Research On Data Integration And Storage Of SCADA System Based On Cloud Computing

SCADA is the main source of real-time operation information in power grid. The complete SCADA data should include all levels of SCADA data. In order to play them in power grid planning, design, renovation, equipment monitoring and other aspects role,they must be integrated. However, the SCADA system is developed by different manufacturers at different time and different database management systems are used.The real-time data can not form valuable decision-making information. And as the different manufacturers don’t have a standard data interfaces, the external interfaces are not consistent. This makes other application system have to develop a separate interface for data transmission when they need to call for the data. Therefore, the system interconnect are very complex, and cannot be reused.To solve the problems above and achieve integration share of data for heterogeneous SCADA systems, the paper study heterogeneous data integration technology based on the IEC61970standard and cloud computing technology for mass data processing and storage and complete the following aspects of work.First of all, to improve information sharing, and to make transparent access to heterogeneous data and other considerations for users, we study data integration technologies and IEC61970standards, and made the design of a standards-based CIM model for heterogeneous data integration platform. We use Hbase to realize the storage of CIM model and complete data mapping between the existing SCADA system and the CIM model. According to the model based on CIM designed in the paper, when a database is added or deleted, we just needs to change the module infotmation and mapping relation stored in global database.So it meet scalability requirement on data of integration platform.Second, based on the integration model of heterogeneous data designed in this paper,I analysised middleware layer’s function of the integrated platform, achieved a MapReduce-based query processing. Template-based mapping method is used to realize the mode conversion function of the wrapper.Finally, in order to verify that the data integration platform can meet the requirements on fast data processing and mass storage, we set up a small Hadoop cluster and distributed database HBase for the experiment. we prompt data import, data query and processing of heterogeneous data integration experiments, and the experimental results were analyzed to verify the feasibility efficiency of this project.