Open access
Date
2023Type
- Conference Paper
ETH Bibliography
yes
Altmetrics
Abstract
The accuracy limitation of physics-driven power flow linearization approaches and the widespread deployment of advanced metering infrastructure render data-driven power flow linearization (DPFL) methods a valuable alternative. While DPFL is still an emerging research topic, substantial studies have already been carried out in this area. However, a comprehensive overview and comparison of the available DPFL approaches are missing in the existing literature. This paper intends to close this gap and, therefore, provides a narrative overview of the current DPFL research. Both the challenges (including data-related and power-system-related issues) and methodologies (namely regression-based and tailored approaches) in DPFL studies are surveyed in this paper; numerous future research directions of DPFL analysis are discussed and summarized as well. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000601298Publication status
publishedExternal links
Book title
2023 IEEE Belgrade PowerTechPages / Article No.
Publisher
IEEEEvent
Subject
Power flow linearization; Data-driven; Machine learning; Regression; ProgrammingOrganisational unit
09481 - Hug, Gabriela / Hug, Gabriela
Notes
Conference lecture held on June 29, 2023.More
Show all metadata
ETH Bibliography
yes
Altmetrics