摘要
为发挥轨道交通的骨干运输功能,优化综合公共交通出行网络,建立了基于既有轨道交通路网的地面道路公交调整双层规划模型。上层规划考虑综合公交网络客运量及轨道交通客运周转量最大化、总社会出行成本及车辆配备成本最小化共4个目标,下层模型以效用理论为基础构建弹性需求下的线路流量分配,其中地面道路公交线路的出行效用考虑了道路通行能力的随机性。采用蒙特卡罗模拟求解下层模型,并采用基于向量的多目标粒子群算法求解整体双层规划,获得最佳调整线路的走向和发车间隔。最后通过算例验证了模型和算法。计算结果表明:所得到的解为一组Pareto解,4个目标间存在相悖关系。在实际问题中,应结合改造成本和现实需要来选取最优解作为调整方案。
Abstract In order to ensure the leading role of railway and optimize the overall public transportation network, a bi-lev- el programming model for bus line adjustment is established based on the existing railway network. The upper planning is aimed to maximize the whole transit network throughput volume, railway turnover volume and minimize the cost of total social travel and equipped vehicles. The lower model is to develop the flow distribution under elastic demand based on utility theory, and in the bus lines＇ travel utility the ran- domness of road capacity us specially considered. Monte Carlo simulation is used to solve the lower model, Vector e- valuated particle swarm optimization （VEPSO） is used to solve the whole bi-level programming, to obtain the adjusted bus line route and departure intervals. At the end, an ex- ample is adopted to test the model and algorithm, the results show that the solution is a group of Pareto solutions, four objections are contradicted. The optimum scheme for network design should be chosen by combining with cost and actual situations.
出处
《城市轨道交通研究》
北大核心
2016年第8期25-30,共6页
Urban Mass Transit
基金
国家自然科学基金项目(51308425)
教育部博士后基金项目(2014M561762)
大学生创新创业训练计划项目(201510488037)
关键词
公交调整
双层规划
蒙特卡罗模拟
多目标粒子
群算法
bus line adjustment
bi-level programming
Monte Carlo simulation
muti-objection particle swarm op-timization
