基于距离度量的差分进化算法

Distance-base Differential Evolution Algorithm

针对差分进化算法求解函数优化问题存在过早收敛和不稳定等缺陷,提出一种基于距离度量的差分进化算法。该算法考虑各粒子的差异,利用欧式距离计算粒子与已知最优粒子的距离,然后根据差异自适应调整自身的交叉概率因子,同时增加柯西变异算子对部分个体进行变异操作,以提高种群多样性,增强算法跳出局部最优解的能力。用三种经典函数检验说明,新算法在收敛精度、速度上优于基本差分进化算法。

For premature convergence and instability of differential evolution in solving function optimiza-tion problem, a distance-base differential evolution algorithm（DDE） is proposed. In order to improve the popu-lation＇ s diversity and the ability of breaking away from the local optimum, the difference between particles was considered, and the Euclidean distance was used to calculate the difference between a particle and the known best global particle, then the particle tuned adaptively the value of the crossover probability factor ac-cording to the difference, at the same time the Cauehy mutation operator is adapted to mutate partial individu- als. The experimental results show that the new algorithm is better than the original differential evolution algo-rithm in convergence rate and accuracy.