Hierarchical guided manta ray foraging optimization for global continuous optimization problems and parameter estimation of solar photovoltaic models

The manta ray foraging optimization (MRFO) algorithm has gained widespread application in engineering problems due to its notable performance and low computational cost. However, the algorithm is prone to getting trapped in local optima and struggles to effectively balance exploration and exploitation, primarily due to the use of fixed parameters and its somersault foraging mechanism, which focuses solely on the current best solution. To address these shortcomings, we propose an improved version called hierarchical guided manta ray foraging optimization (HGMRFO). In this version, the fixed parameters are replaced with an adaptive somersault factor, which helps balance exploration and exploitation during the evolutionary process. A novel hierarchical guidance mechanism is introduced in the somersault foraging strategy. Starting from the population structure, the mechanism guides the search direction of the population individuals through hierarchical interactions. To validate the effectiveness of HGMRFO, we compared it with seven state-of-the-art algorithms on 29 IEEE CEC2017 benchmark functions, achieving an average win rate of 73.15%. On 22 IEEE CEC2011 real-world optimization problems, HGMRFO obtained the most optimal solutions. Additionally, we analyzed the optimal parameter configurations, the impact of hierarchical interactions, and the computational complexity of the proposed method. Finally, when solving the parameter estimation problems for six multimodal photovoltaic models, HGMRFO outperformed other competing methods with a success rate of 97.62%, highlighting its superior performance in the photovoltaic field.