关键词: 组织P系统, 蜣螂算法, 自适应t分布, 动态惯性权重

Abstract: In response to the problem that the original dung beetle optimizer algorithm (DBO) is susceptible to its own influence, resulting in an imbalance between local and global search, and easily falling into the local optima. This paper proposes an adaptive t-distribution DBO with tissue-like membrane (MC-TDBO). Design adaptive inertia factors to change the step sizes of breeding dung beetles and stealing dung beetles, dynamically adjust the exploration range of individual dung beetles, and coordinate and optimize the global search and local development capabilities of the algorithm. Introduce whale optimization algorithm to improve the foraging behavior, promote the population to move closer to the optimal position, and enhance the computational accuracy of the algorithm. Combine success rate with adaptive t-distribution to enhance the ability to escape local optima. Combine tissue-like P system in membrane computing with improved DBO algorithm to enhance algorithm convergence efficiency. Simulated test using 14 benchmark functions shows that compared to the original DBO algorithm, MC-TDBO algorithm and other four algorithms have significantly improved optimization speed, solution accuracy, and stability. Finally, MC-TDBO is used in threshold segmentation for the further validation of its effectiveness.

Key words: tissue-like P system, dung beetle optimizer algorithm, adaptive t-distribution, dynamic inertia weight