CAJ | 학술논문

差分进化算法是一种新的进化计算技术,为解决其早熟问题,提出了一种基于耗散结构理论的改进差分进化算法.在变异成功的个体数和交叉算子之间建立联系,使变异成功的个体影响交叉算子,提高全局收敛能力.仿真实验表明,通过对三个标准测试函数的测试,并与标准遗传算法和差分进化算法相比,所提出的改进差分进化算法是一种收敛速度快、求解精度高、鲁棒性较强的全局优化算法.
차분진화산법시일충신적진화계산기술,위해결기조숙문제,제출료일충기우모산결구이론적개진차분진화산법.재변이성공적개체수화교차산자지간건립련계,사변이성공적개체영향교차산자,제고전국수렴능력.방진실험표명,통과대삼개표준측시함수적측시,병여표준유전산법화차분진화산법상비,소제출적개진차분진화산법시일충수렴속도쾌、구해정도고、로봉성교강적전국우화산법.
Differential evolution algorithm is a new evolutionary computation technology.In order to avoid the premature convergence problem,this paper proposed a modified differential evolution algorithm with the dissipative structure theory.Established the relationship between the number of the successfully mutated individuals and the cross operator.So the number of the successfully mutated individuals could affect the cross operator,improved the global search.In test,used three benchmark functions,and compared the performance of the proposed modified differential evolution algorithm with GA and DE.The result demonstrates that it's a powerful global optimization algorithm with rapid convergence rate,high solution quality and robustness.