Bi-population Cooperative Moth-flame Optimization Algorithm for the Networking Mode Optimization
Shaorong Cao, Yuanyuan Mao, Mingyang Li, Weidong Xie, Chaochao Gao · 2023
The moth-flame optimization (MFO) algorithm is extensively employed to attain the global optimization of the problem. The disadvantages of the original MFO algorithm include poor population variety, a sluggish rate of convergence, and an easy propensity to be drawn in by local optimum. This paper presents a bi-population cooperative moth-flame optimization algorithm (BCMFO) to address the issues. Utilizing the low discrepancy sequence (LDS), a random population with a uniform distribution is produced in the search space. Two subpopulations with similar sizes are updated using Gauss mutation and opposition learning. To improve the algorithm's ability to search globally, the elite method is used to eliminate the subpar solutions from the population. BCMFO is used to optimize the networking mode and is confirmed using the benchmark test suite in CEC 2017. Experimental results show that BCMFO outperforms state-of-the-art algorithms.