Research

Publications

Prospective Students

Events

People

Search

Seminars

Links

Home

2012 -13 Seminars


Temperature-based Model-predictive Cascade Mitigation in Electric Power Systems

Mads Almassalkhi
PhD Candidate, Power & Control Systems Laboratory
University of Michigan

Abstract
This talk proposes a novel model-predictive control scheme for transmission-level operation, which combines both economic and security objectives to mitigate the effects of severe disturbances in electrical power systems. A linear convex relaxation of the AC power flow is employed to model transmission line losses and conductor temperatures. Then, a receding-horizon model predictive control (RHMPC) strategy is developed to alleviate line temperature overloads and prevent the propagation of outages. The RHMPC strategy seeks to alleviate temperature overloads by rescheduling generation, energy storage, and other network elements, subject to ramp-rate limits and network limitations. The RHMPC strategy is illustrated with simulations of an augmented IEEE RTS-96 network with energy storage and renewable generation.