Adjustable robust optimization approach for two-stage operation of energy hub-based microgrids

Growing demand for energy carriers has led to an increased interest in developing and managing multiple energy carrier microgrids. Furthermore, the volatile nature of renewable resources as well as the uncertain electrical and thermal demands imposes significant challenges for the operation of microgrids. Motivated by this, the paper leverages a min max min robust framework for short-term operation of microgrids with natural gas network to capture the uncertainty of wind generation and electrical/thermal loads. The proposed model is linearized and solved using the column-and-constraint generation (C&CG) procedure that decomposes the framework into a master problem and a subproblem. The master problem minimizes the unit commitment cost, while the sub-problem determines the dispatch cost associated with the worst realization of uncertainties via a max min objective function. Also, polyhedral uncertainty sets are defined with budget of uncertainty parameter that adjusts the trade-off between the operation cost and the degree of robustness. The effectiveness of the framework is assessed and discussed via a 21-node energy hub-based microgrid. It can be seen that the solution immunizes against all realizations of uncertainties, whereby increasing the budget of uncertainty and the forecast error, the system robustness is improved. Moreover, the dual variables of the subproblem are converted to the primary variables in order to evaluate the unit commitment and energy dispatch results. (c) 2021 Elsevier Ltd. All rights reserved.

Automated summary

This paper explores the short-term operation of microgrids with natural gas network using a min max min robust framework to address the challenges posed by the uncertainty of renewable resources and electrical/thermal loads. The effectiveness of the proposed model is assessed through a 21-node energy hub-based microgrid, demonstrating improved system robustness with increased budget of uncertainty and forecast error. Converting dual variables of the subproblem to primary variables allows for evaluation of unit commitment and energy dispatch results.