Random Grouping Brain Storm Optimization Algorithm with a New Dynamically Changing Step Size

Finding the global optima of a complex real-world problem has become much more challenging task for evolutionary computation and swarm intelligence. Brain storm optimization (BSO) is a swarm intelligence algorithm inspired by human being’s behavior of brainstorming for solving global optimization problems. In this paper, we propose a Random Grouping BSO algorithm termed RGBSO by improving the creating operation of the original BSO. To reduce the load of parameter settings and balance exploration and exploitation at different searching generations, the proposed RGBSO adopts a new dynamic step-size parameter control strategy in the idea generation step. Moreover, to decrease the time complexity of the original BSO algorithm, the improved RGBSO replaces the clustering method with a random grouping strategy. To examine the effectiveness of the proposed algorithm, it is tested on 14 benchmark functions of CEC2005. Experimental results show that RGBSO is an effective method to optimize complex shifted and rotated functions, and performs significantly better than the original BSO algorithm.