Diversity-Based Evolutionary Population Dynamics: A New Operator for Grey Wolf Optimizer

Evolutionary Population Dynamics (EPD) refers to eliminating poor individuals in nature, which is the opposite of survival fittest. Although this method can improve median whole population meta-heuristic algorithms, it suffers from exploration capability handle high-dimensional problems. This paper proposes a novel EPD operator search process. In other words, as primary mainly improves fitness worst population, and hence we name Fitness-Based (FB-EPD), our proposed diversity best individuals, Diversity-Based (DB-EPD). The applied Grey Wolf Optimizer (GWO) named DB-GWO-EPD. algorithm, three most diversified are first identified at each iteration, then half best-fitted forced be eliminated repositioned around these agents with equal probability. process free merged located closed populated region transfer them and, thus, less-densely regions space. approach frequently employed make explore DB-GWO-EPD tested on 13 shifted classical benchmark functions well 29 test problems included CEC2017 suite, four constrained engineering results obtained by proposal upon implemented compared GWO, FB-GWO-EPD, popular newly optimization including Aquila (AO), Flow Direction Algorithm (FDA), Arithmetic Optimization (AOA), Gradient-based (GBO). experiments demonstrate significant superiority algorithm when majority functions, recommending application any whenever decided ameliorate their performance.