Clarkson University Department of Electrical and Computer Engineering
A Unified Framework for Risk-Aware Commitment and Real-Time Dispatch of Energy Storage in Power System Operations
M.S. Thesis Defense David Gumina
Abstract: This research addresses the critical challenge of reliably integrating large-scale variable renewable energy resources into the power grid, emphasizing the essential role of advanced scheduling methodologies for energy storage resources (ESRs) in achieving grid reliability under high renewable penetration. As renewable generation, particularly wind power, increases, system operators face significant operational complexities, especially during extended periods of low renewable energy production. Traditional resource adequacy frameworks, reliant primarily on deterministic day-ahead and real-time markets, are insufficient for managing the growing uncertainties and variability inherent in renewable energy sources. To bridge this gap, the study develops a multi-day operational resource adequacy framework incorporating probabilistic risk assessments. This framework quantifies loss-of-load risks resulting from forecast uncertainties in renewable generation, load demand, generator availability, and transmission constraints. By leveraging probabilistic metrics such as Loss of Load Probability (LOLP), the developed methodology strategically informs ESR scheduling, particularly focusing on longduration storage technologies. The effectiveness of this approach is demonstrated through detailed simulations of the New York State power grid, developed in collaboration with the New York Independent System Operator (NYISO). Scenarios projected for 2030, including historically low-wind events and significant renewable uncertainty, validate the framework’s capability to enhance operational reliability and resilience. Complementing this multi-day framework, the research introduces an innovative real-time dispatch method integrating day-ahead and real-time market operations via a novel energy “hold-up” algorithm. This algorithm ensures ESRs maintain optimal state-of-charge (SoC) reserves to support peak load demands, dynamically adjusting ESR dispatch in response to real-time operational events and updated forecasts. A mixed-integer linear programming model incorporating NYISO’s operational data demonstrates that this approach significantly improves resource availability during peak periods, reduces battery degradation due to cycling, and enhances the economic viability of ESRs, supporting broader clean energy objectives. Also, the research presents a reliabilityinformed enhancement to NYISO’s existing Capacity Adequacy Commitment Tool (CACT), termed Risk-CACT. This extended tool explicitly integrates probabilistic operational risk metrics derived from iterative Monte Carlo simulations into real-time corrective scheduling decisions. Risk-CACT effectively mitigates operational risks arising from renewable variability, forecast uncertainties, and contingencies by dynamically adjusting generator and ESR schedules. Comprehensive case studies for NYISO’s 2030 scenarios illustrate substantial reductions in reliability risks, notably decreasing peak LOLP values by nearly two orders of magnitude. All together these methodologies form a robust operational toolkit that equips system operators with strategic, adaptive, and economically efficient solutions, significantly enhancing grid reliability and resilience during the transition to a renewable-rich energy system. Future research directions include exploring increased ESR penetration, assessing diversified storage technology portfolios, multi-day operational coordination, and evaluating the impacts of transmission constraints on system effectiveness. This work provides practical and impactful strategies aligned with long-term decarbonization and grid reliability goals. Keywords: Energy Storage Scheduling, Operational Resource Adequacy, Real-Time Unit Commitment.
Wednesday, 05/28/2025 at 11:00 am CAMP 160
Advisors: Prof. Yazhou (Leo) Jiang Other Committee Members: Prof. Tom Ortmeyer Dr. Shubo Zhang