Two-stage dynamic management in energy communities using a decision system based on elastic net regularization

Antonello Rosato, Massimo Panella, Amedeo Andreotti, Osama A. Mohammed and Rodolfo Araneo

Applied Energy, 2021, vol. 291, issue C, No S0306261921003445

Abstract: The modern revolutionary changes in power delivery systems with the advent of smart and flexible grids require systems interoperability, seamless integration of technologies and functionalities. This scenario is paving the way to new prospective of energy clusters that call for new methodologies for the dynamic energy management of distributed energy resources. This paper proposes a new management scheme of energy clusters (e.g., smart building, energy community or virtual power plant) that is based on a data driven decision-maker that daily deploys the activities of the day ahead, providing an optimized scheduling which will be the base for the operations for the next day. The new aggregator relies on some innovative features. It leverages an optimization process based on the elastic net regularization that proves to be an effective support for finding the best scheduling of the distributed resources according to specific key performance indicators, that presently is the unbalance. The decision process works on predicated data obtained through a recently assessed short term forecasting based on long short-term memory neural networks properly adapted to distributed environments. The method allows to extract insight from data and run what-if scenarios to define the best scheduling of resources. The technique uses a set of feasible rules to optimally aggregate the distributed resources and demand-side management programs. We tested the proposed aggregator on a real energy cluster making use of real measured data over two years period. The results show the effectiveness of the proposed approach that is able to predict any sort of unit and manage any sort of program. The significance of this work is to approach the energy management as an optimization and decision problem in a more robust way of reasoning, by employing forecasting/optimization over all the quantities in the community, resulting in a thoroughly complex and efficient method.

Keywords: Distributed generation; Energy storage; Electric vehicles; Forecasting; Neural networks; Long short-term memory networks; Demand response (search for similar items in EconPapers)
Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (7)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0306261921003445
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:appene:v:291:y:2021:i:c:s0306261921003445

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/journaldescription.cws_home/405891/bibliographic
http://www.elsevier. ... 405891/bibliographic

DOI: 10.1016/j.apenergy.2021.116852

Access Statistics for this article

Applied Energy is currently edited by J. Yan

More articles in Applied Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().