Improving Power System Modeling: A Tool to Link Capacity Expansion and Production Cost Models

Victor Diakov, Wesley Cole, Patrick Sullivan, Gregory Brinkman, Robert Margolis

Research output: NRELTechnical Report

Abstract

Capacity expansion models (CEM) provide a high-level long-term view at the prospects of the evolving power system. In simulating the possibilities of long-term capacity expansion, it is important to maintain the viability of power system operation in the short-term (daily, hourly and sub-hourly) scales. Production-cost models (PCM) simulate routine power system operation on these shorter time scales using detailed load, transmission and generation fleet data by minimizing production costs and following reliability requirements. When based on CEM 'predictions' about generating unit retirements and buildup, PCM provide more detailed simulation for the short-term system operation and, consequently, may confirm the validity of capacity expansion predictions. Further, production cost model simulations of a system that is based on capacity expansion model solution are 'evolutionary' sound: the generator mix is the result of logical sequence of unit retirement and buildup resulting from policy and incentives. The above has motivated us to bridge CEM with PCM by building a capacity expansion - to - production cost model Linking Tool (CEPCoLT). The Linking Tool is built to onset capacity expansion model prescriptions onto production cost model inputs. NREL's ReEDS and Energy Examplar's PLEXOS are the capacity expansion and the production cost models, respectively. Via the Linking Tool, PLEXOS provides details of operation for the regionally-defined ReEDS scenarios.
Original languageAmerican English
Number of pages29
DOIs
StatePublished - 2015

NREL Publication Number

  • NREL/TP-6A20-64905

Keywords

  • power system modeling
  • simulation tools

Fingerprint

Dive into the research topics of 'Improving Power System Modeling: A Tool to Link Capacity Expansion and Production Cost Models'. Together they form a unique fingerprint.

Cite this