Handling Multiple Objectives with Integration of Particle Swarm Optimization and Extremal Optimization
In recent years, many hybrid metaheuristic approaches have been proposed to solve multiobjective optimization problems (MOPs). In this paper, we present a novel multiobjective algorithm, so-called MOPSOEO, which combines particle swarm optimization (PSO)
- PDF / 459,361 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 47 Downloads / 159 Views
Abstract In recent years, many hybrid metaheuristic approaches have been proposed to solve multiobjective optimization problems (MOPs). In this paper, we present a novel multiobjective algorithm, so-called MOPSOEO, which combines particle swarm optimization (PSO) with extremal optimization (EO) to solve MOPs. The hybrid approach takes full advantage of the exploration ability of PSO and the exploitation ability of EO, which can overcome the premature convergence of PSO when it is applied to MOPs. The proposed approach is validated by using five benchmark functions and metrics taken from the standard literature on evolutionary multiobjective optimization. Experimental results indicate that the approach is highly competitive with the state-of-the-art evolutionary multiobjective algorithms, and thus, MOPSOEO can be considered a viable alternative to solve MOPs.
Keywords Multiobjective optimization Particle swarm optimization Extremal optimization
M.-R. Chen (&) School of Computer Science, South China Normal University, Guangzhou 510631, China e-mail: [email protected] M.-R. Chen X. Li College of Information Engineering, Shenzhen University, Shenzhen 518060, China e-mail: [email protected] J. Weng Department of Computer Science, Jinan University, Guangzhou 510632, China e-mail: [email protected] X. Zhang College of Computer and Software, Shenzhen University, Shenzhen 518060, China e-mail: [email protected]
Z. Wen and T. Li (eds.), Foundations of Intelligent Systems, Advances in Intelligent Systems and Computing 277, DOI: 10.1007/978-3-642-54924-3_27, Springer-Verlag Berlin Heidelberg 2014
287
288
M.-R. Chen et al.
1 Introduction Most real-world engineering optimization problems are multiobjective in nature, since they normally have several (possibly conflicting) objectives that must be satisfied at the same time. Instead of aiming at finding a single solution, the multiobjective optimization methods try to produce a set of good trade-off solutions from which the decision maker may select one. Evolutionary algorithms (EAs) seem particularly suitable to solve multiobjective optimization problems (MOPs), because they deal simultaneously with a set of possible solutions. This allows us to find several members of the Pareto-optimal set in a single run of the algorithm [1]. Additionally, EAs are less susceptible to the shape or continuity of the Pareto front. Over the past two decades, a considerable amount of multiobjective evolutionary algorithms (MOEAs) have been presented to solve various types of MOPs [1, 2]. Particle swarm optimization (PSO) [3] is a very popular EA during the past decade, and it has been extended to solve MOPs [4, 5]. However, the PSO has been found to be easily trapped into local minima when solving a single objective or multiple objectives. To overcome the limitation of PSO when solving MOPs, we develop a novel multiobjective PSO algorithm hybridized with extremal optimization (EO) [6, 7], called MOPSOEO. The proposed hybrid algorithm perfectly combines the exploration ability of PSO
Data Loading...
