A Chaotic-integrated Harris Hawks Optimization Algorithm for Solving Numerical Optimization Problems

The complexity of the real-world problems is increasing exponentially and prior approaches are losing their efficacy. Therefore, researchers are developing new methods for solving problems. One of these state-of-the-art approaches is the use of meta-heuristic optimization algorithms. Optimization algorithms are the most recent approaches to deal with the complexity of real-world problems, which have shown an abundant performance in solving many issues. Hence, much attention has been paid to these algorithms recently, as numerous optimization algorithms have been developed. The Harris Hawks Optimization (HHO) algorithm is a cutting-edge optimization algorithm that has been inspired by the intelligent behavior and chasing strategies of Harris's hawks in nature and has an extraordinary performance in solving various problems. In this paper, the effect of eleven eminent chaotic maps has been investigated on the proficiency of the HHO. The experiments were performed on fifty-three benchmark functions and the obtained results have been compared with the original HHO in terms of mean and standard deviation. The simulation results show that in the unimodal benchmark functions, Sinusoidal and Singer maps, in the multimodal and the fix-dimension benchmarks, the Leibovitch map, in the shifted and rotated benchmarks, the Intermittency map, and in the hybrid and composite benchmarks Tent map outperformed. In addition, the convergence speed of the chaotic HHO algorithm is discussed in all chaotic maps over each benchmark function. The results represent that chaotic maps can enhance the exploration and exploitation capabilities of the HHO in almost all cases.

Keywords:

Optimization , Chaos theory , Chaotic Map , Global Optimization , Harris Hawks Optimization