Leveraging an Optimization Model for a Virtual Power Plant as an Alternative to Under-Frequency Load Shedding Due to Severe Winter Weather

This praxis explores the suitability of deploying distributed energy resources (DERs) and residentially-sited load controllers in a virtual power plant (VPP) configuration as a resiliency tool for grid operations during severe cold weather events. To determine the efficacy of such an approach, the research entailed gathering historical weather and electrical utility data within the Pennsylvania-Jersey-Maryland (PJM) Interconnection and training regression models for ten (10) of the reporting territories comprising the interconnection. From there, the trained models were used to develop load predictions based on simulated weather and electric utility predictor inputs. The load predictions were then compared against hypothetical day-ahead schedule reserve (DASR) values subject to constraints imposed by modeled failure rates of key electrical generating assets. The capacity shortfalls, when indicated, were then input into an Optimization Dashboard whose objective function was constructed to return the lowest-cost deployment of DERs subject to applied constraints and dependencies. This approach affords the opportunity to determine the costs of enhanced intervention and support during weather-induced electric grid resiliency events when weighed against demonstrated economic loss factors observed following recent named winter storms.

Author

• Alexander, William D

Language

• en

Date created

• 2026

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Engineering Management

Degree

• D.Engr.

Advisor

• Bierce, Jonathan

Committee Member(s)

• Sarkani, Shahram
• Mazzuchi, Thomas

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/3n203z814