Hybrid Multi‐Strategy Improved Wild Horse Optimizer. (7th July 2022)

• Record Type: Journal Article
• Authors: Li, Yancang; Yuan, Qiuyu; Han, Muxuan; Cui, Rong
• Is Part Of: Advanced intelligent systems. Volume 4:Number 10(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-07-07
• Subjects: escaping behavior -- Halton sequence -- mechanical optimization -- nonlinear parameter -- simplex method -- Wild Horse Optimizer; Artificial intelligence -- Periodicals; Robotics -- Periodicals; Control theory -- Periodicals; 006.3
• DOI: 10.1002/aisy.202200097
• Languages: English
• ISSNs: 2640-4567

Abstract

Abstract : Wild Horse Optimizer (WHO), a new metaheuristic algorithm proposed in recent years, has some weaknesses in solving practical problems, such as low searching accuracy and slow convergence speed. Herein, a Hybrid Multi‐Strategy improved Wild Horse Optimizer (HMSWHO) is proposed, which includes four strategies to improve the optimization capability. The Halton sequence is used to initialize the foal population to make the population more diverse. The adaptive parameter TDR is improved to balance the global exploration and local exploitation. The simplex method is used to improve the worst position of the population. Wild horse escaping behavior is added to improve search efficiency and optimization accuracy. The main innovation strategies are the improvement of TDR and the addition of escaping behavior. To verify the effectiveness of the improved strategies, 12 benchmark test functions, CEC2017, and CEC2021 test functions are selected for simulation experiments. Mechanical design examples are added for optimization, and the optimization results are 16.61%, 1.65%, and 0.21% less than that of WHO. The results show that the improved algorithm has obvious advantages in convergence speed, accuracy, and stability. HMSWHO can be applied to more practical engineering optimization problems and provide new ideas for structural optimization methods. Abstract : A Hybrid Multi‐Strategy improved Wild Horse Optimizer is proposed, which includes four strategies to improve the optimization capability. Twelve benchmark test functions, CEC2017, CEC2021 test functions, and three mechanical design examples are selected to verify the effectiveness of the improved strategies. The results show that the improved algorithm has obvious advantages in convergence speed, accuracy, and stability.