本文選題：占道停車(chē)　切入點(diǎn)：智能管理　出處：《北京郵電大學(xué)》2015年碩士論文

【摘要】：隨著經(jīng)濟(jì)的發(fā)展和人們消費(fèi)能力的提高,家用小汽車(chē)的數(shù)量日漸增長(zhǎng),城市中“停車(chē)難”問(wèn)題越來(lái)越嚴(yán)重�，F(xiàn)有的大型停車(chē)場(chǎng)難以滿(mǎn)足停車(chē)需求,增加占道停車(chē)的方式可以大大緩解這一難題。然而因?yàn)檎嫉劳＼?chē)場(chǎng)分布的不規(guī)律性及管理體系的不健全,存在個(gè)人私行收取費(fèi)用的狀況,且無(wú)嚴(yán)格統(tǒng)一的收費(fèi)準(zhǔn)則,管理部門(mén)沒(méi)法及時(shí)掌握泊車(chē)信息及停車(chē)費(fèi)用,駕駛員無(wú)法及時(shí)獲取所在位置附近的停車(chē)場(chǎng)及車(chē)位占用情況信息,所以需要構(gòu)建統(tǒng)一的智能管理系統(tǒng)。此外,國(guó)內(nèi)的大多數(shù)停車(chē)管理系統(tǒng)中的信息發(fā)布模塊,僅能顯示車(chē)位的實(shí)時(shí)信息,未能提供對(duì)短時(shí)間內(nèi)車(chē)位變化情況的預(yù)測(cè),導(dǎo)致駕駛員到達(dá)停車(chē)場(chǎng)后的實(shí)際車(chē)位占用情況可能與在停車(chē)場(chǎng)外看到或者查詢(xún)到的信息差別很大,甚至只能到其他停車(chē)場(chǎng)尋找車(chē)位,帶來(lái)諸多不便。 首先,本文針對(duì)當(dāng)前國(guó)內(nèi)外普遍使用的停車(chē)場(chǎng)系統(tǒng)的情況進(jìn)行了調(diào)研,指出了當(dāng)前工作中存在的問(wèn)題并予以剖析。進(jìn)而討論了時(shí)間序列預(yù)測(cè)的相關(guān)方法,重點(diǎn)討論了基于神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)的方法,針對(duì)BP神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)算法進(jìn)行了深入的研究。 然后,本文對(duì)城區(qū)占道停車(chē)智能管理系統(tǒng)的功能需求進(jìn)行了分析,并完成了整體的設(shè)計(jì)工作。第一,為了使管理系統(tǒng)智能化,將系統(tǒng)劃分為四個(gè)子系統(tǒng)：分別是車(chē)位信息采集子系統(tǒng),手持終端收費(fèi)子系統(tǒng),中心管理子系統(tǒng)以及停車(chē)誘導(dǎo)及車(chē)位預(yù)測(cè)子系統(tǒng)。在各子系統(tǒng)之間定義了通信格式及交互協(xié)議,實(shí)現(xiàn)了數(shù)據(jù)的采集、傳輸、處理及應(yīng)用。第二,在停車(chē)誘導(dǎo)子系統(tǒng)中引入了車(chē)位信息的預(yù)測(cè)功能,目的是對(duì)未來(lái)車(chē)位的變化情況實(shí)現(xiàn)短時(shí)預(yù)測(cè)。通過(guò)對(duì)時(shí)間序列預(yù)測(cè)的傳統(tǒng)方法、時(shí)間序列的非線(xiàn)性預(yù)測(cè)方法以及神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)方法的比較,最終創(chuàng)建了基于BP神經(jīng)網(wǎng)絡(luò)的車(chē)位信息預(yù)測(cè)模型,并用實(shí)際數(shù)據(jù)對(duì)該模型進(jìn)行了驗(yàn)證。 最后,結(jié)合實(shí)際項(xiàng)目需求,完成了對(duì)本系統(tǒng)的開(kāi)發(fā)和實(shí)現(xiàn)。
[Abstract]:With the development of economy and the improvement of people's consumption power, the number of household cars is increasing day by day, and the problem of "parking difficulty" is becoming more and more serious.The existing large parking lot is difficult to meet the parking demand, increasing the parking on the road can greatly alleviate this problem.However, due to the irregular distribution of parking lots and the unsound management system, there is a situation in which private individuals collect fees, and there are no strict and uniform charging criteria, so the management can not grasp parking information and parking fees in a timely manner.The driver can not get the information of parking lot and parking space in time, so it is necessary to construct a unified intelligent management system.In addition, the information release modules in most parking management systems in China can only display the real-time information of parking spaces, and fail to predict the changes of parking spaces in a short period of time.As a result, the actual parking space occupation after the driver arrives in the parking lot may be very different from the information seen or inquired outside the parking lot, even can only look for the parking space in other parking lot, which brings a lot of inconvenience.Firstly, this paper investigates the situation of parking lot system which is widely used at home and abroad, points out the problems existing in the current work and analyzes it.Then, the related methods of time series prediction are discussed, and the methods based on neural network prediction are discussed, and the BP neural network prediction algorithm is studied deeply.Then, this paper analyzes the functional requirements of urban parking intelligent management system, and completes the overall design work.First, in order to make the management system intelligent, the system is divided into four subsystems: parking information collection subsystem, handheld terminal charge subsystem, central management subsystem and parking guidance and parking prediction subsystem.The communication format and interactive protocol are defined among the subsystems, and the data collection, transmission, processing and application are realized.Secondly, the prediction function of parking space information is introduced in the parking guidance subsystem, in order to predict the future parking space changes in a short time.Through the comparison of the traditional methods of time series prediction, the nonlinear prediction methods of time series and the neural network forecasting methods, the vehicle parking information prediction model based on BP neural network is established.The model is validated with actual data.Finally, according to the actual project requirements, the development and implementation of the system is completed.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類(lèi)號(hào)】：U491.7;TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王軍;許宏科;蔡曉峰;孫磊;;基于BP神經(jīng)網(wǎng)絡(luò)的高速公路動(dòng)態(tài)交通流預(yù)測(cè)[J];公路交通技術(shù);2007年01期

2 田穎濤;王軍利;張偉;張宇翔;;先進(jìn)的停車(chē)誘導(dǎo)系統(tǒng)實(shí)施策略研究[J];中國(guó)人民公安大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年01期

3 劉引濤;;基于Spring的MVC模式網(wǎng)上銀行系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[J];電子設(shè)計(jì)工程;2013年07期

4 王健;施浩;;新一代動(dòng)態(tài)交通誘導(dǎo)系統(tǒng)應(yīng)用研究[J];中國(guó)公共安全(學(xué)術(shù)版);2013年04期

5 薛峰;梁鋒;徐書(shū)勛;王彪任;;基于Spring MVC框架的Web研究與應(yīng)用[J];合肥工業(yè)大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年03期

6 鄒德文;張春梅;袁建林;劉燕;;基于BP神經(jīng)網(wǎng)絡(luò)誤差修正的ARIMA模型對(duì)河北省入境游客量的預(yù)測(cè)[J];河北科技師范學(xué)院學(xué)報(bào)(社會(huì)科學(xué)版);2009年04期

7 李萍;曾令可;稅安澤;金雪莉;劉艷春;王慧;;基于MATLAB的BP神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)系統(tǒng)的設(shè)計(jì)[J];計(jì)算機(jī)應(yīng)用與軟件;2008年04期

8 謝宗旺;方旭升;;基于Struts2和Spring框架的Web整合開(kāi)發(fā)研究[J];價(jià)值工程;2011年07期

9 方成;賴(lài)智勇;馬國(guó)潔;;基于BP神經(jīng)網(wǎng)絡(luò)的交通事故易發(fā)段研究[J];山西建筑;2011年28期

10 李松;劉力軍;解永樂(lè);;遺傳算法優(yōu)化BP神經(jīng)網(wǎng)絡(luò)的短時(shí)交通流混沌預(yù)測(cè)[J];控制與決策;2011年10期

，

本文編號(hào)：1701457