Development of new hybrid algorithms based on Bonobo optimization for global optimization and engineering design problems

With the advancement of technology, new algorithms are constantly being introduced for optimization problems that are becoming more difficult day by day. In problems where the solution is very difficult or takes a long time, new methods are developed according to the acceptability of the optimal solution. There are many algorithms in the literature in this field. Since the existence of a single algorithm for each problem is not sufficient, algorithms that can be adjusted to be used to solve more than one problem make problem solving much easier. In this study, the Bonobo Optimization Algorithm, Grey Wolf Optimization Algorithm, Particle Swarm Optimization Algorithm, and Artificial Bee Colony Optimization Algorithm are examined in detail. Additionally, four different hybrid algorithms based on the Bonobo Optimization Algorithm are proposed. BOGKO, obtained by hybridizing the Bonobo Optimization Algorithm and the Grey Wolf Optimization Algorithm, BOPSO, obtained by hybridizing the Bonobo Optimization Algorithm and the Particle Swarm Optimization Algorithm, BOYAK, obtained by hybridizing the Bonobo Optimization Algorithm and the Artificial Bee Colony Optimization Algorithm, and the performance of the HybridBO algorithm obtained by hybridizing these four algorithms are compared by testing them on standard quality test functions, Cec'2019 functions, and engineering design problems. In this study, two separate parameter values used in the Bonobo Optimization Algorithm, the sharing coefficient (scab) for alpha bonobo and the sharing coefficient (scsb) parameters for the selected bonobo, are reproduced with chaotic maps and it is shown that the obtained results give better results than the Classic BO algorithm. These algorithms are also used to solve engineering design problems. The results obtained are explained in tables. According to the results, it is shown that the results obtained from hybrid algorithms often give better results than classical algorithms.