Smart meter data analytics applications for secure, reliable and robust grid system: Survey and future directions

The new power sector scenario has focused on integrating renewable energy sources into the grid and achieving trustworthy consumer-utility-stakeholder relationships using smart meters. Granular consumption data obtained from the data management systems of Advanced Metering Infrastructure helps power utilities build an individual consumer's electricity consumption profile that facilitates better load forecasting, efficient management of demand and supply, and more reliable service. However, different challenges concerned with accessibility of data through different communication layers, such as breaches of privacy of consumer data and security of grid have emerged. Due to this, many customers lose faith in the smart grid. To utilize the full potential of the smart grid system these issues need to be addressed to attain a secure, stable, and reliable grid that brings profit to the utility. This paper provides a comprehensive literature survey to achieve the above-mentioned purposes using data analytics. It also addresses different challenges of using data analytics methods and machine learning applications in mitigating these issues. This study would enable the electric utilities to structure a reliable grid to prevent revenue loss and aid the researchers to recognize the potential and challenges of machine learning techniques in the field of the power sector.