Comparison of variant robust SCUC models for operational security and economics of power systems under uncertainty

In electricity market, security-constrained unit commitment (SCUC) utilizes the detailed market information submitted by market participants to provide a financially viable and physically feasible unit commitment solution for Independent system operators (ISOs)/regional transmission owners (RTOs). With the dramatic increase of renewable energy resources in recent years, the robust optimization approach has been recently explored for solving the SCUC problem with various uncertainties. This paper focuses on the following questions: How variant robust SCUC models in terms of different worst-case definitions could impact operational security and economics of power systems under uncertainties? In addition, what is the proper robust SCUC model that can meet specific market operation needs of ISOs/RTOs for effectively operating the system? Four different robust SCUC models are investigated, including: (1) Minimax load shedding robust model, (2) Minimax economic dispatch (ED) cost robust model, (3) Minimax variance robust model, and (4) Minimax regret robust model. Outer approximation (OA) and Benders' decomposition (BD) are used to solve the four robust models. Numerical case studies compare the performance of the four models and analyze their distinct characteristics for the market economic operation (such as locational marginal prices (LMP), regret costs, and base case costs) and system secure operation (such as load shedding (LS) quantities and Do-Not-Exceed (DNE) limits of power systems under uncertainties.