An Extended Integral Unit Commitment Formulation and an Iterative Algorithm for Convex Hull Pricing

TL;DR

This paper develops new integral formulations and iterative algorithms to efficiently compute exact convex hull prices in large-scale power markets, enhancing market transparency and reducing uplift payments.

Contribution

It introduces novel integral formulations for generators and an efficient iterative algorithm for convex hull pricing in complex power systems.

Findings

Exact convex hull prices obtained within 20 minutes on MISO instances.

The approach works with and without transmission constraints.

Improved computational efficiency over existing methods.

Abstract

To increase market transparency, independent system operators (ISOs) have been working on minimizing uplift payments based on convex hull pricing theorems. However, the large-scale complex systems for ISOs bring computational challenges to the existing convex hull pricing algorithms. In this paper, based on the analysis of specific generator features in the Midcontinent ISO (MISO) system, besides reviewing integral formulations for several special cases, we develop two integral formulations of a single generator that can capture these features. We then build a compact convex hull pricing formulation based on these integral formulations. Meanwhile, to improve the computational efficiency, we propose innovative iterative algorithms with convergence properties, plus a complementary algorithm, to obtain a convex hull price. The computational results indicate that our approach leads to an exact convex hull price on MISO instances with and without transmission constraints and the solutions can be obtained within 20 minutes.