An Improved Dung Beetle Optimization Algorithm Based on Multi-Strategy

Aiming at the shortcomings of slow convergence speed and easy to fall into local optimum of dung beetle optimization algorithm (DBO) in high-dimensional multimodal optimization problems, this paper proposes a multi-strategy improved DBO algorithm (MIDBO). Firstly, the Piecewise mapping and quasi-opposition learning method are introduced to initialize the population, which enriches the diversity and uniform distribution of the initial population. Secondly, the mutation, crossover and selection mechanisms of the differential evolution (DE) algorithm are integrated to effectively enhance the diversity of the population and accelerate the convergence speed of the algorithm. Finally, MIDBO uses adaptive golden sine strategy and Cauchy inverse cumulative distribution function to improve the rolling ball behavior, further balance global exploration and local development, and enhance the ability to jump out of local optimum. The experimental results show that MIDBO is superior to traditional DBO and other classical optimization algorithms in terms of convergence speed, optimization accuracy and result stability on 23 benchmark functions, which verifies the effectiveness and advantages of the improved algorithm.