论文部分内容阅读
并行遗传算法将并行计算机的高速并行性和遗传算法固有的并行性相结合 ,极大地提升了遗传算法的求解速度和质量 .在主从式、细粒度和粗粒度这三类遗传算法并行化模型中 ,粗粒度模型以其较小的通讯开销和对种群多样化 ,获得了最广泛的应用 .本文概括了基于模式定理和有限状态马尔可夫链的遗传算法理论 ,总结了前人在粗粒度模型下开展的理论分析和实践应用 ,并指出并行遗传算法的研究将向异步化 ,理论化和模型化的方向发展 ,而有限状态马尔可夫链是构建并行遗传算法可执行模型的有力工具
The parallel genetic algorithm combines the parallelism of parallel computers with the inherent parallelism of genetic algorithms, which greatly improves the speed and quality of solving genetic algorithms.In the parallelization model of three kinds of genetic algorithms, that is, master-slave, fine-grained and coarse-grained, , The coarse-grained model obtained the most extensive application due to its small communication overhead and diversification of population.This paper summarized the theory of genetic algorithm based on the mode theorem and finite state Markov chain, Model under the theoretical analysis and practical application, and pointed out that the parallel genetic algorithm research will be to the direction of induction, theorization and modeling direction, and the Finite State Markov chain is to build parallel genetic algorithm executable model of a powerful tool