变电设备状态监测大数据的查询优化方法

发布时间：2019-05-29 17:06

【摘要】：变电设备状态监测数据体积大、价值密度低,传统数据处理方法不能很好地满足状态监视、评估与诊断等应用快速查询的需要。文中通过对状态监测数据特点和分布式列数据存储方法的分析,给出了变电设备状态监测的大数据处理框架。通过对监测时间、监测设备编号和设备编号等数据属性的组合,设计了3种状态监测数据复合行键结构,以提高状态监测数据行键查询的灵活性。为了解决在行键未知情况下全表扫描效率低下的问题,提出基于协处理器的二级索引构建方法,实现在非行键约束条件下的快速查询。实验结果表明,基于协处理器的二级索引方法在查询效率上比无索引和IHBase二级索引方式有了明显提高,对状态监测数据写入速度影响较小,能够较好地满足大数据环境下变电设备状态监测大数据快速、灵活查询的需要。
[Abstract]:The data volume of condition monitoring of substation equipment is large and the value density is low. The traditional data processing method can not meet the needs of fast query such as condition monitoring, evaluation and diagnosis. Based on the analysis of the characteristics of condition monitoring data and the distributed column data storage method, the big data processing framework of substation equipment condition monitoring is given in this paper. Through the combination of monitoring time, monitoring equipment number and equipment number, three kinds of composite row key structures of condition monitoring data are designed to improve the flexibility of row key query of condition monitoring data. In order to solve the problem of low efficiency of full table scanning when the row key is unknown, a two-level index construction method based on coprocessor is proposed to realize the fast query under the condition of non-row key constraint. The experimental results show that the query efficiency of the two-level index method based on coprocessor is obviously higher than that of non-index and IHBase secondary index, and has little effect on the writing speed of state monitoring data. It can meet the needs of big data fast and flexible query of substation equipment condition monitoring in big data environment.
【作者单位】：华北电力大学控制与计算机工程学院;
【基金】：国家自然科学基金资助项目(61074078)~~
【分类号】：TM76

【相似文献】