skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: The Use of Statistically Based Rolling Supply Curves for Electricity Market Analysis: A Preliminary Look

Abstract

In light of the changing electricity resource mixes across the United States, an important question in electricity modeling is how additions and retirements of generation, including additions in variable renewable energy (VRE) generation could impact markets by changing hourly wholesale energy prices. Instead of using resource-intensive production cost models (PCMs) or building and using simple generator supply curves, this analysis uses a 'top-down' approach based on regression analysis of hourly historical energy and load data to estimate the impact of supply changes on wholesale electricity prices, provided the changes are not so substantial that they fundamentally alter the market and dispatch-order driven behavior of non-retiring units. The rolling supply curve (RSC) method used in this report estimates the shape of the supply curve that fits historical hourly price and load data for given time intervals, such as two-weeks, and then repeats this on a rolling basis through the year. These supply curves can then be modified on an hourly basis to reflect the impact of generation retirements or additions, including VRE and then reapplied to the same load data to estimate the change in hourly electricity price. The choice of duration over which these RSCs are estimated has a significantmore » impact on goodness of fit. For example, in PJM in 2015, moving from fitting one curve per year to 26 rolling two-week supply curves improves the standard error of the regression from 16 dollars/MWh to 6 dollars/MWh and the R-squared of the estimate from 0.48 to 0.76. We illustrate the potential use and value of the RSC method by estimating wholesale price effects under various generator retirement and addition scenarios, and we discuss potential limits of the technique, some of which are inherent. The ability to do this type of analysis is important to a wide range of market participants and other stakeholders, and it may have a role in complementing use of or providing calibrating insights to PCMs.« less

Authors:
 [1];  [1];  [1];  [2];  [2]
  1. National Renewable Energy Laboratory (NREL), Golden, CO (United States)
  2. U.S. Department of Energy
Publication Date:
Research Org.:
National Renewable Energy Lab. (NREL), Golden, CO (United States)
Sponsoring Org.:
USDOE Office of Energy Efficiency and Renewable Energy (EERE), Strategic Programs (EE-SP)
OSTI Identifier:
1430827
Report Number(s):
NREL/TP-6A20-70954
DOE Contract Number:
AC36-08GO28308
Resource Type:
Technical Report
Country of Publication:
United States
Language:
English
Subject:
29 ENERGY PLANNING, POLICY, AND ECONOMY; electricity markets; renewable energy; retirements; supply curves

Citation Formats

Jenkin, Thomas J, Larson, Andrew, Ruth, Mark F, King, Ben, and Spitsen, Paul. The Use of Statistically Based Rolling Supply Curves for Electricity Market Analysis: A Preliminary Look. United States: N. p., 2018. Web. doi:10.2172/1430827.
Jenkin, Thomas J, Larson, Andrew, Ruth, Mark F, King, Ben, & Spitsen, Paul. The Use of Statistically Based Rolling Supply Curves for Electricity Market Analysis: A Preliminary Look. United States. doi:10.2172/1430827.
Jenkin, Thomas J, Larson, Andrew, Ruth, Mark F, King, Ben, and Spitsen, Paul. Tue . "The Use of Statistically Based Rolling Supply Curves for Electricity Market Analysis: A Preliminary Look". United States. doi:10.2172/1430827. https://www.osti.gov/servlets/purl/1430827.
@article{osti_1430827,
title = {The Use of Statistically Based Rolling Supply Curves for Electricity Market Analysis: A Preliminary Look},
author = {Jenkin, Thomas J and Larson, Andrew and Ruth, Mark F and King, Ben and Spitsen, Paul},
abstractNote = {In light of the changing electricity resource mixes across the United States, an important question in electricity modeling is how additions and retirements of generation, including additions in variable renewable energy (VRE) generation could impact markets by changing hourly wholesale energy prices. Instead of using resource-intensive production cost models (PCMs) or building and using simple generator supply curves, this analysis uses a 'top-down' approach based on regression analysis of hourly historical energy and load data to estimate the impact of supply changes on wholesale electricity prices, provided the changes are not so substantial that they fundamentally alter the market and dispatch-order driven behavior of non-retiring units. The rolling supply curve (RSC) method used in this report estimates the shape of the supply curve that fits historical hourly price and load data for given time intervals, such as two-weeks, and then repeats this on a rolling basis through the year. These supply curves can then be modified on an hourly basis to reflect the impact of generation retirements or additions, including VRE and then reapplied to the same load data to estimate the change in hourly electricity price. The choice of duration over which these RSCs are estimated has a significant impact on goodness of fit. For example, in PJM in 2015, moving from fitting one curve per year to 26 rolling two-week supply curves improves the standard error of the regression from 16 dollars/MWh to 6 dollars/MWh and the R-squared of the estimate from 0.48 to 0.76. We illustrate the potential use and value of the RSC method by estimating wholesale price effects under various generator retirement and addition scenarios, and we discuss potential limits of the technique, some of which are inherent. The ability to do this type of analysis is important to a wide range of market participants and other stakeholders, and it may have a role in complementing use of or providing calibrating insights to PCMs.},
doi = {10.2172/1430827},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Mar 27 00:00:00 EDT 2018},
month = {Tue Mar 27 00:00:00 EDT 2018}
}

Technical Report:

Save / Share: