本文選題：交通安全 + 事故黑點(diǎn)鑒別　； 參考：《東南大學(xué)》2015年碩士論文

【摘要】：已有統(tǒng)計(jì)數(shù)據(jù)顯示,交通事故已成為危害人類健康和社會(huì)發(fā)展的重大問題,也是我國(guó)人群傷害死亡的第一位原因。而事故黑點(diǎn)路段因其事故易發(fā),事故頻發(fā)等特點(diǎn),在道路安全改善項(xiàng)目中是需要優(yōu)先整治的目標(biāo)對(duì)象。然而已有的事故黑點(diǎn)鑒別方法對(duì)事故數(shù)據(jù)完整性要求較高,在實(shí)際工程應(yīng)用中實(shí)施困難并且鑒別結(jié)果多是以文字和表格形式展現(xiàn),可視性較差。因此本文應(yīng)用網(wǎng)絡(luò)核密度估計(jì)方法對(duì)城市路網(wǎng)事故黑點(diǎn)進(jìn)行鑒別研究。核密度估計(jì)方法是空間分析理論中一種非參數(shù)估計(jì)理論,該方法不對(duì)樣本數(shù)據(jù)分布進(jìn)行事先假設(shè),只需要獲得事故發(fā)生的空間位置,并且可以基于GIS平臺(tái)對(duì)鑒別結(jié)果進(jìn)行多角度可視化展現(xiàn)。本文首先對(duì)核密度估計(jì)中的兩個(gè)重要參數(shù)——窗寬和核函數(shù)對(duì)于核密度估計(jì)結(jié)果影響進(jìn)行了討論,接著對(duì)核密度估計(jì)方法應(yīng)用于城市路網(wǎng)事故黑點(diǎn)鑒別中的問題作出簡(jiǎn)要說明,并對(duì)核密度估計(jì)方法做出修正以適應(yīng)城市道路網(wǎng)絡(luò)約束下的事故黑點(diǎn)路段鑒別要求,之后介紹了網(wǎng)絡(luò)核密度估計(jì)方法應(yīng)用于城市路網(wǎng)事故黑點(diǎn)鑒別的簡(jiǎn)要流程。緊接著本文對(duì)修正后的網(wǎng)絡(luò)核密度估計(jì)方法應(yīng)用于城市路網(wǎng)事故黑點(diǎn)鑒別的結(jié)果進(jìn)行有效性檢驗(yàn),檢驗(yàn)結(jié)果表明相比于已有傳統(tǒng)的鑒別方法,網(wǎng)絡(luò)核密度估計(jì)方法鑒別結(jié)果有效性更高,可以被考慮在工程應(yīng)用中加以推廣。本文之后應(yīng)用ArcGIS Desktop軟件,滿足基于網(wǎng)絡(luò)核密度估計(jì)城市路網(wǎng)事故黑點(diǎn)鑒別結(jié)果可視化需求。研究?jī)?nèi)容包括建立道路交通事故數(shù)據(jù)庫(kù),對(duì)道路網(wǎng)絡(luò)對(duì)象進(jìn)行單元?jiǎng)澐?計(jì)算單元節(jié)點(diǎn)的網(wǎng)絡(luò)核密度值,以及基于單元網(wǎng)絡(luò)核密度值大小排序的事故黑點(diǎn)路段鑒別。本文研究的基于網(wǎng)絡(luò)核密度事故城市路網(wǎng)事故黑點(diǎn)鑒別方法,其研究成果將會(huì)給城市道路安全治理中提供技術(shù)依據(jù)。
[Abstract]:Statistics have shown that traffic accidents have become a major problem that endangers human health and social development, and are also the first cause of death among people in China. Because of the characteristics of accident prone and frequent accidents, accident blackspot section is the target of priority treatment in road safety improvement project. However, the existing methods of accident blackspot identification require high integrity of accident data. It is difficult to implement in practical engineering applications and the identification results are mostly presented in the form of words and tables, and the visibility is poor. Therefore, the network kernel density estimation method is used to identify urban road network accident black spots. The kernel density estimation method is a non-parametric estimation theory in spatial analysis theory. The method does not presuppose the distribution of sample data and only needs to obtain the spatial location of the accident. And the discriminant result can be visualized based on GIS platform. In this paper, the influence of two important parameters in kernel density estimation, namely window width and kernel function, on the result of kernel density estimation is discussed, and then the application of kernel density estimation method to the identification of blackspots in urban road network accidents is briefly explained. The kernel density estimation method is modified to meet the identification requirements of accident blackspot sections under the constraints of urban road network. Then the brief flow of the application of network kernel density estimation method to the identification of urban road network accident blackspots is introduced. Then this paper verifies the validity of the modified kernel density estimation method applied to the blackspot identification of urban road network accidents, and the results show that compared with the existing traditional identification methods, The network kernel density estimation method is more effective and can be used in engineering applications. In this paper, ArcGIS Desktop software is applied to meet the visualization requirements of blackspot identification results of urban road network based on network kernel density estimation. The research contents include setting up the road traffic accident database, dividing the road network objects, calculating the network core density value of the unit node, and identifying the accident blackspot section based on the order of the unit network core density value. The black spot identification method of urban road network accidents based on network nuclear density is studied in this paper. The research results will provide technical basis for urban road safety management.
【學(xué)位授予單位】：東南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：U491.31

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 過秀成;盛玉剛;潘昭宇;潘敏榮;盧光明;何明;;公路交通事故黑點(diǎn)總體特征分析[J];東南大學(xué)學(xué)報(bào)(自然科學(xué)版);2007年05期

2 廖志高;柳本民;郭忠印;;基于信息分配的道路黑點(diǎn)鑒別方法[J];中國(guó)公路學(xué)報(bào);2007年04期

3 潘敏榮;過秀成;姜科;潘昭宇;;基于GIS的道路交通事故黑點(diǎn)分析處理系統(tǒng)研究[J];交通與計(jì)算機(jī);2007年02期

4 林冬云;劉慧平;;應(yīng)用空間聚類進(jìn)行點(diǎn)數(shù)據(jù)分布研究[J];北京師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2006年04期

5 裴玉龍,丁建梅;鑒別道路交通事故多發(fā)點(diǎn)的突出因素法[J];中國(guó)公路學(xué)報(bào);2005年03期

6 胡新民,劉濤,張?zhí)烊A,黃勇,謝海巍,于濤;道路黑點(diǎn)識(shí)別與改善[J];交通運(yùn)輸工程學(xué)報(bào);2004年01期

7 胡江碧,劉t，

本文編號(hào)：1821787