FAGA: hybridization of fractional order ABC and GA for optimization

In order to solve problems of optimization, Swarm Intelligence (SI) algorithms are extensively becoming more popular. Many swarm intelligence based optimization techniques are present but most face problems like convergence problem and local minimization problem. In this paper, a hybrid optimization algorithm is proposed using fractional order Artificial Bee Colony (ABC) and Genetic Algorithm (GA) for optimization to solve the existing problems. The proposed algorithm has four phases such as, employee bee, onlooker bee, mutation and scout bee. In employee bee phase, neighbour solution is generated based on ABC algorithm. Then, in onlooker bee, the probability is used to select a solution and new solution is generated based on fractional calculus-dependent neighbor solution. The mutation operation of genetic algorithm is used in the mutation module and then the scout bee phase is carried out. The proposed algorithm is implemented in MATLAB. For experimentation, the unimodal benchmark functions such as: De jong’s, axis parallel hyper-ellipsoid, rotated hyper-ellipsoid and multi-modal functions such as: Griewank and rastrigin are utilzed to anlayse the performance of the algorithm. Then, the comparison of the algorithm is also, carried out with the existing ABC, GA and hybrid algorithm. From the results, we can see that the proposed technique has obtained better results by acquiring better minimization and convergence rate.