Integrated Intelligent Cyber-Physical Fault-Containment Control for Resilient PEPDS
Abstract
Approved for Public ReleaseResearch problem: Power Electronic Power Distribution Systems (PEPDS) are complex cyber-physical systems,(CPS), where the physical power electronics, power distribution, cooling, and communication systems are tightly coupled. The coordin,ation of distributed resources, such as energy storage modules (ESM) and power generation modules (PGM), provides surface combatants, the versatility of operating in different modes, especially in dealing with power system contingencies. The gains in the flexibilit,y of energy management and thus function are driving factors toward a PEPDS where all energy interfaces are modular power electronic,s. More integrated Power Electronics Building Blocks (PEBBs) will continually increase the degrees of freedom of PEPDS with respect,to energy management. However, the resulting complex CPS structure raises critical and interrelated control and fault problems. Firs,d, the simultaneous/rapid succession faults could make it challenging to distinguish fault types, and the system response may deteri,orate. Third, managing the effects of concurrent cyber-physical fault (CPF) events is not well understood even for Integrated Power,and Energy Systems (IPES). Published research with respect to the current IPES topologies has largely been reactive to occurring fau,lt events and addressed single faults. This is a barrier to leveraging the flexibility of PEPDS to circumvent fault conditions with,minimized impact. Current control methods do not respond in a coordinated manner to current and cascading CPF events. For example, i,f there are link outages (i.e., cables, converters) or malicious data propagation due to detection failures, the resulting power rou,ting in other sections will increase. If this redirected flow exceeds capacity limits, load rejections may occur and result in casca,ding failures and even system outages.Technical Approach: To address the above-mentioned challenges, the proposed research will deve,lop an innovative Integrated Cyber-Physical Fault-Containment Control framework by (1) developing novel Deep Graph Neural Network (D,GNN) methods for detecting and managing CPF events, and generating constraint operating spaces and system states to allow for fault-,containment controls; and (2) developing model-predictive constraint methods with D,lt-containment control system to prevent cascading failures and maximize the mission-critical load operability and system resilience,. The proposed project will innovatively addresses disruptive, multi-discipline cyber-physical fault events, which have fast dynamic,llowing major outcomes: 1) Documented vulnerability assessment of consequence to fulfillment ofship missions (resilience) associated, with cyber-physical faults; 2) An integrated cyber-physical fault containment approach based on DGNN and fast DRGL to ensure the re,silience of PEPDS; 3) Input to relevant standards (e.g. IEEE P45.2, IEEE 1676, IEEE 1709, IEEE 1826, IEEE 1662); 4) Publications: AS,NE conferences & journals, and prestigious journals (e.g., IEEE and Elsevier); and 5) Annual joint-reports and presentations in ONR,Controls Program Review and PEPDS Program Review.DoD Impact: This project will deliver to the Navy a new capability for PEPDS in the, preparedness, awareness, and response against cyber-physical threats, which could occur simultaneously and erratically. Therefore,,this proposed project will provide future Naval ship systems with additional resilience for its energy services used in mission-crit,ical loads during various operational stages when subject to disruptive cyber-physical fault events.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 05, 2022
- Source ID
- N000142212239
Entities
People
- Tuyen Vu
Organizations
- Clarkson University
- Office of Naval Research
- United States Navy