一种并行混沌差异演化算法

在提出的并行混沌差异演化算法中，差异演化算法与混沌搜索并行进行，采用基于最优个体的差异演化算法，差异演化的权重因子随种群的聚集度和算法停滞代数作动态调整，而交叉率随种群的进化速度动态调整．利用混沌的随机性和遍历性来扩大算法的搜索范围，与差异演化算法实现功能互补，在每一代并行计算后，引入灾变算子，在达到灾变条件时，对种群实行灾变操作，从而避免了算法陷入局部的最优值．仿真结果表明，与基本差异演化算法相比，并行混沌差异演化算法的收敛速度快，搜索效率高，可用于解决复杂的机器学习问题．

Parallel Chaos Differential Evolution Algorithm

A novel parallel chaos differential evolution algorithm(PCDE) is presented,where differential evolution(DE) and chaos searching run concurrently,and in each iteration process the weighting factor is changed dynamically according to the current aggregation degree and number of stopping generations,the crossover factor is changed dynamically according to the current evolution rate.The randomness and space ergodicity of chaos mapping are considered to enlarge the searching range.The catastrophic factor is introduced to avoid premature convergence.DE and PCDE are tested with three well-known benchmark functions.The experiments show that PCDE is significantly superior to DE with higher efficiency to solve complicated optimization problems.