A systematic literature review on under-frequency load shedding protection using clustering methods

Škrjanc, Tadej (author), Mihalič, Rafael (author), Rudež, Urban (author)

System integrity protection schemes safeguard electric power systems’ overall integrity, among which under-frequency load shedding carries a flagship role. Although triggered rarely, it is irreplaceable in protecting the system from tremendous consequences of a blackout. The search for an optimal strategy has produced numerous innovations over the past 30 years, making it easy to lose track of the state-of-the-art due to the abundance. Given the increasing number of system splits in Europe and the ongoing operational paradigm shift, it is expected that existing load shedding concepts are about to be severely challenged. They are already expected to act more flexibly and, in the future, they may even require a complete redesign to support decarbonization efforts. This is why this research aims to provide a systematic review of existing load shedding algorithms. This is done by categorizing the accessible and adequately documented algorithms using machine learning clustering, more specifically, principal component analysis and t-distributed stochastic neighbour embedding combined with density-based spatial clustering of applications with noise. More than 380 publications were examined and both general and specific features were extracted from each of them. The study provides the description of 54 features along with their pros and cons related to their impact on system frequency stability. These efforts resulted in 28 recognized groups of algorithms, which can be helpful to stakeholders involved in securing and studying electric power system stability. The presented clustering proved very useful and can be extended to any technical field suffering from poor clarity of the state-of-the-art.

• Language: English
• Keywords: literature review, under-frequency load shedding, t-distributed stochastic neighbour embedding, principal component analysis, machine learning, clustering, power system protection, power system stability, blackouts, power system resilience
• Work type: Article
• Typology: 1.02 - Review Article
• Organization: FE - Faculty of Electrical Engineering
• Publication status: Published
• Publication version: Version of Record
• Year: 2023
• Number of pages: 25 str.
• Numbering: Vol. 180, art. 113294
• PID: 20.500.12556/RUL-148679
• UDC: 621.31
• ISSN on article: 1364-0321
• DOI: 10.1016/j.rser.2023.113294
• COBISS.SI-ID: 150505475
• Publication date in RUL: 29.08.2023

Record is a part of a journal

• Title: Renewable & sustainable energy reviews
• Shortened title: Renew. sustain. energy rev.
• Publisher: Elsevier
• ISSN: 1364-0321
• COBISS.SI-ID: 44255233

Licences

• License: CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
• Link: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Description: The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.

Secondary language

• Language: Slovenian
• Keywords: pregled literature, podfrekvenčno razbremenjevanje, t-porazdeljena stohastična vložitev sosedov, analiza glavnih komponent, strojno učenje, gručenje, zaščita elektroenergetskega sistema, stabilnost elektroenergetskega sistema, izpad električne energije, odpornost elektroenergetskega sistema