Publication

Abstract : Exhilarating developments in the renewable energy generation portfolio and the widespread introduction of microgrids have contributed to significant reforms in the existing power grid’s power flow patterns. Power grids around the world have upgraded to smart grid (SG) by introducing advanced monitoring technologies such as Advanced Metering Infrastructure (AMI), smart meters and Phasor Measurement Units (PMUs), which collect high-resolution electrical measurements across the system. This has brought in a massive amount of data, termed big data, that needs to be efficiently processed to derive valuable insights. Specifically, technical sophistication, security, and integration of datasets are the main concerns that need to be tackled to transform the massive dataset into useful insights. This chapter surveys big data analytics (BDA) and its related benefits, challenges, and advancements in SG’s perspective. BDA in combination with visualization tools, help in the predictive decision-making process in the SG. Thus, data analytics play a significant part in the efficient monitoring of the SG. Powered with the integration of information and communication technologies, an information layer has been introduced to the traditional power systems to collect, store, and process data from smart meters and sensor implementations. Big data architecture involves data aggregation, storing, and analytics, which integrates the SG’s need for a range of frameworks with outstanding computational skills to respond to the customer’s needs. Characterization of big data, SGs, and massive volumes of data processing is first addressed as a preface to demonstrate the motivation and possible benefits of integrating advanced data mining in smart grids. Specific principles and standard data analytics techniques for general concerns are also discussed. The chapter’s key section discusses the advanced uses of various data analytics in SG, leveraging machine learning (ML) and artificial intelligence (AI). SG is benefited from the inherent capacity of ML to generalize, delivering reliable and quick power flow predictions from dispersed measurement units, with superior computing performance and interoperability. The chapter explores numerous literary works that utilize various ML methods to enhance SG activity and management. We will present big data analytics and cloud infrastructure in this chapter and address their importance to SG. Specifically, we would concentrate on relevant computational concerns and solutions linked to SG cyber-physical protection and safety.