Price‐based unit commitment with decision‐dependent uncertainty in hourly demand

Abstract

Abstract The price‐based unit commitment (PBUC) problem aims to optimise the power generating units' schedules to meet the system demand with the objective to maximise the generation companies' (GENCOs') profit. State‐of‐the‐art PBUC models have taken into account exogenous uncertainties in renewable generation, demand, and price signals. This study proposes a novel PBUC problem formulation with endogenous or decision‐dependent uncertainty (DDU) in the elastic portion of the demand. The proposed PBUC model is formulated as a mixed‐integer non‐linear programming (MINLP) problem with non‐convex continuous relaxation. A concavification approach is developed to reformulate the non‐convex MINLP model as an equivalent mixed‐integer quadratic programming (MIQP) model whose continuous relaxation is convex. Case studies considering GENCOs owning and operating 3, 12, 19, and 40 generating units demonstrate the efficacy of the proposed DDU‐aware PBUC formulation on the GENCOs' anticipated profits.

Keywords

• power generation economics
• quadratic programming
• power markets
• power generation scheduling
• integer programming
• pricing