A Mix-integer Programming Based Deep Reinforcement Learning Framework for Optimal Dispatch of Energy Storage System in Distribution Networks
Journal of Modern Power Systems and Clean Energy · 2024 · 13 (2) · 597-608
DOI: 10.35833/mpce.2024.000391 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Modern Power Systems and Clean Energy, 2024. Verfasst von Shengren Hou und 4 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref).
Abstract (Originalsprache)
—— The optimal dispatch of energy storage systems (ESSs) in distribution networks poses significant challenges, pri‐ marily due to uncertainties of dynamic pricing, fluctuating de‐ mand, and the variability inherent in renewable energy sources. By exploiting the generalization capabilities of deep neural net‐ works (DNNs), the deep reinforcement learning (DRL) algo‐ rithms can learn good-quality control models that adapt to the stochastic nature of distribution networks. Nevertheless, the practical deployment of DRL algorithms is often hampered by their limited capacity for satisfying operational constraints in re‐ al time, which is a crucial requirement for ensuring the reliabili‐ ty and feasibility of control actions during online operations. This paper introduces an innovative framework, named mixed-integer programming based deep reinforcement learning (MIP-DRL), to overcome these limitations. The proposed MIP-DRL framework can rigorously enforce operational constraints for the optimal dispatch of ESSs during the online execution. This framework involves training a Q-function with DNNs, which is subsequently represented in a mixed-integer programming (MIP) formulation. This unique combination allows for the seamless integration of operational constraints into the decision-making process. The effectiveness of the proposed MIP-DRL framework is validated through numerical simulations, demon‐ strating its superior capability to enforce all operational con‐ straints and achieve high-quality dispatch decisions and show‐ ing its advantage over existing DRL algorithms.
Zitieren
APA
Hou, S., Salazar, E. M., Palensky, P., Chen, Q., & Vergara, P. P. (2024). A Mix-integer Programming Based Deep Reinforcement Learning Framework for Optimal Dispatch of Energy Storage System in Distribution Networks. Journal of Modern Power Systems and Clean Energy, 13(2), 597-608. https://doi.org/10.35833/mpce.2024.000391
Vancouver
Hou S, Salazar EM, Palensky P, Chen Q, Vergara PP. A Mix-integer Programming Based Deep Reinforcement Learning Framework for Optimal Dispatch of Energy Storage System in Distribution Networks. Journal of Modern Power Systems and Clean Energy. 2024;13(2):597-608. doi:10.35833/mpce.2024.000391
BibTeX
@article{Hou2024,
author = {Hou, Shengren and Salazar, Edgar Mauricio and Palensky, Peter and Chen, Qixin and Vergara, Pedro P.},
title = {A Mix-integer Programming Based Deep Reinforcement Learning Framework for Optimal Dispatch of Energy Storage System in Distribution Networks},
journal = {Journal of Modern Power Systems and Clean Energy},
year = {2024},
volume = {13},
number = {2},
pages = {597--608},
doi = {10.35833/mpce.2024.000391}
}
EUVIMED