Operation of electric microgrids under uncertainty
Date
2017
Authors
Panwar, Mayank, author
Suryanarayanan, Siddharth, advisor
Chakraborty, Sudipta, committee member
Hovsapian, Rob O., committee member
Young, Peter M., committee member
Zimmerle, Daniel J., committee member
Journal Title
Journal ISSN
Volume Title
Abstract
Optimization and decision-making are non-trivial in case of multiple, incommensurable, and conflicting objectives. Decision-making becomes more complicated with uncertainty in inputs. Power system operation with electric microgrids subsumes all of the abovementioned aspects. Centralized decision-making in day-ahead dispatch of microgrids with multiple objectives in a grid-connected mode is addressed from the perspective of a power distribution system operator. Uncertainties in the electrical output of variable distributed energy resources and load demand due to forecasting errors are treated statistically by using empirical distributions. Scenarios for simulation are generated using statistics of actual data for solar and load demand forecast. Kantorovich distance measure is used for scenario reduction to maintain computational tractability of the problem. Discrete compromise programming is used for multi-criteria decision-analysis to obtain non-dominated dispatch solutions without generating a computationally expensive Pareto front. Two step look-ahead dynamic program routine is used for dispatch optimization of dispatchable, non-dispatchable solar, and energy storage asset. New performance metrics are developed for reserve management in microgrids using North American Electric Reliability Corporation (NERC) metrics and some previously developed metrics by this researcher. The economic dispatch problem is formulated as a constrained optimization problem with the new metric for reserve as a constraint. Optimization programs are implemented using MATLAB® and power system simulations are performed on standard IEEE 13-node test distribution feeder using the real-time simulation platform—RTDS®. Some potential future developments and applications of performance metrics are presented as future work.
Description
Rights Access
Subject
dispatch
microgrids
uncertainty
metrics
decision making
optimization