Pursuit‐Escape Particle Swarm Optimization
Pursuit‐Escape Particle Swarm Optimization
复制标题
DOI：
10.1002/tee.20245
复制
发表时间：
2008-01
影响因子：
1
通讯作者：
Mitsuharu Higashitani;A. Ishigame;K. Yasuda
中科院分区：
文献类型：
--
作者：
Mitsuharu Higashitani;A. Ishigame;K. Yasuda
提出了一种新的具有追赶和逃逸行为的粒子群优化算法。这种方法从沙丁鱼群和虎鲸群的行为中得到了启示。当沙丁鱼受到虎鲸的攻击时，它们会表现得异常的，那就是沙丁鱼会从虎鲸的手中逃脱，而另一方面，虎鲸会追击沙丁鱼。通过这种方法，粒子被分为两类，称为追逐粒子和逃逸粒子，它们相互作用。它们分别发挥着集约化和多样化的关键作用。这使得粒子能够避免局部最优解并找到全局最优解，并且在搜索过程中在多样化（全局搜索）和强化（局部搜索）之间实现适当的平衡。然后，通过使用几个众所周知的优化基准问题的函数进行数值模拟来验证所提出的方法，并将其与强大的方法（如SAPPO，LDIWM和CFM）进行比较。Copyright © 2007日本电气工程师协会。由John Wiley & Sons公司出版
This paper presents a new Particle Swarm Optimization (PSO) with pursuit and escape behavior. This method takes a cue from the behaviors of schools of sardines and pods of killer whales. When the sardines are attacked by the killer whales, they would behave unusually, that is, the sardines would escape from the killer whales, and on another front, the killer whales would pursue the sardines. By this method, particles are divided into two categories called the pursuit‐particles and the escape‐particles, having interactions with each other. They play the key roles of intensification and diversification, respectively. This allows the particles to avoid local optimal solutions and find a global optimal one, and also achieve an appropriate balance between diversification (global search) and intensification (local search) during the search. Then, the proposed method is validated through numerical simulations using several functions which are well‐known as the optimization benchmark problems by comparing them to powerful methods such as SAPPO, LDIWM, and CFM. Copyright © 2007 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.