Smart Meter Data Sharing for AI-Enhanced Energy Systems: A Review of Relevant Techniques and Detailed Case Studies

Abstract

Digitization is a prevailing trend in modern energy systems. With advancements in information and communications technology (ICT), advanced metering infrastructures, such as electric meters and gas meters, have been developed to record fine-grained energy consumption data. As a result, an increasing amount of data can now be accessed and collected, ranging from basic load, voltages, and gas readings to tamper indication and outage records, etc. With advanced metering infrastructures, companies and operators can collect smart meter data with high accuracy and low latency. The appropriate utilization of smart meter data can result in significant benefits. By analyzing smart meter readings using data mining techniques, various functionalities can be enhanced, including behavior modeling, load forecasting, generation forecasting, etc. The installation of smart meters also enables demand response (DR) programs where utility companies can remotely modify the energy supply to meet demand variations. The increasing availability of smart meter data also boosts the development of artificial intelligence (AI), a fast-developing technology that demands a large amount of data to effectively learn and make accurate decisions. These insights can improve performance and efficiency in the power and energy systems.