[1]熊 伟,李瑞清,陈 荦,等.一种基于空间划分树裁剪外包框的空间索引方法[J].郑州大学学报(工学版),2022,43(03):1-7.
点击复制

一种基于空间划分树裁剪外包框的空间索引方法()
分享到:

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
43
期数:
2022年03期
页码:
1-7
栏目:
出版日期:
2022-04-10

文章信息/Info

Title:
Overview of New Swarm Intelligent Optimization Algorithms
作者:
熊 伟李瑞清陈 荦曹竞之资文杰
文献标志码:
A
摘要:
空间数据库中,基于 R 树的时空索引使用最小外包框对时空数据进行近似以提高查询效率,通过裁剪外包框的冗余空间可以进一步提高索引的效率。 针对这一问题,提出了一种基于 CBB 的改进的时空索引方法。 首先,将优化方法从平面二维拓展到了时空维度中,计算可能的裁剪点,在空间索引中记录外包框中的冗余空间范围,对索引节点外包框的裁剪空间进行优化,减少查询过程中不必要的子节点的计算然后,分析时空维度中查询框与索引节点外包框的相交情况,对查询中后续判断的算法进行研究,避免裁剪过程中冗余的裁剪点比较,优化了基于时空索引进行范围查询的计算过程。 实验结果表明:所提空间索引方法裁剪索引节点外包框大小是 CBB 方法的 3 倍,且减少了 40%的节点计算量,查询耗时降低 了 20%,进 一步提升了 基于空间 划分树 的时空索引的查询性能。
Abstract:
Intelligent o ptimization algorit hms co uld be divi ded into fou r ca tegories : nature-like optimization algorithm, evo lutionar y algorithm , plant grow th simula tion algo rithm , and sw arm i nte lligence op ti mization alg orithm. T he sw ar m intell igence op timization al gorithm wa s the most im porta nt typ e of algorith m. It played an i mport ant role in solving complex e ngineer ing proble ms, and tog ether with im age proc essing, fau lt detec-tion, path planni ng, part icle filter ing, fe ature sel ection, prod uctio n sch eduling, intrusion de tec tion, support ve ctor ma chines, wireless se nsors, neural network m odels , an d g ot more exte nsive app lications in o ther fields. In re cent years, intelligent optim ization algorithms s uch as bat algorith m, fruit fly optimizati on alg orithm, whale optimi zat ion al gorit hm, salp s warm algori thm, and h arris hawks op timizatio n algorithm we re widely used . Based o n the se five new swarm i ntell igence op timiza tion algorit h m, the mo del , characterist ics, improve-men t strate gies an d applic ation fields of the algorithm were reviewed. It analyzed the development opportun ities and futu re trends it fac ed from t heoretical inv estigation s, impr ovem ent str ategy and a pplication stu dies, and provi ded a gu idanc e o n algo rithm ap plication. Find ings showed tha t swarm in tell igen ce opt imi zation algorithm could perform well on many classic proble ms, but still should be exp anded in the f ields of multi -objective optimizati on, mul ti- constraint optim izat ion, dyna mic optimizat ion, and mix ed v ariab le optimiza tion. Ef fective para mete r co ntrol of di ffe rent gr oups of i ntelligent opt imiza tion algorithm in the fa ce of various specific prob-lem s wa s still t he focus o f futur e st udies. C o-evolution from populations, exploring more efficient hybrid meth-ods and sea rch stra tegies could be feasible solutions.
更新日期/Last Update: 2022-05-02