Project summary

Large power transformers are critical for power grids to supply electricity to Australian residents, industries and business. These transformers are connected to condition monitoring and management systems through various communication platforms. This opens potential pathways for cyber-attacks, and may lead to malfunction and damage of power grids. In partnership with Powerlink and Energy Queensland, this project investigates the components of the transformer which are prone to cyber- attacks, modeling potential attack routes and developing machine-learning techniques against these threats. This project paves the way for developing defence strategies to protect transformers from cyber-attacks, ensuring the security of Australia’s power grids.


Project description

Large power transformers are critical and expensive assets (several million dollars each) in power grids. Australia’s power grids rely heavily on these transformers for supplying electricity to residential, commercial, and industrial customers.

With the ongoing digitalization of power grids, various sensor-based condition monitoring and protection systems have been installed on power transformers. In parallel, operation management systems at control centres can remotely monitor and run these transformers through sensing/control signals coming through their communication channels. Therefore, without appropriate protective measures, these transformers are vulnerable to cyber-attacks leading to serious consequences including malfunction, damage and even explosion of transformers. Therefore, it is necessary to investigate transformers’ cyber vulnerabilities and develop defence strategies.

Through condition monitoring systems and communication infrastructure, transformer conditions (voltage/current, temperature, fault gasses etc.) are sent to control centres. If they are tampered with, a false alarm signalling the need for maintenance of a transformer could pose a hazard. Moreover, transformers require some active control mechanisms. If these mechanisms are targeted, transformers can be overloaded, overheated, and explode. Furthermore, transformers are protected against faults through digital relays that are connected to the internal network. If relay settings are altered, transformers may not be disconnected from the power grid when a fault occurs and may result to a catastrophic failure of the whole power grid.

Efforts from the smart grid research community have been undertaken to address the cyber threats to power grids. However, cyber-security of power transformers has not been sufficiently addressed. Cyber-security functions have not been found in commercially available condition monitoring systems in transformers. This project aims to understand whether large power transformers encounter feasible cyber-attacks, whether they are exposed vulnerabilities to attack, what would be the threat, and how and to what extent these could happen. And how these attacks can be detected and mitigated will also be investigated in this project.

This project investigates security vulnerabilities and potential threats to large power transformers through:

1) Cyber-physical components identification including transformer’s components, condition monitoring systems and communication proxy, gateway and data types and flows.

2) Threat modelling based on STRIDE (spoofing, tampering, repudiation, information disclosure, denial of service, and elevation of privilege).

3) Attack modelling for understanding success rates of potential cyber-attack route.

4) Supervised learning algorithms to detect possible cyber intrusion.

The facilities in UQ’s Industry 4.0 Laboratory and Australasian Transformer Innovation Centre will be used as test beds. Transformer condition monitoring datasets collected from industry partners, Powerlink and Energy Queensland, will be utilised.

The research is in line with one of the National Research Priorities - Safeguarding Australia. Protecting critical energy infrastructure has been recognized as a priority goal significant to national security and to the social and economic wellbeing of Australia. This project will prompt collaboration with industry partners to apply for various funding (e.g., ARC Linkage Project, Energy Network Australia fundings, Australian Federal Government Cyber Security Skills Partnership Innovation Funds, Australian Renewable Energy Agency fundings etc) to develop defence strategies to ensure the secure utilisation of power transformers in Australia’s power grids. 


Publication

View publications


Partner organization(s)

Powerlink

Project members

Lead investigator:

Dr Hui Ma

Senior Lecturer in Power & Energy
School of Electrical Engineering and Computer Science

Other investigator(s):

Dr Chandima Ekanayake

Senior Lecturer
School of Electrical Engineering and Computer Science

Associate Professor Guangdong Bai

Associate Professor Software Engineering
School of Electrical Engineering and Computer Science

Dr Nan Ye

Senior Lecturer
School of Mathematics and Physics