基于改进粒子群算法的海上遇险目标搜寻方法  

Searching Method for Maritime Distress Targets Based on Improved Particle Swarm Algorithm

作　　者：孔祥凤 王海红[1] 李盛威 黄伟 KONG Xiangfeng;WANG Haihong;LI Shengwei;HUANG Wei(College of Information Science and Technology,Qingdao University of Science and Technology,Qingdao 266061,China)

机构地区：[1]青岛科技大学信息科学技术学院,山东青岛266061

出　　处：《计算机测量与控制》2025年第3期183-189,共7页Computer Measurement &Control

基　　金：国家自然科学基金项目(61104004,61170258);山东省自然科学基金重大基础研究项目(ZR2021ZD12)。

摘　　要：针对海上遇险目标搜寻范围动态化、影响因素众多导致搜救成功率较低的问题,提出了一种基于改进粒子群算法的海上遇险目标搜寻方法,旨在寻找最佳搜寻路径,提高海上遇险目标的搜救成功率;该方法基于遇险目标的位置信息和搜寻资源参数,构建海上遇险目标搜寻模型,并采用余弦曲线自适应方法改进算法的惯性权重系数,增强粒子群算法的初期全局搜索和后期局部搜索能力;采用自适应策略调整加速度,并保持其总和不变,以避免搜索效率下降或不稳定;引入扰动粒子更新机制来保持种群的多样性,避免陷入局部最优;将改进算法应用于实际搜寻问题,验证了算法的有效性,将改进算法与传统的粒子群算法和遗传算法进行对比,结果表明,改进算法较传统粒子群算法和遗传算法具有更高的搜救成功率。In response to the problems of dynamic search ranges and numerous influencing factors,which leads to lower success rates in search and rescue operations for maritime distress targets,an improved method based on improved particle swarm algorithm is proposed.The method aims to identify the optimal search path to enhance the success rate of rescuing maritime distress targets.Based on the location information of distress targets and parameters of search resources,a searching maritime distress target model is constructed.A cosine curve adaptive method is used to improve the inertia weight coefficient,and enhance the initial global search and later local search capabilities of the particle swarm algorithm.Moreover,to avoid a decrease in search efficiency or instability,an adaptive strategy is employed to adjust acceleration while maintaining the total unchanged.The perturbation particle update mechanism is introduced to maintain population diversity and avoid local optimal solution.The improved algorithm is applied in practical search scenarios,which verifies the effectiveness of the proposed algorithm.The results show that the improved algorithm has a higher success rate in search and rescue operations compared to traditional particle swarm and genetic algorithms.

关 键 词：改进粒子群算法 海上搜寻 自适应 全局最优 惯性权重 扰动粒子 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]