Date of Award
8-2023
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Electrical and Computer Engineering
Committee Chair/Advisor
Dr. Ganesh Kumar Venayagamoorthy
Committee Member
Dr. Rajendra Singh
Committee Member
Dr. Yingjie Lao
Abstract
As modern electric power distribution systems (MEPDS) continue to grow in complexity, largely due to the ever-increasing penetration of Distributed Energy Resources (DERs), particularly solar photovoltaics (PVs) at the distribution level, there is a need to facilitate advanced operational and management tasks in the system driven by this complexity, especially in systems with high renewable penetration dependent on complex weather phenomena.
Digital twins (DTs), or virtual replicas of the system and its assets, enhanced with AI paradigms can add enormous value to tasks performed by regulators, distribution system operators and energy market analysts, thereby providing cognition to the system. DTs of MEPDS assets and the system can be utilized for real-time and faster-than-real-time operational and management task support, planning studies, scenario analysis, data analytics and other distribution system studies.
This study leverages DT and AI to enhance DER integration in an MEPDS as well as operational and management (O&M) tasks and distribution system studies based on a system with high PV penetration. DTs have been used to both estimate and predict the behavior of an existing 1 MW plant in Clemson University by developing asset digital twins of the physical system. Solar irradiance, temperature and wind-speed variations in the area have been modeled using physical weather stations located in and around the Clemson region to develop ten virtual weather stations. Finally, DTs of the system along with virtual and physical weather stations are used to both estimate and predict, in short time intervals, the real-time behavior of potential PV plant installations over the region. Ten virtual PV plants and three hybrid PV plants are studied, for enhanced cognition in the system. These physical, hybrid and virtual PV sources enable situational awareness and situational intelligence of real-time PV production in a distribution system.
Recommended Citation
George, Deborah, "Digital Twins and Artificial Intelligence for Applications in Electric Power Distribution Systems" (2023). All Theses. 4154.
https://open.clemson.edu/all_theses/4154
Author ORCID Identifier
0000-0001-5193-6026