Grey Adaptive Differential Evolution Algorithm

On the basis of our previous work, differential evolution with grey-based adaptive mutation factor, this study attempts to propose a new strategy by updating the crossover rate in a grey-based adaptive manner. Differential evolution with the previous two grey-based parameter automation strategies is termed grey adaptive differential evolution algorithm hereafter. In grey adaptive differential evolution, each individual (target vector) has its own mutation factor and crossover rate whose values are dependent upon the corresponding grey relational grade. Since the relational grade of an individual is varying over the generations, those two control parameters are time-varying. In addition, grey adaptive differential evolution is applied to solve the optimization problems of eight benchmark functions for illustration. Simulation results show that the proposed grey adaptive differential evolution could perform well search performance on most of the test functions. It could also perform better than other self-adaptive differential evolution algorithms with lesser computational time on some test functions.