| 摘 要: 针对原始蜣螂优化算法(DBO)存在的收敛精度低、易陷入局部最优等问题,提出一种改进的蜣螂优化算法。该算法采用混沌映射初始化蜣螂种群以提高种群的多样性,引入北方苍鹰优化算法的勘探策略以增强算法的全局勘探能力,并改进一种非线性边界收敛因子以平衡其收敛速度和收敛精度。同时,采用自适应T分布扰动策略以增强算法跳出局部最优的能力。实验结果表明,改进后的DBO算法在15个基准测试函数的求解寻优中,有13个测试函数的求解结果优于原始蜣螂优化算法、麻雀搜索算法、灰狼优化算法、鲸鱼优化算法和哈里斯鹰优化算法的求解结果,表现出更高的收敛精度、更快的收敛速度及更高的稳定性。 |
| 关键词: 蜣螂优化算法;混沌映射;T分布扰动;基准测试函数 |
|
中图分类号: TP181
文献标识码: A
|
| 基金项目: 国家自然科学基金项目(61273326) |
|
| Dung Beetle Optimization Algorithm Integrating Chaotic Mapping and Adaptive T-Distribution |
|
LI Hongmin, MA Yawei, LIU Ruiyu, WANG Ming
|
(School o f In f ormation and Electrical Engineering, Shandong Jianzhu University, Jinan 250101, China)
lhm640723@126.com; 3455482707@qq.com; 1261129450@qq.com; 11201@sdjzu.edu.cn
|
| Abstract: Aiming at the poor convergence accuracy and the tendency to fall into local optima in the original Dung Beetle Optimization(DBO) algorithm, this paper proposes an improved DBO algorithm. This algorithm utilizes chaotic mapping to initialize the dung beetle population, thereby enhancing population diversity. It incorporates an exploration strategy from the Northern Goshawk Optimization(NGO) algorithm to strengthen the global exploration capability of the algorithm and improves a nonlinear boundary convergence factor to balance its convergence speed and accuracy. Additionally, an adaptive T-distribution disturbance strategy is employed to enhance the algorithm's ability to escape local optima. The experimental results show that the improved DBO algorithm outperforms the original DBO, Sparrow search algorithm, Grey Wolf Optimization algorithm, Whale Optimization algorithm, and Harris Hawk Optimization algorithm in solving and optimizing 13 out of 15 benchmark test functions, exhibiting higher convergence accuracy, faster convergence speed, and higher stability. |
| Keywords: Dung Beetle Optimization; chaotic mapping; T-distribution disturbance; benchmark test function |
