基于模糊群决策算法的高性能数据库优选  

Optimization of High-performance Database Based on Fuzzy Group Decision-making Algorithm

作　　者：胡致杰 黄小虎[2] HU Zhi-jie;HUANG Xiao-hu(Department of Information Technology,Guangdong Technology College,Zhaoqing 526100,China;School of Mathematics and Information,South China Agricultural University,Guangzhou 510000,China)

机构地区：[1]广东理工学院信息技术学院,广东肇庆526100 [2]华南农业大学数学与信息学院,广东广州510000

出　　处：《控制工程》2022年第7期1302-1309,共8页Control Engineering of China

基　　金：广东省教育厅教学改革项目(2018-661)。

摘　　要：模糊群决策的两个主要问题包括完全一致模糊偏好关系(FPR)的构造和专家共识水平的提升,因此,构建了基于乘性一致性和共识调整模型的群决策算法,并将其运用于高性能数据库优选问题。首先,运用专家提供的FPR,设计一种完全乘性一致FPR的构造方法,进而引入同时衡量一致性和共识性的一致性-共识性指数;其次,提出基于最小调整成本策略的群共识调整模型,以共识调整总成本最小化为目标函数建立最优化模型,进而得到每个专家的最优调整参数和共识阈值下界;最后,构建一种具有收敛性的模糊群决策算法。通过高性能数据库产品的优选实验表明,所构建的群决策算法在成本和效率方面更为有效。The two main problems of fuzzy group decision-making include the construction of completely consistent fuzzy preference relation(FPR) and the improvement of expert consensus level.Therefore,a group decision-making algorithm based on multiplicative consistency and consensus adjustment model is constructed and applied to the optimization problem of high-performance database.Firstly,by using the FPR provided by experts,a construction method of completely multiplicative consistent FPR is designed,and then a consistency-consensus index is introduced to measure the consistency and consensus at the same time.Secondly,a group consensus adjustment model based on the minimum adjustment cost strategy is proposed.Taking the minimization of the total cost of consensus adjustment as the objective function,an optimization model is established,and then the optimal adjustment parameters and the lower bound of consensus threshold of each expert are obtained.Finally,a convergent fuzzy group decision-making algorithm is constructed.The optimization experiment of high-performance database products shows that the constructed group decision-making algorithm is more effective in terms of cost and efficiency.

关 键 词：群决策 模糊偏好关系 乘性一致性 群体共识调整算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]