Analyses of guide update approaches for vector evaluated particle swarm optimisation on dynamic multi-objective optimisation problems

The vector evaluated particle swarm optimisation (VEPSO) algorithm is a multi-swarm variation of particle swarm optimisation (PSO) used to solve static multi-objective optimisation problems (SMOOPs). Recently, VEPSO was extended to the dynamic VEPSO (DVEPSO) algorithm to solve dynamic multi-objective optimisation problems (DMOOPs) that have at least one objective that changes over time. The search process of DVEPSO is driven through local and global guides that can be updated in various ways. This paper investigates the influence of various guide update approaches on the performance of DVEPSO. DVEPSO is also compared against a competitive-cooperative evolutionary algorithm. The results indicate that DVEPSO performs well in fast changing environments, but struggles to converge to discontinuous Pareto-optimal fronts (POFs).

[1]  Michael N. Vrahatis,et al.  Recent approaches to global optimization problems through Particle Swarm Optimization , 2002, Natural Computing.

[2]  Kalyanmoy Deb,et al.  Dynamic multiobjective optimization problems: test cases, approximations, and applications , 2004, IEEE Transactions on Evolutionary Computation.

[3]  Gerry Dozier,et al.  Adapting Particle Swarm Optimizationto Dynamic Environments , 2001 .

[4]  Andries Petrus Engelbrecht,et al.  CIlib: A collaborative framework for Computational Intelligence algorithms - Part I , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[5]  Andries Petrus Engelbrecht,et al.  CIlib: A collaborative framework for Computational Intelligence algorithms - Part II , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[6]  Frans van den Bergh,et al.  An analysis of particle swarm optimizers , 2002 .

[7]  Kay Chen Tan,et al.  A predictive gradient strategy for multiobjective evolutionary algorithms in a fast changing environment , 2010, Memetic Comput..

[8]  Andries Petrus Engelbrecht,et al.  Solving dynamic multi-objective problems with vector evaluated particle swarm optimisation , 2008, 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence).

[9]  Andries Petrus Engelbrecht,et al.  Dynamic Multi-objective Optimisation Using PSO , 2010, Multi-Objective Swarm Intelligent System.

[10]  Konstantinos E. Parsopoulos,et al.  MULTIOBJECTIVE OPTIMIZATION USING PARALLEL VECTOR EVALUATED PARTICLE SWARM OPTIMIZATION , 2003 .

[11]  Andries Petrus Engelbrecht,et al.  Archive management for dynamic multi-objective optimisation problems using vector evaluated particle swarm optimisation , 2011, 2011 IEEE Congress of Evolutionary Computation (CEC).

[12]  Mario Cámara Sola,et al.  Parallel processing for dynamic multi-objetive optimization , 2010 .

[13]  Kay Chen Tan,et al.  A Competitive-Cooperative Coevolutionary Paradigm for Dynamic Multiobjective Optimization , 2009, IEEE Transactions on Evolutionary Computation.

[14]  Julio Ortega Lopera,et al.  Approaching Dynamic Multi-Objective Optimization Problems by Using Parallel Evolutionary Algorithms , 2010, Advances in Multi-Objective Nature Inspired Computing.