Opposition-Based Hybrid Strategy for Particle Swarm Optimization in Noisy Environments

Qi Kang,Mengchu Zhou,Lingpeng Meng,Caifei Xiong

doi:10.1109/access.2018.2809457

Abstract

Particle swarm optimization (PSO) is a population-based algorithm designed to tackle various optimization problems. However, its performance deteriorates significantly when optimization problems are subjected to noise. PSO is strongly influenced by its previous best particles and global best one, which may lead to premature convergence and fall into local optima. This also holds true for various PSO variants dealing with optimization problems in noisy environments. Opposition-based learning (OBL) is well-known for its ability to increase population diversity. In this paper, we propose hybrid PSO algorithms that introduce OBL into PSO variants for improving the latter’s performance. The proposed hybrid algorithms employ probabilistic OBL for a swarm. In contrast to other integrations of PSO and OBL, we select the top fittest particles from the current swarm and its opposite swarm to improve the entire swarm’s fitness. Experiments on 20 benchmark functions subject to different levels of noise show that the proposed hybrid PSO algorithms outperform their counterpart PSO variants as well as composite differential evolution in most cases.

Highlights

Particle Swarm Optimization (PSO) [1] is a stochastic optimization algorithm inspired by the behaviors of birds and fish in order to find desired solutions to complex optimization problems
The same conclusion is applicable to composite DE (CoDE) and CoDE
Because of external factors such as imprecise measurements and communication/computing errors, many real-life optimization problems are subject to various noise

Summary

Introduction

Particle Swarm Optimization (PSO) [1] is a stochastic optimization algorithm inspired by the behaviors of birds and fish in order to find desired solutions to complex optimization problems. It has shown excellent performance in many problems [2]–[11] and achieved the performance comparable to other evolutionary algorithms’ [12]. Realworld optimization problems subject to noise are commonly responsible for inaccurate and uncertain information such as deviations and measurement errors, which deteriorate the performance of PSO significantly [13] In this type of problems, the true objective function values of solutions are disturbed by noise. Recent studies [13] on large-scale optimization problems subject to noise have revealed the necessity of explicitly addressing the noise issue rather than relying on PSO itself to mitigate it

Methods

Results

Conclusion