Near-Optimal Scheduling in Day-Ahead Markets: Pricing Models and Payment Redistribution Bounds

Near-optimal unit commitment (UC) scheduling is a practical reality in wholesale electricity markets. This paper revisits previous work that has found that minor differences in cost among near-optimal schedules can result in a large redistribution of market payments (i.e., changes in generator profi...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on power systems 2020-05, Vol.35 (3), p.1684-1694
Main Authors: Eldridge, Brent, O'Neill, Richard, Hobbs, Benjamin F.
Format: Article
Language:English
Subjects:
Electricity supply industry
Generators
Incentives
Integers
market design
mixed integer programming
nonconvex pricing
Pricing
Production
Schedules
Scheduling
Software
Unit commitment
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Near-optimal unit commitment (UC) scheduling is a practical reality in wholesale electricity markets. This paper revisits previous work that has found that minor differences in cost among near-optimal schedules can result in a large redistribution of market payments (i.e., changes in generator profits and consumer surplus). It has been believed that this instability is unavoidable, but previous studies have only calculated prices using what we call the Restricted pricing model. This paper compares previous results to three additional models that are based on integer relaxation, and which we call the Partial, Tight, and Loose Dispatchable pricing models. Results are presented for a suite of test cases, including four ISO-scale cases. Similar to previous findings, the Restricted and Partial Dispatchable models both result in large payment redistributions among alternative solutions. In contrast, theoretical and experimental results for the Tight and Loose Dispatchable models show that pricing models with unconditional integer relaxation will have bounded payment redistributions, and, further, this bound can become quite small by tightening the UC problem's convex relaxation. In the presence of market power, stable financial outcomes may improve market efficiency by reducing incentives to bid strategically.
ISSN:0885-8950
1558-0679
DOI:10.1109/TPWRS.2019.2947400