Solving the DynamicsAware Economic Dispatch Problem with the Koopman Operator
Abstract
The dynamicsaware economic dispatch (DED) problem embeds lowlevel generator dynamics and operational constraints to enable near realtime scheduling of generation units in a power network. DED produces a more dynamic supervisory control policy than traditional economic dispatch (TED) that reduces overall generation costs. However, in contrast to TED, DED is a nonlinear, nonconvex optimization problem that is computationally prohibitive to solve. We introduce a machine learningbased operatortheoretic approach for solving the DED problem efficiently. Specifically, we develop a novel discretetime Koopman Operator (KO) formulation that embeds domain information into the structure of the KO to learn highfidelity approximations of the generator dynamics. Using the KO approximation, the DED problem can be reformulated as a computationally tractable linear program (abbreviated DEDKO). We demonstrate the high solution quality and computationaltime savings of the DEDKO model over the original DED formulation on a 9bus test system.
 Authors:

 BATTELLE (PACIFIC NW LAB)
 Publication Date:
 Research Org.:
 Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
 Sponsoring Org.:
 USDOE
 OSTI Identifier:
 1819913
 Report Number(s):
 PNNLSA159920
 DOE Contract Number:
 AC0576RL01830
 Resource Type:
 Conference
 Resource Relation:
 Conference: Proceedings of the Twelfth ACM International Conference on Future Energy Systems (eEnergy '21), June 28July 2, 2021, Virtual, Online
 Country of Publication:
 United States
 Language:
 English
Citation Formats
King, Ethan, Bakker, Craig KR, Bhattacharya, Arnab, Chatterjee, Samrat, Pan, Feng, Oster, Matthew R., and Perkins, Casey J. Solving the DynamicsAware Economic Dispatch Problem with the Koopman Operator. United States: N. p., 2021.
Web.
King, Ethan, Bakker, Craig KR, Bhattacharya, Arnab, Chatterjee, Samrat, Pan, Feng, Oster, Matthew R., & Perkins, Casey J. Solving the DynamicsAware Economic Dispatch Problem with the Koopman Operator. United States.
King, Ethan, Bakker, Craig KR, Bhattacharya, Arnab, Chatterjee, Samrat, Pan, Feng, Oster, Matthew R., and Perkins, Casey J. 2021.
"Solving the DynamicsAware Economic Dispatch Problem with the Koopman Operator". United States.
@article{osti_1819913,
title = {Solving the DynamicsAware Economic Dispatch Problem with the Koopman Operator},
author = {King, Ethan and Bakker, Craig KR and Bhattacharya, Arnab and Chatterjee, Samrat and Pan, Feng and Oster, Matthew R. and Perkins, Casey J.},
abstractNote = {The dynamicsaware economic dispatch (DED) problem embeds lowlevel generator dynamics and operational constraints to enable near realtime scheduling of generation units in a power network. DED produces a more dynamic supervisory control policy than traditional economic dispatch (TED) that reduces overall generation costs. However, in contrast to TED, DED is a nonlinear, nonconvex optimization problem that is computationally prohibitive to solve. We introduce a machine learningbased operatortheoretic approach for solving the DED problem efficiently. Specifically, we develop a novel discretetime Koopman Operator (KO) formulation that embeds domain information into the structure of the KO to learn highfidelity approximations of the generator dynamics. Using the KO approximation, the DED problem can be reformulated as a computationally tractable linear program (abbreviated DEDKO). We demonstrate the high solution quality and computationaltime savings of the DEDKO model over the original DED formulation on a 9bus test system.},
doi = {},
url = {https://www.osti.gov/biblio/1819913},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2021},
month = {7}
}