Henry L. Pierce Laboratory Seminar Series

Data-Driven Learning in Power Distribution Networks
Increase in supply side variability due to increases in renewable generation require demand side management strategies to reduce electricity delivery costs. Smart grid technologies provide opportunities for measuring and controlling distributed energy resources such as loads and storage at an unprecedented scale reducing the electricity delivery cost. Enabling this solution requires accurate models of the power distribution network and how consumer loads respond to various signals. This talk introduces data driven tools to resolve these two challenges. VISDOM (Visualization and Insight for Demand Operations and Management) is an open source framework to learn consumer behavior utilizing real experiments and observing smart meter data. The proposed methodology utilizes features derived from the data to postulate behavior models, and various algorithms to characterize consumption statistics, segment consumers and target the appropriate ones to programs. VADER (Visualization and Analytics for Distributed Energy Resources) is a platform that learns models of the distribution network fusing traditional utility data such as from SCADA with novel sources of information such as measurements from inverters and smart meters. We demonstrate how machine learning and statistical analysis can yield theoretical sound and more accurate models of the system as compared to traditional approaches.