Attributes of Big Data Analytics for Data-Driven Decision Making in Cyber-Physical Power Systems

Big data analytics is a virtually new term in power system terminology. This concept delves into the way a massive volume of data is acquired, processed, analyzed to extract insight from available data. In particular, big data analytics alludes to applications of artificial intelligence, machine learning techniques, data mining techniques, time-series forecasting methods. Decision-makers in power systems have been long plagued by incapability and weakness of classical methods in dealing with large-scale real practical cases due to the existence of thousands or millions of variables, being time-consuming, the requirement of a high computation burden, divergence of results, unjustifiable errors, and poor accuracy of the model. Big data analytics is an ongoing topic, which pinpoints how to extract insights from these large data sets. The extant article has enumerated the applications of big data analytics in future power systems through several layers from grid-scale to local-scale. Big data analytics has many applications in the areas of smart grid implementation, electricity markets, execution of collaborative operation schemes, enhancement of microgrid operation autonomy, management of electric vehicle operations in smart grids, active distribution network control, district hub system management, multi-agent energy systems, electricity theft detection, stability and security assessment by PMUs, and better exploitation of renewable energy sources. The employment of big data analytics entails some prerequisites, such as the proliferation of IoT-enabled devices, easily-accessible cloud space, blockchain, etc. This paper has comprehensively conducted an extensive review of the applications of big data analytics along with the prevailing challenges and solutions