Quantum mechanics

Engineering

Cyber-physical system

Automation

Embedded system

Computer security

Network security

Operating system

Anomaly detection

Power (physics)

Data mining

Computer science

Protocol (science)

Computer network

Mechanical engineering

Alternative medicine

Physics

Malware

Electric power system

Artificial intelligence

Intrusion detection system

Medicine

IEC 61850

Artificial neural network

Recurrent neural network

Pathology

IEEE Access

The introduction of the cyber-physical system (CPS) into power systems has created a variety of communication requirements and functions that existing legacy systems do not support. To this end, the IEEE 1815.1 standard defines the mapping between existing distributed network protocol networks and IEC 61850 networks that reflect new requirements. However, advanced CPS cyberattacks have been reported, and in order to address cyberattacks, security research on new power systems that use network devices and heterogeneous communication is necessary. In this study, we propose an intrusion detection system for an IEEE 1815.1-based power system using CPS. We 1) analyze an IEEE 1815.1-based power system network and propose a suitable application method for an intrusion detection system, 2) suggest a bidirectional recurrent neural network-based anomaly detection system for an IEEE 1815.1-based network, and 3) demonstrate the verification of the proposed technique using various power system-specific attack data, including real power system using CPS network traffic, CPS malware behavior (CMB), false data injection (FDI), and disabling reassembly (DR) attacks. Proposed technique successfully detected five types of CMB attacks, three types of FDI and DR attacks.

IEEE 1815.1-Based Power System Security With Bidirectional RNN-Based Network Anomalous Attack Detection for Cyber-Physical System