(shí),訓(xùn)練時(shí)間和訓(xùn)練精度往往不能同時(shí)得到保證。首先,本文分析傳統(tǒng)BP神經(jīng)網(wǎng)絡(luò)的預(yù)測(cè)模式,提出一種將輸出層的動(dòng)態(tài)值變?yōu)槎ㄖ档念A(yù)測(cè)模式加速收斂。其次,在定值輸出層的基礎(chǔ)上,針對(duì)神經(jīng)網(wǎng)絡(luò)訓(xùn)練時(shí)間長(zhǎng)的缺點(diǎn),提出一種利用K近鄰算法優(yōu)化訓(xùn)練數(shù)據(jù)集的K-BP預(yù)測(cè)模型。該模型在提前考慮預(yù)測(cè)數(shù)據(jù)與訓(xùn)練數(shù)據(jù)匹配度的前提下,進(jìn)行BP神經(jīng)網(wǎng)絡(luò)的訓(xùn)練數(shù)據(jù)集篩選。相比于傳統(tǒng)神經(jīng)網(wǎng)絡(luò),該模型在縮短訓(xùn)練時(shí)間的前提下減小了訓(xùn)練誤差。隨著信息技術(shù)與物聯(lián)網(wǎng)技術(shù)在城市交通領(lǐng)域的廣泛應(yīng)用,城市交通流量的數(shù)據(jù)已經(jīng)呈現(xiàn)出大數(shù)據(jù)的諸多特征。傳統(tǒng)的神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型在小規(guī)模訓(xùn)練樣本的前提下,還能滿足交通流預(yù)測(cè)需求。但隨著訓(xùn)練樣本的維度和數(shù)據(jù)量不斷增多,傳統(tǒng)的神經(jīng)網(wǎng)絡(luò)在訓(xùn)練樣本方面往往會(huì)消耗過(guò)長(zhǎng)時(shí)間,不利于實(shí)現(xiàn)實(shí)時(shí)的短時(shí)交通預(yù)測(cè)。本文提出了在Hadoop環(huán)境下利用MapReduce的分布式處理框架與BP神經(jīng)網(wǎng)絡(luò)相結(jié)合的預(yù)測(cè)模型,該模型利用BP神經(jīng)網(wǎng)絡(luò)的MapReduce并行化在保證預(yù)測(cè)精度的同時(shí)減小預(yù)測(cè)時(shí)間,達(dá)到預(yù)測(cè)的實(shí)時(shí)性。
[Abstract]:With the rapid development of economy, transportation has become the lifeline of the development of national economy. Although the rapid development of the transportation industry has brought great convenience to people, it is followed by serious traffic congestion. Real-time and accurate traffic flow prediction is a key technology in traffic guidance system, and traffic route system is an important part of intelligent transportation system. Because the traffic system is a non-stationary stochastic system with the participation of artificial factors, the traditional linear model prediction is becoming more and more unsuitable for nonlinear traffic forecasting, and intelligent forecasting and combinatorial optimization models have attracted more and more attention. In this paper, the representative forecasting methods which accord with the characteristics of traffic flow data are deeply studied. The classical artificial intelligence method BP neural network is selected as the basic algorithm of traffic flow prediction. The traditional BP neural network can not guarantee the training time and precision simultaneously in traffic flow prediction. Firstly, this paper analyzes the prediction model of traditional BP neural network, and proposes a prediction model which can change the dynamic value of the output layer into a fixed value. Secondly, on the basis of constant output layer, aiming at the disadvantage of long training time of neural network, a K-BP prediction model is proposed to optimize the training data set using K-nearest neighbor algorithm. The training data set of BP neural network is filtered on the premise of considering the matching degree between prediction data and training data in advance. Compared with the traditional neural network, the model reduces the training error on the premise of shortening the training time. With the wide application of information technology and Internet of things technology in the field of urban transportation, the data of urban traffic flow have shown many characteristics of big data. The traditional neural network forecasting model can meet the demand of traffic flow forecasting on the premise of small scale training samples. However, with the increasing of dimension and data volume of training samples, the traditional neural networks often consume too long time in training samples, which is not conducive to real-time short-term traffic prediction. In this paper, a prediction model based on the distributed processing framework of MapReduce and BP neural network is proposed in Hadoop environment. In this model, the MapReduce parallelization of BP neural network is used to ensure the prediction accuracy and reduce the prediction time to achieve the real-time prediction.
【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP183;U491.1

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 馮青平;李星毅;;基于大數(shù)據(jù)的非參數(shù)回歸短時(shí)交通流預(yù)測(cè)方法[J];無(wú)線通信技術(shù);2015年03期

2 錢(qián)偉;楊慧慧;孫玉娟;;相空間重構(gòu)的卡爾曼濾波交通流預(yù)測(cè)研究[J];計(jì)算機(jī)工程與應(yīng)用;2016年14期

3 關(guān)學(xué)忠;佟宇;高哲;皇甫旭;聶品磊;白云龍;;基于分形理論的短期電力負(fù)荷預(yù)測(cè)[J];計(jì)算機(jī)與數(shù)字工程;2014年11期

4 王彥明;;近年來(lái)Hadoop國(guó)內(nèi)研究進(jìn)展[J];現(xiàn)代情報(bào);2014年08期

5 ;大數(shù)據(jù)構(gòu)建智慧醫(yī)療[J];IT經(jīng)理世界;2014年13期

6 謝海紅;戴許昊;齊遠(yuǎn);;短時(shí)交通流預(yù)測(cè)的改進(jìn)K近鄰算法[J];交通運(yùn)輸工程學(xué)報(bào);2014年03期

7 陳星;武麗芳;王福明;;基于GA-BP神經(jīng)網(wǎng)絡(luò)的股票預(yù)測(cè)研究[J];山西電子技術(shù);2014年01期

8 李軍懷;高瞻;王志曉;張t，

本文編號(hào)：1945710