一种改进的粒子群算法在交通分配上的应用  

Application of an Improved Particle Swarm Algorithm in Traffic Assignment

作　　者：李晓君 赵晓蕾 赵洪銮[1,3] 宿梦梦 邹炜 LI Xiao-jun;ZHAO Xiao-lei;ZHAO Hong-luan;SU Meng-meng;ZOU Wei(School of Computer Science and Technology,Shandong Jianzhu University,Jinan 250101,China;School of Architecture and Urban Planning,Shandong Jianzhu University,Jinan 250101,China;School of Science,Tianjin Chengjian University,Tianjin 300384,China)

机构地区：[1]山东建筑大学计算机科学与技术学院,山东济南250101 [2]山东建筑大学建筑城规学院,山东济南250101 [3]天津城建大学理学院,天津300384

出　　处：《计算机技术与发展》2023年第4期140-145,共6页Computer Technology and Development

基　　金：山东省专业学位研究生教学案例库(SDYAL20157)。

摘　　要：针对粒子群算法收敛速度慢、求解精度低和算法在迭代后期容易陷入局部最优的问题,首先,采用仅以位置项来控制粒子进化方向的简化粒子群算法,以此避免粒子速度过大时导致的粒子发散的现象;其次,引入随迭代次数增加自适应改变的线性惯性权重来消除惯性分量的影响,同时引入莱维飞行策略来改变粒子位置以帮助粒子逃离局部最优;最后,通过四种测试函数对固定权重的粒子群算法、标准粒子群算法和改进算法的性能进行比较。实验证明,改进后的算法在收敛速度、精度和稳定性上都有所提升。在验证了改进算法的有效性后,使用改进后的算法求解单一OD对多路径路网的用户最优模型并与标准粒子群算法求解结果进行对比,改进后的算法求解结果更加稳定均衡,验证了算法的可行性。Aiming at the problems of slow convergence speed,low solution accuracy,and easy to fall into local optimum in the later iteration of particle swarm optimization,we firstly adopt a simplified particle swarm optimization algorithm that only uses the position term to control the evolution direction of particles,so as to avoid the problem of particle divergence caused by excessive particle speed.Secondly,the linear inertia weight that adaptively changes with the increase of the number of iterations is introduced to eliminate the influence of the inertial component,and the Levy flight strategy is introduced to change the particle position to help the particle escape from the local optimum.Finally,through four test function,we compare the performance of fixed-weight particle swarm optimization,standard particle swarm optimization and improved algorithms.Experiments show that the improved algorithm has improved convergence speed,accuracy and stability.After verifying the effectiveness of the improved algorithm,the improved algorithm is used to solve the user equilibrium model of a single OD to a multi-path road network and compared with the results of the standard particle swarm optimization algorithm.The results of the improved algorithm are more stable and balanced.The feasibility of the improved algorithm is verified.

关 键 词：简化的粒子群算法 非线性递减惯性权重 莱维飞行 单一OD对多路径路网 用户最优模型 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]