Project Details
Abstract
Recent investments in power grid infrastructure are moving the traditional power grid toward the adoption of smarter and secure devices. These devices i.e. smart meters, phasor measurement units, smart transformers, and smart inverters possess many attractive features that have not been yet fully utilized. They can significantly enhance the flexibility, observability, and controllability of the power grid, which consequently helps in transferring the existing power infrastructure to the next generation with advanced ancillary services for the grid. However, such power grid with smart devices could be a subject of cyber and incident attacks since there is a communication medium among these devices in order to operate smartly and in harmony. Hence, there is a need for a self-learning, attack-resilient power grid and smart devices that are able to effectively detect and protect from cyber attacks, which is the aim of this project. The proposed project aims to leverage the team’s extensive research and experience in cybersecurity, deep learning, smart inverters and transformers, big data, autonomous control, and multi-agent distributed networked control to enhance the situational awareness and resiliency in power grid with critical infrastructure. The proposed multi-layer cyber-physical intrusion detection monitors both cyber and physical data (i.e. data from smart devices, PMUs, etc) to perform effective intrusion detection and proactive control response while providing secure communication and data integrity. Conventional intrusion detection based on cyber or physical data analytics could result in bad data going undetected, or in bad data detection occurring at a stage at which the malicious event already had severe effect on power grid and the country critical infrastructure. The proposed approach is to monitor both cyber and physical data from smart meters, PMUs, smart PV inverters, and other existing smart devices in the network to conduct a proactive, predictive, and effective intrusion detection at early stage and consequently execute autonomous corrective control action by smart devices. We envision the smart power grid supported by the proposed hierarchical system of cybersecurity analytics and secure communication, enabling quick cyber event detection, and an optimal response mechanism that ensures the minimal possible service disruption subject to customizable rules for timely cyber event isolation and handling. The proposed approach integrates deep learning methods, cybersecurity analytics at system level, high-fidelity model development, continuous monitoring of smart devices cyber and physical data for in-depth intrusion detection, and multi-agent distributed networked proactive control strategy to enhance the smart grid security and resiliency. The project clusters are spanning over three major goals: (G1) infrastructure cyber security and data integrity, (G2) cyber-physical intrusion detection, and (G3) proactive control actions and attack resiliency. The main concept of the proposed project to achieve these three major goals is to firstly outline the Qatar power grid and discover all the operational technology (OT) devices as well as IT devices; then secure communications and data collection will be developed; after that minimization of attack surface will be developed; the intrusion detection from system level to power grid fraction and finally at device level is proposed which provides proactive cyber-physical security for Qatar critical infrastructure; finally a recovery and autonomous control schemes will be developed in response to cyber-physical attacks. The unique aspect of the proposed project clusters (PC) are listed as: (PC1) discovering and protecting OT devices; (PC2) adaptive key exchange communication protocol; (PC3) big data security to enhance the data integrity at the system level and minimize private data exposure; (PC4) predictive and proactive security via smart deception for the smart grid infrastructure; (PC5) enhancing power grid situational awareness and cybersecurity analytics at system level using existing smart devices in the network while optimizing the number and locations of PMUs; (PC6) in-depth cyber-physical attack detection via smart devices and deep learning; and (PC7) autonomous control actions for smart devices in response to cyber-physical attack and other anomalies with resilient solutions for critical infrastructure. This project is a collaborative effort among Texas A&M University at Qatar (TAMUQ), Hamad Bin Khalifa University (HBKU), Qatar University (QU), Qatar Environment & Energy Research Institute (QEERI), Kansas State University (KSU), and Texas A&M University (TAMU) in close collaboration and support from industry partners particularly KAHRAMAA. The industry partners will provide the necessary feedback, data, and models to align the project outcomes with Qatar national vision 2030 and need. Furthermore, the industrial collaborators and co-funding partners will support the proposed project towards high technology readiness level (TRL), the expected TRL is 5/7. The QNRF support for this partnership enables collaboration between the Qatar academia. industry, and international institutions to move the frontier of technology in cybersecurity, intrusion detection, and situational awareness particularly for Qatar’s current power grid and future smart grid.
Submitting Institute Name
Texas A&M University at Qatar
Sponsor's Award Number | NPRP12C-0814-190012 / NPRP12C-33905-SP-71 |
---|---|
Proposal ID | EX-QNRF-NPRPC-23 |
Status | Active |
Effective start/end date | 1/01/21 → 1/01/26 |
Primary Theme
- Sustainability
Primary Subtheme
- SU - Sustainable Energy
Secondary Theme
- Sustainability
Secondary Subtheme
- SU - Resource Security & Management
Keywords
- Cyber security
- Smart grid
- Machine learing applications
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.