Research
Research Focus
My research group focuses on the cyber-physical vulnerability, security, and resiliency of smart infrastructures, particularly with artificial intelligence systems. Since 2018, I have received over $3.5 million in research funds as a PI or co-PI in these areas. The research aims to develop scientific knowledge and engineering solutions against known vulnerabilities that have long challenged critical infrastructure security and zero-day vulnerabilities that can be exploited by cyber and/or physical attacks amidst the digitalization of various systems and processes.Â
The key methodology will consist of power system security, complex network theories, and computational intelligence algorithms, which will assist in vulnerability assessment, attack analysis, and defence response of the smart critical infrastructures. To this end, my group has been investigating computational intelligence techniques, including deep learning, feature learning, reinforcement learning, and transfer learning, and has developed advanced vulnerability analysis, adaptive penetration tests, and accurate attack detectors, among others.
The ultimate goal of my research is to 1) establish a comprehensive knowledge base for complex and intelligent attack threats; 2) develop automated threat monitoring and risk assessment strategies; and 3) facilitate rapid decision supports for the protection, detection, mitigation, and restoration against contingencies and adversaries for smarter and securer infrastructures.
CPS Vulnerability
Physical systems have inherent structural and dynamic vulnerabilities, as we have seen in wide-area blackouts triggered by single contingencies. Cyber integrations and physical interconnections of our infrastructures have made modern CPSs increasingly interdependent. Through the inter-dependencies, a single contingency can propagate across multiple regions and infrastructures. To prevent and mitigate failure propagation within and across infrastructures, my research works on vulnerability identification, risk assessment, and situation awareness, which are becoming crucial in modern power systems and other CPS.
My research on CPS vulnerability investigates:
Multi-contingency screening and analysis
Modeling and risk analysis of cascading outages
Interdependent networks and resilient smart grid
Topological and hybrid analysis of CPS as complex networks
Vulnerability and risk visualization
CPS Security
With integrations of information and communication technologies, cyber-physical systems like the Smart Grid are no longer independently operated islands off-the-grid. The emerging access via communication channels and advanced metering infrastructures (AMI) unfolds unforeseen cyber-security challenges, as in the alerting 2015 cyber-attack on the Ukraine power grid. To protect our critical infrastructures against malicious attacks, my research works on the analysis of potential attack threats built on knowledge of system states and vulnerabilities and proposes effective strategies to detect, mitigate, and resolve the threats.
My research on CPS security investigates:
Multi-target synchronous attacks
Multi-target sequential attacks
Multi-type coordinated attacks
False data injection attacks
Spatial-temporal analysis of post-attack behavior
Data-driven intelligent detector design