Anomaly Detection Algorithms for Smart Metering using Swarm Intelligence

Abstract

Advancement in the information and communication technology has introduced Advanced Metering Infrastructure (AMI) in the electricity metering system, which has replaced old mechanical meters with smart electric meters. This modernization also introduced a lot of scope for the different anomalies and attacks on smart meters. Hence to tackle these challenges, we have proposed three anomaly detection algorithms (VBA, HBA, KBA) which are truly based on the principles of Swarm Intelligence (SI). The swarm intelligence is the emerging subbranch of artificial intelligence which studies the collective intelligence of groups of simple agents. The theory is corroborated by its performance in terms of probability of detection and probability of false alarm. The proposed algorithms entrust the probability of detection and probability of false alarm close to 1.00 and 0.17 respectively.