论文部分内容阅读
Anthony Chen(2002)提出的基于起讫点的Frank-Wolfe(ODBFW)算法,对同一起点的所有OD对搜索最短路树,采用one-OD-at-a-time流量更新策略,是一种收敛性能较好的均衡交通分配算法.主要对ODBFW算法的初始化、线性搜索及最短路搜索算法进行改进,提出ODBFW改进算法.将ODBFW改进算法应用于国际上同类研究常用的Sioux Falls网络及我国山东省诸城市实际道路网,比较标准FW算法、ODBFW算法及ODBFW改进算法的可行性、效率及稳定性.结果显示,改进算法效率较标准FW算法有显著提高,稳定性较好.在小型测试路网上改进算法较ODBFW算法达到收敛的时间减少近15%,在大型测试路网上减少近5%.
Anthony Chen (2002) put forward the point-based Frank-Wolfe (ODBFW) algorithm, which uses a one-OD-at-a-time traffic update strategy for searching all shortest-path trees for all OD pairs in the same starting point. Better balanced traffic assignment algorithm.The main improvement of the ODBFW algorithm initialization, linear search and shortest path search algorithm, proposed ODBFW improved algorithm.Application of improved ODBFW algorithm to similar international research Sioux Falls network and our country Zhucheng, Shandong Province The actual road network, the comparison of the standard FW algorithm, the ODBFW algorithm and the improved ODBFW algorithm are presented.The results show that the efficiency of the improved algorithm is significantly improved and the stability is better than that of the standard FW algorithm.Improvement on the small test road network Compared with ODBFW algorithm, the algorithm can reduce the time of convergence by nearly 15% and reduce by nearly 5% in the large test road network.