论文部分内容阅读
并行计算与演化计算同为复杂问题寻优的现代技术。两者的结合不仅减少运行时间、提高解的质量,还增加计算能力,即增加应用领域。并行演化算法除了主从式、粗粒度、细粒度和分层四类并行化模型外,还有改进的并行演化算法。本文分析介绍加速比定义分类、迁移策略与并行演化算法的应用情况,总结前人对并行演化算法开展的理论分析与实践应用研究进展,并就该领域的进一步研究方向提出了一些看法,指出细胞状模型和并行遗传程序设计将成为两大研究热点。
Parallel computing and evolutionary computation are the same as the complex problems of modern technology. The combination of the two not only reduces run-time, improves the quality of solution, but also increases the computing power, ie increases the application area. In addition to the main parallel, coarse-grained, fine-grained and hierarchical four parallel models, the parallel evolutionary algorithm also has an improved parallel evolutionary algorithm. This paper introduces the application of classification, migration strategy and parallel evolution algorithm, summarizes the previous theoretical and practical applications of parallel evolution algorithm, and puts forward some suggestions for further research in this field. It is pointed out that cells Shaped model and parallel genetic programming will become two major research hot spots.