A Theoretical Guideline for Designing an Effective Adaptive Particle Swarm

被引:48
作者
Bonyadi, Mohammad Reza [1 ,2 ,3 ]
机构
[1] Rio Tinto, Data Sci & AI, Brisbane, Qld 4000, Australia
[2] Univ Queensland, Ctr Adv Imaging, Brisbane, Qld 4072, Australia
[3] Univ Adelaide, Optimisat & Logist Grp, Adelaide, SA 5005, Australia
关键词
Particle swarm optimization; Convergence; Acceleration; Linear programming; Guidelines; Correlation; Indexes; covariance; particle swarm optimization (PSO); stability; OPTIMIZATION ALGORITHM; CONVERGENCE ANALYSIS; STRATEGIES;
D O I
10.1109/TEVC.2019.2906894
中图分类号
TP18 [人工智能理论];
学科分类号
140502 [人工智能];
摘要
In this paper, the underlying assumptions that have been used for designing adaptive particle swarm optimization (PSO) algorithms in the past years are theoretically investigated. I relate these assumptions to the movement patterns of particles controlled by coefficient values (inertia weight and acceleration coefficients) and introduce three factors, namely the autocorrelation of the particle positions, the average movement distance of the particle in each iteration, and the focus of the search, that describe these movement patterns. I show how these factors represent movement patterns of a particle within a swarm and how they are affected by particle coefficients (i.e., inertia weight and acceleration coefficients). I derive equations that provide exact coefficient values to guarantee to achieve the desired movement pattern defined by these three factors within a swarm. I then relate these movements to the searching capability of particles and provide a guideline for designing potentially successful adaptive methods to control coefficients in particle swarm. Finally, I propose a new simple time adaptive particle swarm and compare its results with previous adaptive particle swarm approaches. Experiments show that the theoretical findings indeed provide a beneficial guideline for the successful adaptation of the coefficients in the PSO algorithm.
引用
收藏
页码:57 / 68
页数:12
相关论文
共 39 条
[1]
Impacts of Coefficients on Movement Patterns in the Particle Swarm Optimization Algorithm [J].
Bonyadi, Mohammad Reza ;
Michalewicz, Zbigniew .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2017, 21 (03) :378-390
[2]
Particle Swarm Optimization for Single Objective Continuous Space Problems: A Review [J].
Bonyadi, Mohammad Reza ;
Michalewicz, Zbigniew .
EVOLUTIONARY COMPUTATION, 2017, 25 (01) :1-54
[3]
Stability Analysis of the Particle Swarm Optimization Without Stagnation Assumption [J].
Bonyadi, Mohammad Reza ;
Michalewicz, Zbigniew .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2016, 20 (05) :814-819
[4]
Analysis of Stability, Local Convergence, and Transformation Sensitivity of a Variant of the Particle Swarm Optimization Algorithm [J].
Bonyadi, Mohammad Reza ;
Michalewicz, Zbigniew .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2016, 20 (03) :370-385
[5]
Nonlinear inertia weight variation for dynamic adaptation in particle swarm optimization [J].
Chatterjee, A ;
Siarry, P .
COMPUTERS & OPERATIONS RESEARCH, 2006, 33 (03) :859-871
[6]
Novel inertia weight strategies for particle swarm optimization [J].
Chauhan, Pinkey ;
Deep, Kusum ;
Pant, Millie .
MEMETIC COMPUTING, 2013, 5 (03) :229-251
[7]
Chen GM, 2006, WCICA 2006: SIXTH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION, VOLS 1-12, CONFERENCE PROCEEDINGS, P3672
[8]
Cleghorn C.W., 2016, Proceedings of the IEEE Symposium Series on Swarm Intelligence, P1
[9]
Particle swarm stability: a theoretical extension using the non-stagnate distribution assumption [J].
Cleghorn, Christopher W. ;
Engelbrecht, Andries P. .
SWARM INTELLIGENCE, 2018, 12 (01) :1-22
[10]
The particle swarm - Explosion, stability, and convergence in a multidimensional complex space [J].
Clerc, M ;
Kennedy, J .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 (01) :58-73