Abstract
This paper aims to address the problems of low initial population diversity and being prone to falling into a local optimum in the basic Black-winged kite optimization algorithm (BKA) and proposes an improved multi-strategy hybrid black-winged kite optimization algorithm (IMBKA) and its application. Firstly, in the process of generating the initial group, the optimal point set model is adopted for optimization; Secondly, an adaptive weighting method has been added to the attack behavior; Then, alert behaviors that can significantly enhance the robustness and optimize the performance of the algorithm were introduced; Finally, the Levy flight strategy was combined with the migration behavior to prevent the algorithm from becoming trapped in a local optimum. In this study, the Markov chain was constructed to prove the convergence of the improved algorithm, and a test function was used to conduct a comparative test of IMBKA with five other algorithms. The results demonstrate that the performance of the improved algorithm surpasses that of the other algorithms. In practical applications, a model for predicting pantograph-catenary contact resistance is constructed by optimizing the parameters of the Support Vector Machine (SVM) through IMBKA. The model’s prediction results further demonstrate that the improved algorithm is practical. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1038/s41598-026-36871-x.