Grasshopper optimization algorithm with principal component analysis for global optimization

As one of the latest meta-heuristic algorithms, the grasshopper optimization algorithm (GOA) has extensive applications because of its efficiency and simplicity. However, the basic GOA still has enough room for improvement. Therefore, a new variant GOA algorithm which combines two strategies, namely PCA–GOA, is proposed. Firstly, principal component analysis strategy is employed to obtain the grasshoppers with minimally correlated variables, which can improve the exploitation capability of the GOA. Then, a novel inertia weight is proposed to balance exploration and exploitation in an intelligent way, which makes the GOA to have better search capability. Furthermore, the performance of PCA–GOA is evaluated by solving a series of benchmark functions. The experimental results manifest that the PCA–GOA provides better outcomes than the basic GOA and other state-of-the-art algorithms on the majority of functions, which demonstrates the superiority of the PCA–GOA.