Improving Wild Horse Optimizer: Integrating Multistrategy for Robust Performance across Multiple Engineering Problems and Evaluation Benchmarks

Abstract

In recent years, optimization problems have received extensive attention from researchers, and metaheuristic algorithms have been proposed and applied to solve complex optimization problems. The wild horse optimizer (WHO) is a new metaheuristic algorithm based on the social behavior of wild horses. Compared with the popular metaheuristic algorithms, it has excellent performance in solving engineering problems. However, it still suffers from the problem of insufficient convergence accuracy and low exploration ability. This article presents an improved wild horse optimizer (I-WHO) with early warning and competition mechanisms to enhance the performance of the algorithm, which incorporates three strategies. First, the random operator is introduced to improve the adaptive parameters and the search accuracy of the algorithm. Second, an early warning strategy is proposed to improve the position update formula and increase the population diversity during grazing. Third, a competition selection mechanism is added, and the search agent position formula is updated to enhance the search accuracy of the multimodal search at the exploitation stage of the algorithm. In this article, 25 benchmark functions (Dim = 30, 60, 90, and 500) are tested, and the complexity of the I-WHO algorithm is analyzed. Meanwhile, it is compared with six popular metaheuristic algorithms, and it is verified by the Wilcoxon signed-rank test and four real-world engineering problems. The experimental results show that I-WHO has significantly improved search accuracy, showing preferable superiority and stability.

Research and Design of Improved Wild Horse Optimizer-Optimized Fuzzy Neural Network PID Control Strategy for EC Regulation of Cotton Field Water and Fertilizer Systems

“…The single-horse vector contains a total of 175 elements. The vector is initialized using Xavier [23]. Set the population's size, the stallion percentage, and the subgroup number.…”

Optimizing Multi-Layer Perovskite Solar Cell Dynamic Models with Hysteresis Consideration Using Artificial Rabbits Optimization

“…Broadly, these algorithms can be classified into evolutionary algorithms (EA) [20,21] and swarm-based techniques [22]. These algorithms have proven to be more versatile and effective in addressing engineering challenges than conventional optimization methods [23], as subsequently discussed.…”