粒子群優(yōu)化算法及其參數(shù)設(shè)置
專 業(yè)：信息與計算科學(xué)
學(xué) 生： xx
指導(dǎo)教師： 徐小平
摘 要
粒子群優(yōu)化是一種新興的基于群體智能的啟發(fā)式全局搜索算法，粒子群優(yōu)化算法通過粒子間的競爭和協(xié)作以實現(xiàn)在復(fù)雜搜索空間中尋找全局最優(yōu)點。它具有易理解、易實現(xiàn)、全局搜索能力強等特點，倍受科學(xué)與工程領(lǐng)域的廣泛關(guān)注，已經(jīng)成為發(fā)展最快的智能優(yōu)化算法之一。論文介紹了粒子群優(yōu)化算法的基本原理，分析了其特點。論文中圍繞粒子群優(yōu)化算法的原理、特點、參數(shù)設(shè)置與應(yīng)用等方面進行全面綜述，重點利用單因子方差分析方法，分析了粒群優(yōu)化算法中的慣性權(quán)值，加速因子的設(shè)置對算法基本性能的影響，給出算法中的經(jīng)驗參數(shù)設(shè)置。最后對其未來的研究提出了一些建議及研究方向的展望。
關(guān)鍵詞：粒子群優(yōu)化算法；參數(shù)；方差分析；最優(yōu)解














Particle swarm optimization algorithm and its parameter set
Speciality: Information and Computing Science
Student: Ren Kan
Advisor: Xu Xiaoping
Abstract
Particle swarm optimization is an emerging global based on swarm intelligence heuristic search algorithm, particle swarm optimization algorithm competition and collaboration between particles to achieve in complex search space to find the global optimum. It has easy to understand, easy to achieve, the characteristics of strong global search ability, and has never wide field of science and engineering concern, has become the fastest growing one of the intelligent optimization algorithms. This paper introduces the particle swarm optimization basic principles, and analyzes its features. Paper around the particle swarm optimization principles, characteristics, parameters settings and applications to conduct a thorough review, focusing on a single factor analysis of variance, analysis of the particle swarm optimization algorithm in the inertia weight, acceleration factor setting the basic properties of the algorithm the impact of the experience of the algorithm given parameter setting. Finally, its future researched and prospects are proposed.
Key word：Particle swarm optimization; Parameter; Variance analysis; Optimal solution






目 錄
摘 要 II
Abstract III
1.引言 1
1.1 研究背景和課題意義 1
1.2 參數(shù)的影響 1
1.3 應(yīng)用領(lǐng)域 2
1.4 電子資源 2
1.5 主要工作 2
2.基本粒子群算法 3
2.1 粒子群算法思想的起源 3
2.2 算法原理 4
2.3 基本粒子群算法流程 5
2.4 特點 6
2.5 帶慣性權(quán)重的粒子群算法 7
2.7 粒子群算法的研究現(xiàn)狀 8
3.粒子群優(yōu)化算法的改進策略 9
3.1 粒子群初始化 9
3.2 鄰域拓撲 9
3.3 混合策略 12
4.參數(shù)設(shè)置 14
4.1 對參數(shù)的仿真研究 14
4.2 測試仿真函數(shù) 15
4.3 應(yīng)用單因子方差分析參數(shù)對結(jié)果影響 33
4.4 對參數(shù)的理論分析 34
5結(jié)論與展望 39
致謝 43
附錄 44