MIP formulation improvement for large scale security constrained unit commitment with configuration based combined cycle modeling

Abstract

As a part of the day-ahead market clearing process, Midcontinent Independent System Operator (MISO) solves one of the largest and most challenging Security Constrained Unit Commitment (SCUC) models. Better computational performance of SCUC models not only improves the market efficiency but also facilitates future market developments. This paper introduces recent developments in SCUC formulation with configuration based combined cycle modeling in MISO, which include the improvement on constraints associated with binary variables, the improvement of formulation on piecewise linear incremental energy curve (PWL) and reduction of non-zeros by aggregating variables (AGG). Furthermore, mathematical proof shows that the proposed enhanced locally ideal PWL model is the convex envelope of the PWL cost function. To illustrate the effectiveness of the proposed formulation, numerical results and analysis based on MISO system are presented.