Abstract: With the development of smart grid, the mass data collected for the equipment conditioning requires higher performance of data store and data inquiry. A storage system based on Hadoop for smart grid condition monitoring was designed and achieved. The system consists of three parts; Hadoop cluster, storage client and inquiry client. A quantity of experiments were carried out including benchmark on hadoop cluster, verification for storage effect and analysis on query performance, which validate that the system has the advantages of distributed, mass storage and efficient inquiry,and is suitable for dealing with the mass data in condition monitoring of power equipments in smart grid.

Key words: Hadoop, HDFS, MapReduce, Power equipment, Condition monitoring, Smart grid