论文部分内容阅读
为了解决粒子群算法收敛速度慢和早熟收敛等问题,根据生物免疫系统理论中的克隆选择学说,提出一种量化正交免疫克隆粒子群算法.给出正交子空间分割算法,并采用正交交叉策略来增强子代个体解分布的均匀性.为避免个体邻域内最优解的丢失,提出一种自学习算子.并证明该算法的全局收敛性.实验中对标准测试函数进行20～1000维的测试,分别与5种算法进行比较,并给出算法参数对计算复杂度的影响.结果表明,本文方法有效克服早熟收敛,并且在保持种群多样性的同时提高收敛速度.
In order to solve the problems of slow convergence and premature convergence of particle swarm optimization, a novel orthogonal immune clonal particle swarm optimization algorithm is proposed based on the theory of clonal selection in biological immune system theory. The orthogonal subspace segmentation algorithm is given and orthogonal Cross strategy to enhance the homogeneity of the solution distribution of the offspring individuals.In order to avoid the loss of the optimal solution in the individual neighborhood, a self-learning operator is proposed and the global convergence of the algorithm is proved.In the experiment, 1000-dimensional test, and compared with the five algorithms respectively, and gives the influence of the algorithm parameters on the computational complexity.The results show that the proposed method can effectively overcome the premature convergence and improve the convergence rate while keeping the population diversity.