-
摘要: 为了通过路段检测交通流量计算拥挤条件下多种交通模式需求, 提出了一个随机用户平衡条件下的多模式路径流量估计模型, 并给出了相应模型的增广拉格朗日乘子算法, 算法将模型中的路段容量、观测路段流量平衡与估计需求的范围等约束条件转化为相应的惩罚函数项, 并将原先的有约束优化流量估计模型转化为一个无约束优化模型, 最后应用一个简单的投影迭代算法求解无约束优化模型。仿真结果表明: 先验需求误差对模型的需求估计结果有重要影响, 误差越小估计结果越准确, 而先验需求误差对路段流量估计结果几乎没有影响, 因此, 模型和算法简单可用。Abstract: In order to calculate multi-mode traffic demand from traffic counts under congested conditions, a multi-mode path flow estimation model with stochastic user equilibrium was presented, and an augmented Lagrange dual algorithm was given. Link capacity, flow equilibrium and priori demand constraints were transformed into penalty functions. The original constraint optimal flow estimation model was transformed into non-constraint optimal model, and a simple projection algorithm was adopted for non-constraint optimal model. Simulation result shows that the smaller the priori demand error is, the better OD estimation results are; the priori demand error has little influence on link flow estimation result, therefore, the estination model and algorithm are simple and applicable.
-
[1] 安梅, 高自友. OD需求估计的多模式双层规划模型新算法[J]. 公路交通科技, 1999, 16 (4): 56-60. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK199904014.htmAN Mei, GAO Zi-you. A new algorithmfor the interaction multi-modal traffic origin-destination demands esti mation[J]. Journal of Highway and Transportation Research and Devel-opment, 1999, 16 (4): 56-60. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK199904014.htm [2] 安梅, 高自友. 模式间相互影响时估计OD需求的双层规划模型及求解算法[J]. 系统工程理论与实践, 2001, 21 (4): 100-105. https://www.cnki.com.cn/Article/CJFDTOTAL-XTLL200104008.htmAN Mei, GAOZi-you. Bi-level programming model and algo-rithmfor the interaction multi-modal traffic origin-destination demands esti mation[J]. Systems Engineering Theory and Practice, 2001, 21 (4): 100-105. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XTLL200104008.htm [3] SEUNGKIRL B, HYUNMYUNG K, YONGTAEK L. Mul-tiple-vehicle origin-destination matrix esti mation fromtraffic counts using genetic algorithm[J]. Journal of Transportation Engineering, 2004, 130 (3): 339-347. doi: 10.1061/(ASCE)0733-947X(2004)130:3(339) [4] WONG S C, TONG C O. Esti mation of multi-class origin-destination matrices fromtraffic counts[J]. Journal of Urban Planning and Development, 2005, 131 (1): 19-29. doi: 10.1061/(ASCE)0733-9488(2005)131:1(19) [5] NIE Y, ZHANG H M. Inferring origin-destination trip ma-trices with a decoupled GLS path flow esti mator[J]. Trans-portation Research Part B, 2005, 39 (3): 497-518. [6] BELL MG H, SHIELD C M, BUSCHF, et al. Astochastic user equilibrium path flow esti mator[J]. Transportation Research Part C, 1997, 5 (1): 197-210. [7] 任刚, 王炜. 带转向延误的非对称多模式用户平衡模型及算法[J]. 中国公路学报, 2006, 19 (2): 80-85. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL200602013.htmREN Gang, WANG Wei. Model and algorithmfor asymmet-ric multi modal user equilibrium with turn delays[J]. China Journal of Highway and Transport, 2006, 19 (2): 80-85. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL200602013.htm [8] 王选仓, 于江霞, 王秉纲, 等. 基于需求函数模型的公路网规模预测[J]. 长安大学学报: 自然科学版, 2006, 26 (3): 59-62. https://www.cnki.com.cn/Article/CJFDTOTAL-XAGL200603014.htmWANG Xuan-cang, YUJiang-xia, WANG Bing-gang, et al. Reasonable highway network scale prediction based on demand function[J]. Journal of Chang an University: Natural Science Edition, 2006, 26 (3): 59-62. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XAGL200603014.htm [9] LARSSON T, PATRIKSSON M. An augmented lagrangian dual algorithm link capacity side constrained traffic assign-ment problems[J]. Transportation Research Part B, 1995, 29 (2): 433-455. [10] NIE Y, ZHANG H M. Models and algorithms for the traffic assignment problem with link capacity constraints[J]. Transportation Research Part B, 2004, 38 (2): 285-312. -
下载:
点击查看大图
图(6)
计量
- 文章访问数: 334
- HTML全文浏览量: 144
- PDF下载量: 852
- 被引次数: 0
