基于状态空间模型智能算法的公交优化调度

发布时间：2019-03-17 13:20

【摘要】：随着我国城市化进程的快速推进,机动车保有量呈急剧增长势态。然而,城市的综合服务能力及基础设施建设的效率却敢不上汽车保有量的增长速度,致使道路交通紧张,制约城市可持续发展。因此,大力发展城市交通系统成为当务之急。公交调度系统是交通系统的子系统之一,公交调度系统通过合理的分配人力、物力资源,以实现对公交车辆的最优化调度和管理。由于公交优化调度能带来良好的经济效益和社会效益,故国内外学者相继提出了很多优化方法来解决这一问题。公交优化调度问题属于多目标、多约束优化问题,建模和模型求解都存着困难。这一问题至今没有获得很好的解决方法,针对这一问题的研究仍在不断进行。本文提出了一种基于状态空间模型的智能算法来进行公交优化调度,该算法是采用实数编码方式且引入遗传算法基本理念的一种智能优化算法。利用状态进化矩阵的构造来指导算法搜索方向,突破了传统遗传算法的求解模式,使问题的求解过程可以用离散系统动力学过程表示。再通过选种池的优胜劣汰的选择机理来实现算法朝问题的最优解逼近。该算法操作简单,且寻优效果好。本文在考虑发车时间间隔、公交公司盈利和满载率约束的条件下,给出了兼顾公交公司和乘客利益的数学模型,该模型通过权重系数来反映主体利益倾向。将基于状态空间模型的智能算法运用到公交优化调度问题的设计中,并借助仿真平台,通过算例比较,验证了该算法在求解问题时的有效性。
[Abstract]:With the rapid development of urbanization in China, the number of motor vehicles is increasing rapidly. However, the city's comprehensive service capacity and the efficiency of infrastructure construction dare not exceed the growth rate of car ownership, which leads to the shortage of road traffic and restricts the sustainable development of the city. Therefore, it is urgent to develop the urban transportation system. The bus dispatching system is one of the subsystems of the traffic system. Through the rational allocation of human and material resources, the bus dispatching system can realize the optimal scheduling and management of public transport vehicles. Because the optimal dispatching of public transportation can bring good economic and social benefits, scholars at home and abroad have put forward a lot of optimization methods to solve this problem. Bus scheduling optimization problem belongs to multi-objective, multi-constraint optimization problem, modeling and model solving are difficult. So far, there is no good solution to this problem, and the research on this problem is still going on. In this paper, an intelligent algorithm based on state space model is proposed to optimize bus scheduling. This algorithm is an intelligent optimization algorithm based on real coding and introducing the basic concept of genetic algorithm (GA). The construction of the state evolution matrix is used to guide the search direction of the algorithm, which breaks through the solving mode of the traditional genetic algorithm and makes the solving process of the problem can be represented by the dynamic process of the discrete system. Then, the optimal solution of the problem is approximated by the selection mechanism of the survival of the fittest in the seed selection pool. The operation of the algorithm is simple and the optimization effect is good. In this paper, considering the constraints of departure time interval, profit and full load ratio of bus companies, a mathematical model considering the interests of both bus companies and passengers is presented. The model reflects the interest tendency of the main body through the weight coefficient. The intelligent algorithm based on state space model is applied to the design of bus optimal scheduling problem, and the effectiveness of the algorithm in solving the problem is verified by the comparison of numerical examples with the help of simulation platform.
【学位授予单位】：长沙理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491.17;TP18

【相似文献】