Nature-inspired metaheuristic algorithms are often based on the first-order difference hypercube search style to search for optimum solutions. In contrast, the spherical evolution algorithm (SE) employs a spherical search style. SE is very effective; however, there is still room for improvement. In this study, we added a chaotic local search (CLS) to the SE to improve its performance. This CLS uses information from several chaotic maps and records each instance of success. The recorded historical success information guides the CLS to choose the chaotic map for the next iteration. In our experiment, we compare the chaotic spherical evolution algorithm (CSE) with the original SE and other metaheuristic algorithms. The test set consists of 29 benchmark functions from the CEC2017 benchmark set and 22 real-world optimization problems from the CEC2011 set. Additionally, the new parameter introduced in the CSE has also been briefly discussed. Experimental results indicate that the proposed CSE significantly performs better than its competitors.

受自然启发的元启发式算法通常基于一阶差分超立方体搜索风格来搜索最优解。相比之下，球形进化算法 (SE) 采用球形搜索风格。SE 非常有效；然而，仍有改进的余地。在本研究中，我们向 SE 添加了混沌局部搜索 (CLS) 以提高其性能。此 CLS 使用来自多个混沌地图的信息并记录每个成功实例。记录的历史成功信息指导CLS为下一次迭代选择混沌图。在我们的实验中，我们将混沌球形进化算法 (CSE) 与原始 SE 和其他元启发式算法进行了比较。测试集由来自 CEC2017 基准集的 29 个基准函数和来自 CEC2011 集的 22 个实际优化问题组成。此外，还简要讨论了 CSE 中引入的新参数。实验结果表明，所提出的 CSE 的性能明显优于其竞争对手。