基于Paramics仿真平臺的交通誘導策略評估研究
發(fā)布時間：2018-12-10 17:13
【摘要】：摘要：隨著城市交通問題的日益突出,動態(tài)交通誘導系統(tǒng)作為智能交通系統(tǒng)的重要組成部分,日漸成為目前解決城市交通問題的重要手段。通過交通誘導引導出行者的路徑選擇行為可以發(fā)揮道路交通基礎設施的最大效能,對均衡路網(wǎng)交通流、減少延誤、提高交通系統(tǒng)的運行效率具有重要意義。在眾多理論研究層面的誘導策略中,采用何種誘導策略更適合現(xiàn)實的交通環(huán)境值得關(guān)注,而通過微觀交通仿真模型對誘導策略進行驗證評估,可以將理論研究向?qū)嵺`推進,具有一定的現(xiàn)實意義。本文利用Paramics微觀仿真平臺對多種動態(tài)誘導策略進行仿真實現(xiàn)并評估分析,以期實現(xiàn)理論研究向?qū)嶋H應用的過渡。本文的主要工作有： (1)對各種微觀仿真模型進行對比,結(jié)果表明Paramics微觀仿真軟件能較好的實現(xiàn)動態(tài)交通誘導功能,在此基礎上詳細研究了Paramics軟件中各模塊的主要功能及其高級程序接口API。 (2)重點研究了動態(tài)誘導策略在Paramics微觀仿真平臺上的實現(xiàn)方法,建立了兩條路徑的簡單路網(wǎng)模型,設置合理的參數(shù)并加載車流,在C++語言環(huán)境中利用Paramics豐富的API函數(shù)編寫誘導策略插件實現(xiàn)誘導策略。 (3)在不同交通需求場景下利用Paramics仿真實現(xiàn)了三種經(jīng)典反饋型誘導控制策略,即Bang-bang、P和PI控制。從系統(tǒng)和用戶兩種角度考慮,選擇了平均旅行時間和路徑旅行時間偏差作為評價指標,對輸出數(shù)據(jù)進行處理分析后比較了三種反饋策略的誘導效果,得出P誘導策略的效果最好,P1次之,而Bang-bang最差。 (4)提出了離散化色彩類VMS誘導策略,出行者由誘導信息的被動執(zhí)行者變?yōu)橹鲃記Q策者,根據(jù)對圖形式VMS誘導信息進行速度感知(服從均勻分布和三角分布)做出路徑選擇決策。通過Logit離散選擇模型,實現(xiàn)了兩種感知分布對應的誘導策略的Paramics仿真,以平均旅行時間和路徑旅行時間偏差作為評價指標,得出交通需求、誘導周期、參數(shù)對兩種感知分布對應的誘導策略均有不同影響。
[Abstract]:Abstract: with the increasingly prominent urban traffic problems, dynamic traffic guidance system, as an important part of intelligent transportation system, has become an important means to solve urban traffic problems. The route selection behavior guided by traffic guidance can give full play to the maximum efficiency of road traffic infrastructure, which is of great significance to balance the traffic flow, reduce delays and improve the efficiency of traffic system. Among the many guidance strategies at the theoretical level, it is worth paying attention to which one is more suitable for the real traffic environment. However, through the microscopic traffic simulation model to verify and evaluate the guidance strategy, the theoretical research can be promoted to practice. Has certain realistic significance. In this paper, a variety of dynamic induction strategies are simulated and analyzed based on Paramics micro-simulation platform, in order to realize the transition from theoretical research to practical application. The main work of this paper is as follows: (1) comparing various microscopic simulation models, the results show that the Paramics micro-simulation software can achieve the function of dynamic traffic guidance. On this basis, the main functions of each module in Paramics software and its advanced program interface API. are studied in detail. (2) the realization method of dynamic induction strategy on Paramics microcosmic simulation platform is studied, and a simple road network model with two paths is established, and reasonable parameters are set up and the traffic flow is loaded. In C language environment, the rich API function of Paramics is used to write the inductive policy plug-in to realize the induction policy. (3) under different traffic demand scenarios, three kinds of classical feedback induction control strategies, namely Bang-bang,P and PI control, are realized by using Paramics simulation. Considering the system and the user, the average travel time and the path travel time deviation are selected as the evaluation indexes. After processing and analyzing the output data, the inductive effects of the three feedback strategies are compared. The results showed that P induction strategy was the best, P1 was the second, and Bang-bang was the worst. (4) A discrete color class VMS induction strategy is proposed, in which the traveller changes from the passive executor of the induced information to the active decision maker. The path selection decision is made according to the velocity perception (uniform distribution and triangular distribution) of VMS induced information in graph form. Through the Logit discrete selection model, the Paramics simulation of two inductive strategies corresponding to the perceptual distribution is realized. The average travel time and the path travel time deviation are taken as the evaluation indexes, and the traffic demand and the induction period are obtained. Parameters have different effects on the induction strategies corresponding to the two perceptual distributions.
【學位授予單位】：北京交通大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：U495
本文編號：2370931
[Abstract]:Abstract: with the increasingly prominent urban traffic problems, dynamic traffic guidance system, as an important part of intelligent transportation system, has become an important means to solve urban traffic problems. The route selection behavior guided by traffic guidance can give full play to the maximum efficiency of road traffic infrastructure, which is of great significance to balance the traffic flow, reduce delays and improve the efficiency of traffic system. Among the many guidance strategies at the theoretical level, it is worth paying attention to which one is more suitable for the real traffic environment. However, through the microscopic traffic simulation model to verify and evaluate the guidance strategy, the theoretical research can be promoted to practice. Has certain realistic significance. In this paper, a variety of dynamic induction strategies are simulated and analyzed based on Paramics micro-simulation platform, in order to realize the transition from theoretical research to practical application. The main work of this paper is as follows: (1) comparing various microscopic simulation models, the results show that the Paramics micro-simulation software can achieve the function of dynamic traffic guidance. On this basis, the main functions of each module in Paramics software and its advanced program interface API. are studied in detail. (2) the realization method of dynamic induction strategy on Paramics microcosmic simulation platform is studied, and a simple road network model with two paths is established, and reasonable parameters are set up and the traffic flow is loaded. In C language environment, the rich API function of Paramics is used to write the inductive policy plug-in to realize the induction policy. (3) under different traffic demand scenarios, three kinds of classical feedback induction control strategies, namely Bang-bang,P and PI control, are realized by using Paramics simulation. Considering the system and the user, the average travel time and the path travel time deviation are selected as the evaluation indexes. After processing and analyzing the output data, the inductive effects of the three feedback strategies are compared. The results showed that P induction strategy was the best, P1 was the second, and Bang-bang was the worst. (4) A discrete color class VMS induction strategy is proposed, in which the traveller changes from the passive executor of the induced information to the active decision maker. The path selection decision is made according to the velocity perception (uniform distribution and triangular distribution) of VMS induced information in graph form. Through the Logit discrete selection model, the Paramics simulation of two inductive strategies corresponding to the perceptual distribution is realized. The average travel time and the path travel time deviation are taken as the evaluation indexes, and the traffic demand and the induction period are obtained. Parameters have different effects on the induction strategies corresponding to the two perceptual distributions.
【學位授予單位】：北京交通大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：U495
【參考文獻】
相關(guān)期刊論文 前10條
1 干宏程;;VMS誘導信息影響下的路徑選擇行為分析[J];系統(tǒng)工程;2008年03期
2 鄭為中,史其信,李德};應用PARAMICS微觀交通仿真對交通分流計劃的比較評價[J];公路交通科技;2004年11期
3 趙曉華;陳陽舟;石建軍;高俊俠;;交通控制算法在Paramics交通仿真軟件中的實現(xiàn)[J];公路交通科技;2006年06期
4 賀正冰;關(guān)偉;;考慮信號交叉口影響的分散路徑誘導策略[J];北京工業(yè)大學學報;2013年10期
5 吳文祥,黃海軍;平行路徑網(wǎng)絡中信息對交通行為的影響研究[J];管理科學學報;2003年02期
6 姜桂艷;鄭祖舵;白竹;代磊磊;趙佳琪;;擁擠條件下可變信息板交通誘導信息對駕駛行為的影響[J];吉林大學學報(工學版);2006年02期
7 李振龍;趙曉華;;基于模糊聚類的快速路VMS信息發(fā)布方法[J];計算機工程;2008年08期
8 閆友彪,陳元琰;機器學習的主要策略綜述[J];計算機應用研究;2004年07期
9 王笑京;智能運輸系統(tǒng)體系框架研究[J];交通與計算機;2001年04期
10 胡明偉,郭秀芝;用微觀交通仿真軟件實現(xiàn)ITS模擬的比較研究[J];交通與計算機;2004年04期
，本文編號：2370931
本文鏈接：http://sikaile.net/kejilunwen/jiaotonggongchenglunwen/2370931.html
教材專著
