Prof. Sarika Khushalani Solanki
Associate Professor
Lane Department of Computer Science and Electrical Engineering
West Virginia University,
Morgantown, WV, USA
Sarika Khushalani Solanki (Senior Member, IEEE) received Ph.D. degree in electrical and computer engineering from Mississippi State University, Starkville, MS, USA, in 2006. She is currently an Associate Professor with the Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV, USA. Prior to that, she worked for Open Systems International Inc., Minneapolis, MN as a Senior Engineer. Her research interests include smart grid, modeling and analysis of distribution systems and microgrids, and data mining and AI applications to power systems. Dr. Solanki is the recipient of prestigious NSF CAREER award, Statler College best educator award and her students have received best paper and poster awards at IEEE conferences. She served as Chairman for IEEE PES Distribution Systems subcommittee and Power Engineering Career Promotion and Workforce Development Subcommittee. She also served as editor for IEEE Transactions on Smart Grid and MDPI Energies and was a co-chair of 49th North American Power Symposium.
Keynote Speech: Artificial Intelligence Driven Distribution Grid
Abstract: Increased penetration of renewables (wind and solar) will continue to stress grid and our objective for future distribution systems is seamless operation and high reliability even under this growing stress. A key challenge with high penetration of these renewable sources and inverter-based distributed energy resources in the power distribution systems is that inverters normally have zero inertia causing a high rate of voltage and frequency variability and hence faster dynamics. The bane of this variability, leading to uncertain,unpredictable and uncontrollable system response could eclipse the very benefits of environmentally friendly and low-cost renewable energy. It is imperative for distribution system analysis to account for the variability and faster dynamics thereby requiring quick and smart decision making. It is challenging to account for variabilities and hence model-free solutions are desired. Can machine learning, deep learning and in general artificial intelligence allow real time decision making and analysis while also mitigating the challenges posed above? This talk is thus focused on data-driven, scalable, computationally-efficient, distributed analysis methods that provide situational awareness through the exploitation of underutilized massive synchronized data now available through various measuring devices.