Application of Data Mining in Smart Grid Technology

Introduction

Smart grid technology is a radical and creative approach for improvisation in existing passive power grid. Assimilation of power and communication infrastructures is inevitable for the design and deployment of an imminent smart grid technology. The grid is called ‘Smart’ as it is accomplished with intelligent devices for decision making. Some of the substantial features of smart grid technology are bidirectional communication, renewable integration, active consumer participation, distribution automation, advanced metering infrastructure, supervisory control and data acquisition and complete management of entire power grid through wireless communication protocols and technologies. Smart grid communication infrastructure is a gigantic architecture with massive amount of data communication between various hierarchical network layers (Farooq & Jung, 2014). Full duplex data communication is required for monitoring, control, security, management and fault diagnosis of entire power grid (Mahmood et al., 2015). Management of complex, hierarchical and heterogeneous smart grid communication infrastructure necessitates data collection, storage, processing, analysis, retrieval and communication for self-healing and complete automation. Collection, interpretation and analysis of massive quantity of data generated in smart grid operation are critical, inevitable and complex tasks (Gungor et al., 2010). Data mining techniques can be an effective solution for smart grid operation and management. Data mining is essential for converting information into knowledge (Wijaya, T., 2013). Knowledge discovery and data mining are interdisciplinary tasks in the context of smart grid network. Database management and data mining are the two significant aspects of a database system. Database management deals with processing and storage of data and data mining deals with discovery and abstraction of knowledge for the purpose of decision making. Data mining is a computational process for data scrutiny and analysis. Data analysis is unavoidable for unambiguous knowledge discovery as well as decision making practices. Data mining is necessary for analysis of various statistics associated with power generation, distribution automation, data communications, billing, consumer participation, and fault diagnosis in smart power grid. In Smart grid, accumulation of real time data is inevitable. Data mining techniques are required when data is continuously collected on real time basis (Atzmueller et al., 2013). The pattern of gigantic sets of data can be effectively extracted and scrutinized with the use of different data mining techniques for energy efficiency, reliability and real time decision making. This chapter is expected to serve as a comprehensive analysis and review of application of data mining techniques in generation, transmission, distribution and utilization of energy in smart grid infrastructure.