ABSTRACT

The Satin Bowerbird Optimisation (SBO) was inspired by the Satin Bowerbirds living in Australia’s rainforests and other mesic habitats. Like other meta-heuristic algorithms, the main problem faced by the SBO is that it has been empirically demonstrated to become easily trapped into local optimal solutions, creating low precision and slow convergence speeds. To overcome these deficiencies, we propose herein the Adaptive Chaotic Satin Bowerbird Optimisation algorithm (AC-SBO). Within the AC-SBO algorithm, a chaotic map is introduced to modify the search process, with which to enhance global convergence speeds, and to obtain better performance. We introduced the chaos theory into the SBO optimisation process, in order to replace the main parameter’s greatest step size ($\alpha$), which assists in controlling the balance between both exploration and exploitation. The search accuracy and performance of the AC-SBO algorithm were verified on ten classical benchmark functions. In addition, in the experimental CEC2014 results showed that for almost all functions, the AC-SBO technique proved superior to the other comparative algorithms optimisations. The Wilcoxon rank-sum statistical test was performed in order to judge the significance of the results, and further demonstrated the improved performance of the proposed AC-SBO algorithm.

摘要

Satin Bowerbird Optimization (SBO) 的灵感来自生活在澳大利亚热带雨林和其他热带栖息地的 Satin Bowerbird。与其他元启发式算法一样，SBO 面临的主要问题是经验证明它很容易陷入局部最优解，导致精度低和收敛速度慢。为了克服这些缺陷，我们在此提出了自适应混沌缎面园丁优化算法（AC-SBO）。在 AC-SBO 算法中，引入了混沌映射来修改搜索过程，从而提高全局收敛速度并获得更好的性能。我们将混沌理论引入到 SBO 优化过程中，以替换主要参数的最大步长（$\alpha$)，这有助于控制探索和开发之间的平衡。AC-SBO 算法的搜索精度和性能在十个经典基准函数上得到验证。此外，CEC2014 的实验结果表明，对于几乎所有功能，AC-SBO 技术都证明优于其他比较算法优化。为了判断结果的显着性，进行了 Wilcoxon 秩和统计检验，并进一步证明了所提出的 AC-SBO 算法的改进性能。