Enhancing dung beetle optimization algorithm with hybrid multi-strategy and its engineering applications

Enhancing dung beetle optimization algorithm with hybrid multi-strategy and its engineering applications

연구 분야: Analysis

학회: Neural Computing and Applications

An enhanced dung beetle optimization algorithm with hybrid multi-strategy (PCTDBO) is proposed, to balance the convergence speed and diversity of the algorithm. Firstly, a convex lens imaging opposition-based learning strategy is employed to reduce the probability of the algorithm falling into local optima, enhancing the algorithm's global exploration capability. Secondly, inspired by the prairie dog optimization (PDO) algorithm, the strategy of alerting similar predators is used to replace the foraging stage strategy of the dung beetle optimization algorithm, to improve the excessive randomness during the foraging stage, which may lead to instability and slow convergence or even failure to converge. Lastly, to prevent the algorithm from getting stuck in local optima during the exploitation phase, tent chaotic mapping and Cauchy mutation are introduced to enhance the algorithm's ability to escape from local optima. To validate the effectiveness of PCTDBO, benchmark and CEC2017 test functions are tested and compared with state-of-the-art algorithms, and the experimental findings demonstrate that PCTDBO brings about notable enhancements in both convergence speed and optimization accuracy, while also showcasing strong robustness. Additionally, the algorithm has been successfully applied to five practical engineering application problems, and the results indicate that PCTDBO demonstrates good applicability and effectiveness in solving project optimization problems.