并行化的多目标优化海缆路由规划算法研究  

Research on Parallel Multi-objective Optimal Submarine Cable Route Planning Algorithm

作　　者：蒋佳芮 赵赞善 段茂生 高冠军[1] 

机构地区：[1]北京邮电大学信息光子学与光通信国家重点实验室,北京100876 [2]中国科学院声学研究所南海研究站,海口570105 [3]陵水海洋信息海南省野外科学观测研究站,海南陵水572423

出　　处：《光通信研究》2025年第2期105-109,共5页Study on Optical Communications

基　　金：国家自然科学基金资助项目(62371064);国家重点研发计划资助项目(2022YFB2903303);北京市自然科学基金资助项目(4232050)。

摘　　要：【目的】文章为了解决传统蚁群优化(ACO)算法更新同一张地图导致无法并行规划的缺陷,提出了一种并行多目标优化海缆路由规划算法,实现了局部区域的精细规划。【方法】文章采用分治思想将目标海域的栅格地图分割成多个栅格子图,建立了并行化多目标优化海缆路由规划算法模型,并对模型关键参数进行优化,然后在最佳模型参数下,利用并行化蚁群优化(PACO)算法进行海底光缆路由规划,统计了Pareto前沿解下的海底光缆路由方案。【结果】仿真结果表明,并行多目标优化算法模型在分块数量为6,蚁群规模大小为150时,获得最佳的搜索能力和效率。PACO算法规划的海底光缆路由与传统ACO算法相比在相同风险条件下节省了33.9%的成本,且路由成本均小于传统ACO算法,路由最大成本与传统ACO算法的最小成本相比还降低了20.6%,同时相应的风险降低了65.8%。【结论】在多目标海底光缆路由规划中,与传统ACO算法相比,PACO算法不仅在规划结果上更优,而且运算时间效率提高至少8倍。【Objective】In order to solve the problem that the traditional Ant Colony Optimization(ACO)algorithm updates the same map,resulting in the inability of parallel planning,a parallel multi-objective optimization submarine cable route planning algorithm is proposed in this paper,which realizes the precise planning of local areas.【Methods】In this paper,the grid map of the target sea area is divided into multiple grid subgraphs by the idea of divide and conquer,and a parallel multi-objective optimization submarine cable route algorithm model is established,and the key parameters of the model are optimized.Then,the Parallel Ant Colony Optimization(PACO)algorithm is used to carry out the submarine cable route planning under the optimal model parameters,and the submarine cable route scheme solved by Pareto frontier is counted.【Results】The simulation results show that the parallel multi-objective optimization model obtains the best search ability and efficiency when the number of blocks is 6 and the size of ant colony is 150.The PACO algorithm can save 33.9%of the cost of submarine cable route compared with the traditional ACO algorithm under the same risk conditions,and the cost of routes is smaller than the traditional ant colony algorithm.The maximum cost of routes is also reduced by 20.6%compared with the minimum cost of the traditional ACO algorithm,and the coresponding risk is reduced by 65.8%.【Conclusion】In multi-objective submarine cable route planning,compared to the traditional ACO algorithm,the PACO algorithm not only achieves better planning results but also improves computational efficiency by at least 8 times.

关 键 词：海缆路由规划 并行蚁群优化算法 多目标优化 

分 类 号：TN929.11[电子电信—通信与信息系统]