In this article we describe a Particle Swarm Optimization (PSO) approach to handling constraints in Multi-objective Optimization (MOO). The method is called Constrained Adaptive Multi-objective Particle Swarm Optimization (CAMOPSO). CAMOPSO is based on the Adaptive Multi-objective Particle Swarm Optimization (AMOPSO) method proposed in [1]. As in AMOPSO, the inertia and the acceleration coefficients are determined adaptively in CAMOPSO, while a penalty based approach is used for handling constraints. In this article, we first review some existing MOO approaches based on PSO, and then describe the AMOPSO method in detail along with experimental results on six unconstrained MOO problems [1]. Thereafter, the way of handling constraints in CAMOPSO is discussed. Its performance has been compared with that of the NSGA-II algorithm, which has an inherent approach for handling constraints. The results demonstrate the effectiveness of CAMOPSO for the test problems considered.
CITATION STYLE
Hiot, L. M., Ong, Y. S., Panigrahi, B. K., Shi, Y., Lim, M.-H., Tripathi, P. K., … Pal, S. K. (2010). Handbook of Swarm Intelligence. Springer (Vol. 8, pp. 221-239–239). Retrieved from http://www.springerlink.com/content/r4k7w76467163180/
Mendeley helps you to discover research relevant for your work.