CAREER:Probe-to-Learn Power Distribution Networks

Virginia Polytechnic Institute And State University, Blacksburg VA

Investigators

Abstract

The goal of this CAREER proposal is to develop a comprehensive data analytics framework for distribution grids, engage students in the exciting field of energy engineering, and advance education on distribution systems. To accomplish any meaningful grid-wide optimization task, the distribution system operator will need to precisely know the power consumed or generated at every node, the line and transformer parameters, and the topology of the grid. However currently, limited instrumentation and their sheer size render power distribution grids unobservable in general. To complement the smart metering and grid sensing infrastructure, the novel technique of grid probing is introduced here. Grid probing can be accomplished by commanding smart inverters residing at a subset of nodes to intentionally yet instantaneously perturb their power injections. Actuating the networked physical system and subsequently sensing the incurred voltages at metered nodes can unveil non-metered loads and network topologies. The original and potentially transformative idea of engaging smart inverters outside their intended control functionality pioneers new grid data analytics. This program goes beyond probing and maps the observability limits by processing smart meter data too. Contemporary optimization schemes together with their real-time and decentralized variants will deal with streaming and spatially incomplete voltage and power injection data collected at large-scale unbalanced distribution networks. On the broader impact, this program lies at the nexus of power systems, nonlinear system identification, and statistical learning; advances our understanding of unique patterns in grid data; and discovers novel uses for inverters in grid monitoring and control. Enhancing situational awareness in distribution grids benefits society by enabling the efficient and reliable integration of distributed renewable energy, electric vehicles, and customer participation. The expected benefits to industry are cutting-edge grid analytics solutions, advanced asset management, and value added for inverters and smart meters. The proposed activities bring about a potentially transformative paradigm for grid data processing, neatly integrated with educational and outreach objectives. The program reaches pre-college female students through hands-on learning activities and collaborative app-based games on electric grids. It also involves undergraduate research and continues our outreach to undergraduate students through visits to residential communities. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.