Parameters Optimization for Improved Grey Wolf Optimizer by Using Uniform Experimental Design
Ming-Zhen Tsai, Po-Yuan Yang, Fu-I Chou, Jyh‐Horng Chou · 2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) · 2021
This paper utilizes an improved strategy and a uniform experimental design (UED) to enhance the algorithmic efficiency of a grey wolf optimizer (GWO) algorithm. The original GWO algorithm is explored and developed by a linear descent in the convergence factor. However, the linear descent may not be efficient in the solution search, and it is easy to fall into the local optimum. For improving the weakness, this paper refers to an improved GWO. The IGWO adopts a convergence factor with a nonlinear strategy, which is based on a sine function, to balance the abilities of exploration and development. In addition, three parameters of ainitial, μ, and C in GWO that need to be optimized are considered. Therefore, the UED method is applied to find the best combination of parameters. From the experimental results, the proposed algorithm can obtain a better performance.