融合组织P系统的自适应t分布蜣螂算法  

Fusion of Adaptive t-Distribution Dung Beetle Optimizer Algorithm with Tissue P System

作　　者：许家昌[1,2] 江琳 苏树智 XU Jiachang;JIANG Lin;SU Shuzhi(The First Affiliated Hospital of Anhui University of Science and Technology(Huainan First People’s Hospital),Huainan,Anhui 232001,China;School of Computer Science and Engineering,Anhui University of Science and Technology,Huainan,Anhui 232001,China)

机构地区：[1]安徽理工大学第一附属医院(淮南市第一人民医院),安徽淮南232001 [2]安徽理工大学计算机科学与工程学院,安徽淮南232001

出　　处：《计算机工程与应用》2025年第4期99-113,共15页Computer Engineering and Applications

基　　金：安徽理工大学医学专项培育项目(YZ2023H2B008,YZ2023H2A007);南方林业与生态应用技术国家工程实验室开放基金项目(2023NFLY08);国家自然科学基金面上项目(52374155);安徽省自然科学基金面上项目(2308085MF218);安徽省高等学校自然科学研究基金重大项目(2022AH040113)。

摘　　要：针对原始蜣螂算法(dung beetle optimizer,DBO)易受自身影响,导致局部搜索和全局搜索不平衡,容易陷入局部最优的问题,提出一种结合组织膜的自适应t分布蜣螂算法(adaptive t-distribution DBO with tissue-like membrane,MC-TDBO)。设计自适应惯性因子改变繁育蜣螂和小偷蜣螂的步长,动态调节蜣螂个体的探索幅度,协调并优化算法的全局搜索和局部开发能力;引入鲸鱼算法改进觅食行为,促使种群向最优位置靠近,提高算法的计算精度;结合成功率和自适应t分布,提升算法跳出局部最优的能力;引入组织P系统与改进后的DBO算法结合,增强算法收敛效率。采用14个基准函数进行仿真测试,实验结果表明,MC-TDBO算法和原始DBO算法等四种算法相比,寻优速度、求解精度和稳定性均得到了显著提升。将MC-TDBO算法在阈值分割中进行应用测试,进一步验证其有效性。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 indi-vidual 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 opti-mal 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 com-pared to the original DBO algorithm,MC-TDBO algorithm and other four algorithms have significantly improved optimi-zation speed,solution accuracy,and stability.Finally,MC-TDBO is used in threshold segmentation for the further valida-tion of its effectiveness.

关 键 词：组织P系统 蜣螂算法 自适应t分布 动态惯性权重 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]