Due to increasing penetration of solar photovoltaic generation and electric vehicle charging loads, there is a genuine need to closely monitor the voltage over the entire length of the distribution feeder. Smart meters, present only at the terminal nodes of the feeder, cannot fulfill this need; they also have high reporting delays.  Distribution phasor measurement units have the necessary speed, but it is cost- prohibitive to place them in bulk. Thus, monitoring voltages in real-time unobservable distribution systems is challenging. This talk will describe how the use of machine learning can help overcome this challenge by performing high-speed voltage estimation while accounting for the physical attributes and operational characteristics of modern distribution systems. To ensure trust in the machine learning-based approach, formal guarantees of performance will also be provided.

Anamitra Pal is an Associate Professor in the School of Electrical, Computer, and Energy Engineering at Arizona State University (ASU). His research interests include data analytics with a special emphasis on time-synchronized measurements, artificial intelligence applications in power systems, renewable generation integration studies, and critical infrastructure resilience. Dr. Pal has received the 2018 Young CRITIS Award for his contributions to the field of critical infrastructure protection, the 2019 Outstanding Young Professional Award from the IEEE Phoenix Section, the National Science Foundation CAREER Award in 2022, and the 2023 Centennial Professorship Award from ASU.

Host: Rajasekhar Anguluri