Profiling Users in the Smart Grid

The implementation of the smart grid brings with it many new components that are fundamentally different to traditional power grid infrastructures. The most important addition brought by the smart grid is the application of the Advanced Metering Infrastructure (AMI). As part of the AMI, the smart meter device provides real time energy usage about the consumer to all of the smart grids stakeholders. Detailed statistics about a consumer’s energy usage can be accessed by the end user, utility companies and other parties. The problem, however, is in how to analyse, present and make best use of the data. This paper focuses on the data collected from the smart grid and how it can be used to detect abnormal user behaviour for energy monitoring applications. The proposed system employs a data classification technique to identify irregular energy usage in patterns generated by smart meters. The results show that it is possible to detect abnormal behaviour with an overall accuracy of 99.45% with 0.100 for sensitivity, 0.989 for specificity and an error of 0.006 using the LDC classifier.