Recently, the research team of Professor Zhao Junhua in the School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, publishes a paper titled “An Inertia-based Data Recovery Scheme for False Data Injection Attack” in IEEE Transactions on Industrial Informatics, a top journal of IEEE Trans series.

Journal Introduction

IEEE Transactions on Industrial Informatics focuses on knowledge-based factory automation as a means to enhance industrial fabrication and manufacturing processes. This embraces a collection of techniques that use information analysis, manipulation, and distribution to achieve higher efficiency, effectiveness, reliability, and/or security within the industrial environment. The scope of the Transaction includes reporting, defining, providing a forum for discourse, and informing its readers about the latest developments in intelligent and computer control systems, robotics, factory communications and automation, flexible manufacturing, vision systems, and data acquisition and signal processing. IEEE Transactions on Industrial Informatics is the JCR Q1 level and its latest impact factor is 10.215.

Research Background

Due to vulnerabilities exposed to cyberattacks in the cyber physical power system, increasing concerns have been paid to its cybersecurity, especially on the so-called false data injection attack. Timely recovering true values of measurements and states after encountering cyber-attacks is of paramount importance for ensuring the subsequent controls and operations of the cyber physical power system. However, most of the existing related researches focus on the detection methods while studies of data recovery are rarely considered.

Research Method

This paper, among the first, discovers a power system operation phenomenon defined as measurement data inertia (MDI). Different from the well-known inertia in power systems, this phenomenon focuses on the measurement data, and mathematically describes that the measurement data, in the long run, alters regularly and smoothly, and the current measurement value is greatly associated with the preceding values.

The author uses the MDI phenomenon to derive the coarse values of the pre-attack measurements, and further considers the power system operating states and constraints, proposing a data recovery optimization model to accurately recover measurements and state variables to the pre-attack level.

Moreover, the author considers that each state variable has different variation rules, using common error indicators such as relative error (RE) solely may cause unfair evaluation, especially on the data whose amplitude is close to 0. Therefore, an interval error (IE) criterion is proposed based on the state operation interval to evaluate the data recovery quality.

Research Conclusion

In this paper, we propose a data recovery scheme to recover measurements and states contaminated by cyberattacks. The proposed method takes advantage of the discovered MDI effect to deduce coarse values for the pre-attack measurements, then these coarse values are used in the proposed recovery model to approach the true pre-attack values. Meanwhile, the predetermined state bounds are also exploited as constraints in the proposed model to suggest normal fluctuations for pre-attack state variables. An error criterion called IE is also proposed to assess the performance of the proposed recovery scheme. Moreover, the proposed recovery scheme is extensively evaluated on the IEEE 30-bus test benchmark. From the numerical results, we can conclude that the proposed recovery scheme can exactly recover contaminated measurements and states to the pre-attack values, exhibiting high recovery accuracy for cyberattack contaminations. Meanwhile, the case studies also show that the required time for the recovery processes is short enough without impacting on the later operations, thus ensuring the proposed method to be an in-time recovery strategy after encountering cyberattacks.

Author Biography

Prof. Junhua Zhao and Prof. Gaoqi Liang of The Chinese University of Hong Kong (Shenzhen) are the co-corresponding authors of this paper.

Professor Junhua Zhao is an Associate Professor in CUHK(SZ), the Director of Energy Markets and Finance Lab, Shenzhen Finance Institute, and a Scientist at Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS). He joined CUHKSZ in 2015. Before joining CUHKSZ, He was a Senior Lecturer and also acted as the Principal Research Scientist of Centre for Intelligent Electricity Networks, the University of Newcastle, Australia. He has 11 years of experience in the power industry in Australia. His research area includes smart grid, electricity market, energy economics, data mining and artificial intelligence. He published more than 200 research papers, including more than 100 papers in SCI, and 50 papers in IEEE Transactions. His published papers have been cited more than 9600 times, with an H-index of 49 (based on Google Scholar statistics).

He has rich practical experience in the power and energy industry, and his research results have had a significant impact on the industry. He deeply participated in the rule design and market construction of China's first electricity spot market. The three software products he was involved have been applied in leading energy companies such as New York Con Edison, Hong Kong Electric Company, Guangdong Energy Group Corporation Ltd, China Datang Corporation Ltd, and China National Offshore Oil Corporation. He has led or participated in more than 30 important research and industrial projects, which include the 'Smart Grid, Smart City' trial program funded by Australian government, two flagship cluster programs funded by Commonwealth Science and Industrial Research Organization (CSIRO), and one Electric Power Research Institute (EPRI) project.

Professor Zhao is the external experts of 'Australian National Outlook', the co-chair of the IEEE Special Interest Group (SiG) on Active Distribution Grids and Microgrids as well as the Secretary of the Asia Pacific Working Group of the IEEE PES SBLC (Smart Building, Load and Customer). And he is a member of a Global Smart Grid Federation (GSGF) working group. Besides, he is a member of the 'Interfaces of Grid Users/ Focus on EV and Local Storage' working group, Smart Grid Australia (SGA) working group, 'Cyber Physical Security of the Smart Grid' group and 'Critical Infrastructure Program for Modelling and Analysis'(CIPMA) expert group. He is the vice-chair of the Shenzhen AI industry society's expert committee. He is the editorial board members of IET ENERGY CONVERSION AND ECONOMICS, JOURNAL OF MODERN POWER SYSTEM AND CLEAN ENERGY, ELECTRIC POWER COMPONENTS AND SYSTEMS and POWER SYSTEM PROTECTION AND CONTROL. And he is the expert reviewer of Australian Research Council (ARC), National Natural Science Foundation of China Reviewer and Hong Kong Research Grants Committee (RGC). Also, he is the reviewer of IEEE Transactions on Power Systems, IEEE Transactions on Smart Grid, IEEE Transactions on Neural Networks and Learning Systems, Applied Energy, IET Generation, Transmission & Distribution. In addition, Professor Zhao also won the second prize of Zhejiang Province Natural Science Award in 2020. He is also an energy industry expert specially appointed by China Merchants Bank.

Professor Gaoqi Liang received her B.S. and Ph.D. degrees from North China Electric Power University and University of Newcastle in Australia in 2012 and 2017 respectively. She is currently a Research Assistant Professor in the School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen. She has published 34 academic papers. At present, 4 papers have been selected as "Essential Science Indicators (ESI) Highly cited articles". Her Google Scholar have been cited for more than 1900 times, with h-index of 10 and i10-index of 13. Her research interests include smart grid, electricity market, and their cyber-physical security; machine learning and its cyber security in smart grid, etc. She had participated in writing White Book on The Development of Artificial Intelligence Industry in Shenzhen and Blue Book on Green Development of Guangdong-Hong Kong-Macao Greater Bay Area (2019); participated in the establishment of IEEE Standard P2781 "Load Modeling and Simulation"; participated in the development of the first set of commercial electric power spot market simulation software platform NEMS in China, and has been successfully applied in Guangdong Energy Group, Cnooc, Datang Power Generation and other large energy enterprises.

The first author of the paper: Jiaqi Ruan

Jiaqi Ruan graduated from Shenzhen University with a master's degree in 2019. In the same year, he joined the research group of Prof. Junhua Zhao in The Chinese University of Hong Kong, Shenzhen, to pursue the Ph.D. degree. His research interests include cybersecurity and demand response in smart grids, and currently he has published 4 SCI papers.