Impacts of UC formulation tightening on computation of convex hull prices

In U.S. electricity markets, energy prices are traditionally determined by optimal dual variables corresponding to system demand constraints of the economic dispatch problem. With fixed unit commitment (UC) decisions, however, such prices cannot reflect actual marginal costs of serving load, and market participants might not want to follow commitment and dispatch decisions made by Independent System Operators. To address this, convex hull pricing provides a promising way to accurately calculate prices and reduce uplift payments in electricity markets. One way to find the convex hull prices is to take the dual of the UC problem and solve the resulting problem by Lagrangian relaxation. In this study, impacts of tightening unit formulations on prices from integer relaxation are investigated, where formulation tightening is combined with the surrogate Lagrangian relaxation method to calculate convex hull prices. The performance of the proposed model is compared with the conventional UC formulation and the effect of integer-relaxation is also studied. IEEE 118 bus system with 54 units is used for testing. Results demonstrate that the tightened formulation can produce accurate prices upon integer-relaxation for a smaller 3-hour example and approximate prices for a larger 24-hour example. This demonstrates the potential of formulation tightening on the calculation of convex hull prices.