Research And Application Of Smart Grid Data Cloud Storage Based On NoSQL

Due to the depletion of energy recently, the research on smart grid are undertaken by many countries.Smart grid is the intelligent power grid which needs to master various types of more real-time and more comprehensive grid data in order to support a more secure and reliable operation of the grid. Massive grid data are generated in transmission, substation, distribution, utilization of electric power, scheduling and other aspects of the smart grid power.high performance in storing these data in a efficient,reliable, inexpensive way and querying quickly is an important research topic currently.High scalability is required in storage system for storing massive smart grid data clearly, and No SQL storage is designed to meet the demand. According to a variety of smart grid data structures, huge amounts of data volume and high real-time requirements, Mongo DB database is put forward to store data which will store and query massive smart grid data efficiently.This dissertation describes Cloud computing, cloud storage and No SQL database technology. Then Mongo DB database, a kind of No SQL database which is popular nowadays is studied deeply on data structure, replication set and fragmentation mechanism. And a improved data balancing algorithm is proposed. Later cloud storage platform based on Mongo DB database is designed and Mongo DB database cluster is built. Taking electricity data of residential customers as example, finally the mass data is stored in the database cluster, storing and querying performance are compared between Mongo DB database and the traditional relational database. YCSB is used for the benchmark test of Mongo DB. The result indicates the advantages of No SQL databases in the mass data storage.