In this study a novel population based meta-heuristic, called controlled showering optimization (CSO) algorithm, for global optimization of unconstrained problems is presented. Modern irrigation systems are equipped with smart tools manufactured and controlled by human intelligence. The proposed CSO algorithm is inspired from the functioning of water distribution tools to model search agents for carrying out the optimization process. CSO imitates the mechanism of projection of water units by sprinklers and the movements of their platforms to the desired locations for scheming optimum searching procedures. The proposed method has been tested using a number of diverse natured benchmark functions with low and high dimensions. Statistical analysis of the empirical data demonstrates that CSO offers solutions of better quality in comparison with several well-practiced algorithms like genetic algorithm (GA), particle swarm optimization (PSO), differential evolution (DE), artificial bee colony (ABC), covariance matrix adaptation evolution strategy (CMA-ES), teaching and learning based optimization (TLBO) and water cycle algorithm (WCA). The experiments on high-dimensional problems reveal that CSO algorithm also outperforms significantly a number of algorithms designed specifically for high dimensional global optimization problems.

中文翻译：

---

受控淋浴优化算法：用于全局优化的智能决策工具

在这项研究中，提出了一种新颖的基于种群的元启发式算法，称为可控淋浴优化（CSO）算法，用于无约束问题的全局优化。现代灌溉系统配备了由人类智能制造和控制的智能工具。所提出的CSO算法是从配水工具的功能中获得灵感的，它可以对搜索代理进行建模以执行优化过程。CSO模仿了洒水装置投射水单元的原理，以及其平台向理想位置移动以设计最佳搜索程序的机制。所提出的方法已经使用多种具有低维和高维特性的基准函数进行了测试。对经验数据的统计分析表明，与几种良好实践的算法（例如遗传算法（GA），粒子群优化（PSO），差异进化（DE），人工蜂群（ABC），协方差）相比，CSO提供了质量更高的解决方案矩阵适应进化策略（CMA-ES），基于教与学的优化（TLBO）和水循环算法（WCA）。对高维问题的实验表明，CSO算法的性能也明显优于许多专为高维全局优化问题设计的算法。基于教学的优化（TLBO）和水循环算法（WCA）。对高维问题的实验表明，CSO算法的性能也明显优于许多专为高维全局优化问题设计的算法。基于教学的优化（TLBO）和水循环算法（WCA）。对高维问题的实验表明，CSO算法的性能也明显优于许多专为高维全局优化问题设计的算法。