论文部分内容阅读
提出一种求解动态优化问题的改进差分进化算法.新算法将种群分为跟踪和搜索两个种群.通过监测跟踪种群的当前最优解和次优解来判断环境是否发生变化.发现环境变化时,重新计算种群适应值,分别找出变化后两个种群新的最优解.最优解好的种群,变为跟踪种群,保持不变,采用DE/best/1变异策略,在其最优解附近进行局部搜索;最优解差的种群,变为搜索种群,重新初始化,采用DE/rand/1变异策略全局搜索,扩大搜索范围,寻找新的最优解.搜索过程中,跟踪种群和搜索种群各负其责,相互配合提高了算法的搜索效率.比较跟踪和搜索种群的最优解,好的最优解作为动态优化问题的解.最后,用Dynamic Function1(DF1)函数对算法进行了验证,实验结果表明该算法可行有效.
This paper proposes an improved differential evolution algorithm to solve the dynamic optimization problem.The new algorithm divides the population into two groups: tracking and searching.Through monitoring and tracking the current optimal solution and the sub-optimal solution of the population to determine whether the environment changes, , Recalculate the fitness of the population and find the new optimal solution of the two populations after the change respectively.The optimal solution of the population becomes the tracking population and remains the same.Using DE / best / 1 mutation strategy, The population with the optimal solution becomes the search population and reinitialize, and the global search with DE / rand / 1 mutation strategy is used to expand the search range and search for the new optimal solution.In the search process, The searching population is responsible for each other, which improves the searching efficiency of the algorithm.Compared with the optimal solution of tracking and searching population, good optimal solution is used as the solution to the dynamic optimization problem.Finally, we use Dynamic Function1 (DF1) function to carry out the algorithm Validated, the experimental results show that the algorithm is feasible and effective.