An improved self-adapting differential evolution algorithm
Wenjing Jin, Yanfeng Ge, Liqun Gao, Yang Zhang · 2010
To improve the convergence speed and obtain global optimization successfully, an improved self-adapting differential evolution algorithm (IADE) with parameters self-adaptation and a new mutation strategy is given in this paper. Utilizing the diversity of population information, IADE algorithm divides the population into three sub-populations according to fitness function values and applies different differential arithmetic to different sub-populations. The parameters self-adaptation updates the control parameters automatically to appropriate value based on historical date obtained in process. It thus helps to improve the robustness of the algorithm and avoid premature convergence. And the investigation of IADE algorithm with a set of ten standard benchmark problems shows IADE algorithm outperforms, or at least comparable to the DE algorithms and some other adaptive and self-adaptive differential evolution algorithms in terms of average fitness function value, number of function evaluations and convergence time.