基于改进烟花优化算法的三维空间气体源定位
Localization of gas source in three-dimensional space based on improved fireworks optimization algorithm
冯崧 1曾祥进 1黄瑜豪1
作者信息
- 1. 武汉工程大学 计算机科学与工程学院,湖北 武汉 430205;湖北三峡实验室,湖北 宜昌 445804;武汉工程大学 荆门化工新材料产业技术研究院,湖北 荆门 448000
- 折叠
摘要
为探究三维空间中气体源定位及其源强反算问题,提出1 种改进烟花爆炸算法(GWOFA).将定位过程分为全局定位过程和局部定位过程.全局定位过程即结合灰狼优化算法和莱维飞行在三维空间中的全局搜索过程;局部定位过程是在全局定位的结果上的进一步开发过程,其通过引入边界条件的爆炸半径选取方式和选择策略更加高效地改进烟花优化算法实现.研究结果表明:本文算法相比于支持向量机回归模型(LinearSVR)、GWO算法和粒子群算法具有更高精确度,相比于GWO算法和粒子群算法具有更高稳定性和更低随机性;在气体源定位问题上,本文算法整体表现优于LinearSVR、GWO算法和粒子群算法.研究结果可为解决三维空间中气体源定位问题和相关参数估计问题提供新的思路方法.
Abstract
In order to investigate the problems of gas source localization and back-calculation of source intensity in the three-dimensional space,an improved fireworks explosion algorithm(GWOFA)was proposed.The localization process is divided into two stages:the global localization and the local localization.The global localization stage combines the grey wolf optimiza-tion(GWO)algorithm with the Lévy flight for global exploration in three-dimensional space.The local localization process re-fers to the further development process based on the results of global localization,which is achieved through the introduction of boundary conditions for selecting the explosion radius and a more efficient selection strategy through an improved fireworks optimization algorithm.The results show that the algorithm proposed in this paper has higher accuracy compared to support vector machine regression model(LinearSVR),GWO algorithm,and particle swarm algorithm,as well as higher stability and lower randomness compared to GWO algorithm and particle swarm algorithm.In terms of gas source localization,the overall performance of this algorithm is superior to LinearSVR,GWO algorithm,and particle swarm algorithm.The results can provide new ideas and methods for solving the problems of gas source localization and related parameters estimation in the three-di-mensional spaces.
关键词
气体源定位/烟花爆炸算法/莱维飞行/灰狼优化算法Key words
gas source localization/fireworks algorithm/Lévy flight/grey wolf optimization algorithm引用本文复制引用
基金项目
湖北省湖北三峡实验室创新基金(SC215001)
武汉工程大学荆门化工新材料产业技术研究院开放基金(JM2023006)
出版年
2024
NETL
NSTL
万方数据