Adaptive Accelerating Differential Evolution
Qian De-ling · Fuza xitong yu fuzaxing kexue · 2008
Differential evolution(DE) is a new evolutionary computation technology and exhibits good performance on optimization.However the algorithm,to the high dimension and high multi-modal function,will fall into premature convergence.And its performance is strongly influenced by the differential strategy and the value of each strategy parameter including scale factor.Therefore,Adaptive Accelerating Differential Evolution(AADE) which is proposed to solve the optimization problems.The basic principle of AADE is that chaos initialization is adopted to improve individual quality,the scale factor and differential strategy adjusted randomly generation by generation to avoid the artificial factors,to enhance the searching capacity.Experiments on benchmark functions show that AADE outperform standard DE in function optimization.