Analyzing the Effect of Constraint-handling Methods on Artificial Bee Colony Algorithm

44928

Different modifications of the Artificial Bee Colony algorithm, which stands out with its performance among swarm intelligence algorithms, have been developed to solve the constrained optimization problems in which the design parameters are restricted by certain conditions and the optimum value should be located within the feasible region. In order to solve the problems by considering the constraints, there are methods based on penalty functions, methods based on preserving feasibility of solutions, methods which distinguish between feasible and infeasible solutions and hybrid methods. In this study, in order to solve the constrained optimization problems, new methods have been proposed by integrating penalty functions, stochastic ranking and stochastic Deb’s rules into the basic Artificial Bee Colony algorithm. The proposed methods have been tested on well-known constrained test problems in the literature, and the results have been compared with other state-of-the-art algorithms. The results indicate that proposed methods showed similar or better performance to solve constrained optimization problems when compared to the other algorithms considered in this study.

385

Artificial Bee Colony algorithm , Constrained optimization , constraint , handling methods , PEN , ABC , SR , ABC , DEB , SR , ABC

Erciyes University Journal Of The Institute Of Science an‎d Technology

399

Erciyes University Journal Of The Institute Of Science an‎d Technology