Hierarchical and Compositional Strategies for Control and Protection in PEPDS
Abstract
Approved for Public ReleaseThe objective of this proposal is to develop theory, algorithms and demonstrations for the automated synthesis of hierarchical and compositional control strategies for power distribution systems that are dense in power-electronic building blocks (PEBBs), also known as "power-electronic power distribution systems" (PEPDSs).The coordination PEBBs must be tightly performed at timescales under 10 microseconds for their groupings to perform power converter functions such as those needed for source and load interfaces. These tightly coordinated groups of PEBBs or clusters can also be coordinated at slower timescales to perform high-level system functions such as the steering of electrical energy within the power distribution system. The size of PEBB clusters and their modes of interactions are interrelated with the framework and algorithms utilized for system-wide configuration of the system and determination of energy flow paths to meet system level goals.The strategies that will be developed over the course of the proposed effort will enhance the system-level performance, e.g., load operability, in PEPDSs while respecting other safety- and mission-related specifications. They will address several cross-cutting concerns in the operation of such power distribution networks: (i) The hierarchical abstractions will help scale up the applicability to large number (e.g., 1000s to 10000s) of power electronics elements. (ii) They will help identify the level of abstraction (i.e., the extent of clustering of the PEBBs) necessary for each layerof the hierarchy so that a given power system functionality, or a collection of functionalities, can be delivered with the necessary performance and reliability. (iii) The compositional nature of the algorithms will help formalize the interfaces between the PEBBsand between the clusters of PEBBs and the information flow requirements (e.g., the content and bandwidth) so that the resulting distributed control strategies can be implemented. The proposed effort will build on existing low-level control strategies for the coordination of PEBBs so that a cluster of PEBBs deliver the intended functionality low-level functionality. It will then focus on higher-level distributed strategies that will control the clusters and/or set the functionality they need to deliver so that system-wide requirements are met. To this end, the team will build on their recent work on automated and compositional synthesis of control strategies for modular systems with models that allow for a wide range of uncertainties (e.g., stochastic and non-deterministic) and rich set of safety- and mission-related specifications (e.g., in so-called graph temporal logic with probabilistic modalities).
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 06, 2021
- Source ID
- N000142112164
Entities
People
- Ufuk Topcu
Organizations
- Office of Naval Research
- United States Navy
- University of Texas at Austin