摘要
结合多连接查询的特点,提出将侧重于全局搜索的遗传算法与侧重于局部搜索的模拟退火算法相结合的数据库多连接查询优化算法:先从一组随机产生的初始种群开始全局最优解的搜索,通过选择、交叉、变异等遗传操作产生新一代种群,然后对新个体进行模拟退火操作,将结果作为下一代种群中的个体.如此反复迭代进行,到满足最终条件为止.仿真实验验证了该算法的有效性.
The algorithm of multi-joint query optimization based on genetic algorithm with overall search ability and simulated annealing with local search ability are proposed by combining with characteristic of multi-join optimization：Starting an optimal-solution-search to the overall situation in a group of initial population, which is random selected. A new generation of population will be produced after the selection strategy, crossover and mutation. And then the simulated annealing is applied to those new populations, and the result is used as the unit of the next generation population. The above process is operated repeatedly and iterativey, until the result meets the the final qualification. The simulation experiment has proved its efficiency.
出处
《郑州轻工业学院学报（自然科学版）》
CAS
2007年第6期44-47,50,共5页
Journal of Zhengzhou University of Light Industry:Natural Science
基金
黑龙江省教育厅科学技术研究项目(11521006)
黑龙江省自然科学基金项目(2005G3674-00)
关键词
查询优化
遗传算法
模拟退火
多连接查询
query optimization
genetic algorithm
simulated annealing
multi-joint query
