Methods for generating TLPs (typical load profiles) for smart grid-based energy programs

Abstract

Most electric power companies implement an automatic meter reading (AMR) system operating at quarter-hour intervals. The companies install the system (an electric meter) at the homes of their high-voltage (HV) customers who consume a significant amount of power each month. When first introduced to the industry, the AMR system simply measured customers' peak power and billed them. Recent studies are examining the same system's applicability in the cutting-edge smart grid technology. A growing number of studies are focusing on AMR-based distribution network load analysis and demand prediction to promote the dissemination of smart grid-based services. Researchers are basically using AMR customers' usage data to analyze loads and generate the virtual load profile (VLP) of non-automatic meter reading (nAMR) customers. Generating VLP requires clustering and classification that are among the various data mining techniques adopted by researchers. This study reviewed previous research findings that reported AMR-based typical load profile (TLP) generation, and utilized the AMR data of some KEPCO HV customers for TLP generation. Analyses were performed via three clustering techniques, and the strengths of the techniques were compared.