A Planning Model of Standalone Hydrogen-Based Carbon-Free Microgrid Through Convex Relaxation

Hydrogen-based carbon-free microgrids (CFMs) show great potential in alleviating the pressure of climate changes. To reduce the computational burden, representative days are usually selected to simulate the yearly operation in the planning models of CFMs. This method, however, fails to capture the l...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on smart grid 2023-07, Vol.14 (4), p.2668-2680
Main Authors: Wu, Xiong, Cao, Binrui, Liu, Bingwen, Wang, Xiuli
Format: Article
Language:English
Subjects:
Carbon
Computing time
Distributed generation
Hydrogen
Hydrogen storage
Integer programming
Linear programming
Mixed integer
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:Hydrogen-based carbon-free microgrids (CFMs) show great potential in alleviating the pressure of climate changes. To reduce the computational burden, representative days are usually selected to simulate the yearly operation in the planning models of CFMs. This method, however, fails to capture the long-term dynamics of renewable power generation and behaviors of hydrogen storage due to the loss of data. To effectively reflect these characteristics, a convex relaxation-based planning model using yearly data is proposed. The model is developed for the planning of a standalone hydrogen-based CFM. The model reduces the computational time compared to the mixed integer linear programming (MILP) one. Specifically, the convex relaxed operation models of devices are introduced through convex relaxation. The proposed planning model of the CFM is obtained by replacing the original model of each device with the corresponding relaxed model. In addition, a method of constructing feasible solutions is developed to make the results practical. The proposed model is adequately investigated in the case studies based on the yearly data. The results indicate that the proposed model is solved at least 2 times faster than the conventional MILP model since some MILP models fail to be solve in a suitable time. The proposed method obtains more economical planning results than that using the representative day-based method. Using yearly data to simulate long-term operation also leads to lower costs. The planning results are also trustable since all the operation states are binary and feasible.
ISSN:1949-3053
1949-3061
DOI:10.1109/TSG.2022.3224900