Information Technology Reference
In-Depth Information
A Hybrid Global Optimization Algorithm:
Particle Swarm Optimization
in Association with a Genetic Algorithm
M. Andalib Sahnehsaraei, M.J. Mahmoodabadi, M. Taherkhorsandi,
K.K. Castillo-Villar and S.M. Mortazavi Yazdi
Abstract The genetic algorithm (GA) is an evolutionary optimization algorithm
operating based upon reproduction, crossover and mutation. On the other hand,
particle swarm optimization (PSO) is a swarm intelligence algorithm functioning by
means of inertia weight, learning factors and the mutation probability based upon
fuzzy rules. In this paper, particle swarm optimization in association with genetic
algorithm optimization is utilized to gain the unique bene
ts of each optimization
algorithm. Therefore, the proposed hybrid algorithm makes use of the functions and
operations of both algorithms such as mutation, traditional or classical crossover,
multiple-crossover and the PSO formula. Selection of these operators is based on a
fuzzy probability. The performance of the hybrid algorithm in the case of solving
both single-objective and multi-objective optimization problems is evaluated by
utilizing challenging prominent benchmark problems including FON, ZDT1,
ZDT2, ZDT3, Sphere, Schwefel 2.22, Schwefel 1.2, Rosenbrock, Noise, Step,
Rastrigin, Griewank, Ackley and especially the design of the parameters of linear
feedback control for a parallel-double-inverted pendulum system which is a com-
plicated, nonlinear and unstable system. Obtained numerical results in comparison
to the outcomes of other optimization algorithms in the literature demonstrate the
Search WWH ::
Custom Search