基于Hadoop的智能电网时序大数据处理方法  

Hadoop-Based Smart Grid Time Series Big Data Processing Method

作　　者：孙利宏[1] SUN Li-hong(Ningixa Medical University,Yinchuan Ningxia 50001,China)

机构地区：[1]宁夏医科大学,宁夏银川750001

出　　处：《计算机仿真》2020年第12期67-71,共5页Computer Simulation

摘　　要：为了提高电网数据处理的安全性和效率,提出基于Hadoop的智能电网时序大数据处理方法。依据智能电网时序大数据简析,结合Map、Reduce及Partition三个函数具备的过滤器和工厂法以及监听器等一系列模式,实现数据清洗处理。依据分布式计算思想,结合近邻分类法和Map-Reduce模型设计的并行分类混合法,实现数据分类处理。对分类数据进行安全存储,通过消息摘要算法针对需要存储的智能电网时序大数据生成相应数字摘要;根据密钥生成函数获取随机密钥,同时利用上述密钥针对待存储数据实行加密,获取对应密文。针对获取的随机密钥实行信息隐藏处理;把密文存储至云中;当密文成功存储至云之后,把获取的密钥和数字摘要两种信息并同文件名至HBase中,实现数据存储。仿真结果表明,上述方法具有较强的安全性与时效性。In order to improve the security and efficiency of processing power grid data, a method of processing the time series big data in smart grids based on Hadoop was presented. Based on a brief analysis for time series big data, the filters, factory methods and listeners provided by the Map, Reduce, and Partition functions, the data cleaning was completed. According to the idea of distributed computation, the neighborhood classification method was combined with the parallel classification hybrid method designed by Map-Reduce model to achieve data classification. The secure storage of the classified data was carried out. For the smart grid time series big data to be stored, the message digest algorithm was adopted to form corresponding digital abstract. The random key was generated by key function. Based on this key, the data that would be stored was encrypted to obtain the corresponding ciphertext. For the random key, the information hiding was implemented. The ciphertext was stored in the cloud. After that, the key and digital abstract were combined into the same file name in HBase. Thus, the data storage was achieved. Simulation results prove that the proposed method has strong security and temporal sequence.

关 键 词：智能电网 时序大数据 处理 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]