Intrusion Detection System for Smart Meters in Power Distribution Network Based on Statistical Anomaly Detection

Intrusion Detection System for Smart Meters in Power Distribution Network Based on Statistical Anomaly Detection

Mirza Šarić

Amila Jašari

Ammar Arpadžić

Abstract – Recently, there have been significant changes in the power system operation and management. These changes are driven by deregulation and market liberalization policies and include significant technical innovations and introduction of information and communication technologies to power systems. The introduction of smart meters and advanced metering infrastructure (AMI) presents new challanges. Two most sensitive aspects of smart meters are electricity consumption data and remote switch access. This paper presents and overview of security challanges for the smart meters used in power distribution system and proposes a fuzzy intrusion detection system for smart meters in power distribution system based on statistical anomaly detection, based on Mamdani type fuzzy inference system with two input (Elapsed Session Time and Login Frequency) variables and one output variable used as intrusion detection control signal. It is argued that this paper makes a contribution towards understanding and design of comprehensive intrusion detection technologies for smart meters.

Keywords - AMI, Power distribution network, fuzzy logic, intrusion detection, risk model, smart meter.

International Burch University

Journal of Natural Sciences and Engineering