An improved Harris Hawks optimizer combined with extremal optimization

Harris Hawks optimizer (HHO) is a new swarm intelligence optimization algorithm proposed in recent years. It seeks the optimal solution by simulating the predation strategy of Harris hawks and many previous experiments show that HHO has a good effect on solving optimization problems. However, HHO also has the shortcomings of low convergence accuracy and easy to fall into local optimum. In order to improve the performance of HHO, an improved HHO hybridized with extremal optimization (IHHO-EO) is proposed. Aiming at the defect of insufficient information utilization and excessive randomization in the exploration phase of the algorithm, the own historical optimal position of Harris hawks is introduced to better guide the individuals to search for better positions and improve the global search ability. Secondly, a nonlinear prey energy escaping factor is proposed to better balance the exploration and exploitation phases. Thirdly, refracted opposition-based learning (ROBL) with a dynamic parameter is proposed and combined with HHO, which can improve the quality of solutions and convergence speed. Finally, the exploitation ability is improved by performing EO operation which has strong local search ability. The proposed algorithm is applied to 23 classical benchmark test functions and 29 CEC2017 test functions. IHHO-EO is compared with HHO, other newly proposed optimization algorithms and some improved variants of HHO. The experimental results verify the effectiveness of the added strategies. In addition, the proposed approach is applied to solving the pressure vessel design problem. The results show that IHHO-EO has an excellent performance in terms of accuracy, reliability and statistical tests.