计算机工程与应用 ›› 2024, Vol. 60 ›› Issue (10): 76-87.DOI: 10.3778/j.issn.1002-8331.2309-0314

• 理论与研发 • 上一篇    下一篇

求解约束优化问题的改进蛇优化算法

梁昔明,史兰艳,龙文   

  1. 1.北京建筑大学 理学院,北京 102616
    2.贵州财经大学 数学与统计学院,贵阳 550025
  • 出版日期:2024-05-15 发布日期:2024-05-15

Improved Snake Optimization Algorithm for Solving Constrained Optimization Problems

LIANG Ximing, SHI Lanyan, LONG Wen   

  1. 1.School of Science, Beijing University of Civil Engineering and Architecture, Beijing 102616, China
    2.School of Mathematics and Statistics, Guizhou University of Finance and Economics, Guiyang 550025, China
  • Online:2024-05-15 Published:2024-05-15

摘要: 结合外点罚函数法与改进蛇优化算法求解约束优化问题,得到一种新的求解约束优化问题的算法WDFSO。算法WDFSO首先通过外点罚函数法将约束优化问题转化为一系列界约束优化问题,然后运用基于变异质心的对立学习策略与种群分类策略改进的蛇优化算法对所得界约束优化问题进行求解,进而获得所求约束优化问题的解。为验证算法WDFSO的有效性,选取CEC2006中19个标准约束优化问题进行数值实验,并使用Wilcoxon秩和检验来证明算法的显著性。实验结果表明,与对比算法相比,算法WDFSO求解约束优化问题具有更高的收敛精度和更好的稳定性。最后应用算法WDFSO求解两个工程约束优化问题,结果表明算法WDFSO求解性能更好。

关键词: 约束优化问题, 外点罚函数法, 蛇优化算法, 对立学习, 种群分类策略, 数值实验

Abstract: To solve the constrained optimization problem, a new algorithm WDFSO is obtained by combining the exterior penalty function method and an improved snake optimization algorithm. Firstly, the constrained optimization problem is transformed into a series of bound-constrained optimization problems by the exterior penalty function method. Then, the improved snake optimization algorithm based on the oppositional learning of the centroid variation strategy and the population classification strategy is used to solve the bound-constrained optimization problem, and obtain the solution of the constrained optimization problem. In order to verify the effectiveness of WDFSO algorithm, 19 benchmark constrained optimization problems in CEC2006 are selected for numerical experiments, and the Wilcoxon rank sum test is used to prove the algorithm significance. The experimental results show that WDFSO algorithm has higher convergence accuracy and better stability than the comparison algorithms. Finally, WDFSO algorithm is applied to solve two engineering constraint optimization problems, and the results show that WDFSO algorithm has better performance.

Key words: constrained optimization problem, exterior penalty function method, snake optimization algorithm, opposition-based learning, population classification strategy, numerical experiment