摘要
随着故障录波和行波装置的广泛应用和技术进步,IEEE COMTRADE 1999格式暂态数据(简称COMTRADE数据)已经呈现出海量化发展的趋势。如何对海量COMTRADE数据进行快速检索和分析也成为当前亟待解决的问题。文中研究并设计了一种高效的索引框架,该框架构建了以词对齐混合编码(WAH)为基础的位图索引,通过对COMTRADE数据构建元数据并建立相关索引,能够实现高速的电力数据检索,可以解决传统搜索技术的性能问题。该框架通过引入两阶段提交协议(2PC)来保证原始数据和元数据,元数据和索引文件的一致性。在与传统的MySQL关系数据库的比较测试中,所提方法性能更为优越,特别在对多维数据多条件和浮点数检索时,优势更为明显。该框架能够满足当前海量COMTRADE数据检索的需要,并可应用于电力行业其他大数据的存储检索和分析。
With the wide application and technical progress of fault recording and travelling wave devices, the IEEE COMTRADE 1 999 is apparently going massive.How to rapidly retrieve and analyze massive COMTRADE data has become a problem demanding prompt solution. An index framework based on word-aligned hybrid (WAH) bitmap for massive COMTRADE data and capable of solving the traditional indexing defect latency and delayed response is built.By introducing the two phase commit protocol(2PC),the consistence between index and metadata,metadata and data is ensured.The test is done between bitmap based on WAH and MySQL,and the results show that the bitmap index based on WAH is more effective than MySQL”s,especially for multi-dimensional,multi-condition and floating-point COMTRADE data indexing.This index framework meets the need of COMTRADE data as well as those other big data of power industry.
出处
《电力系统自动化》
EI
CSCD
北大核心
2015年第14期126-132,共7页
Automation of Electric Power Systems
基金
国家自然科学基金资助项目(U1202233
51267009)~~
