Evaluation of Power System Identification Methods and Their Applicability for Online PSS Tuning

Diego Cifelli; Catalin Gavriluta; Adolfo Anta Martinez

doi:10.1109/PowerTech55446.2023.10202959

Evaluation of Power System Identification Methods and Their Applicability for Online PSS Tuning

Research output: Chapter in Book or Conference Proceedings › Conference Proceedings with Oral Presentation › peer-review

Abstract

This paper compares four methods and their accuracy in the identification of a simplified power system model.
The identified model is to be used for power system stabilizer (PSS) tuning. Therefore, the focus is on identifying a low-order linear model that is able to accurately capture the dynamics of the system in the frequency bandwidth in which the PSS has typically to act in order to damp low frequency oscillations.
We evaluate polynomial transfer function models and state space model structures under different estimation methods are considered. The two-area four-generator power system model is used as test case, and the results of the four methods are compared to the analytical results in terms of both low-frequency oscillation modes and frequency response in the 0.1 - 2Hz interval.
The combination of the Predictor Based Subspace IDentification (PBSID) method and the Prediction-Error Minimization (PEM) method showed the better performance.

Original language	English
Title of host publication	2023 IEEE Belgrade PowerTech
Place of Publication	Belgrade
Number of pages	7
DOIs	https://doi.org/10.1109/PowerTech55446.2023.10202959
Publication status	Published - 2023
Event	2023 IEEE Belgrade PowerTech - Belgrade, Serbia Duration: 25 Jun 2023 → 29 Jun 2023

Conference

Conference	2023 IEEE Belgrade PowerTech
Country/Territory	Serbia
City	Belgrade
Period	25/06/23 → 29/06/23

Research Field

Power System Digitalisation

Access to Document

10.1109/PowerTech55446.2023.10202959

Cite this

@inproceedings{7d795b293d0c45528602626e6639e111,

title = "Evaluation of Power System Identification Methods and Their Applicability for Online PSS Tuning",

abstract = "This paper compares four methods and their accuracy in the identification of a simplified power system model. The identified model is to be used for power system stabilizer (PSS) tuning. Therefore, the focus is on identifying a low-order linear model that is able to accurately capture the dynamics of the system in the frequency bandwidth in which the PSS has typically to act in order to damp low frequency oscillations. We evaluate polynomial transfer function models and state space model structures under different estimation methods are considered. The two-area four-generator power system model is used as test case, and the results of the four methods are compared to the analytical results in terms of both low-frequency oscillation modes and frequency response in the 0.1 - 2Hz interval. The combination of the Predictor Based Subspace IDentification (PBSID) method and the Prediction-Error Minimization (PEM) method showed the better performance.",

author = "Diego Cifelli and Catalin Gavriluta and \{Anta Martinez\}, Adolfo",

year = "2023",

doi = "10.1109/PowerTech55446.2023.10202959",

language = "English",

isbn = "9781665487788",

booktitle = "2023 IEEE Belgrade PowerTech",

note = " 2023 IEEE Belgrade PowerTech ; Conference date: 25-06-2023 Through 29-06-2023",

}

TY - GEN

T1 - Evaluation of Power System Identification Methods and Their Applicability for Online PSS Tuning

AU - Gavriluta, Catalin

AU - Anta Martinez, Adolfo

A2 - Cifelli, Diego

PY - 2023

Y1 - 2023

N2 - This paper compares four methods and their accuracy in the identification of a simplified power system model. The identified model is to be used for power system stabilizer (PSS) tuning. Therefore, the focus is on identifying a low-order linear model that is able to accurately capture the dynamics of the system in the frequency bandwidth in which the PSS has typically to act in order to damp low frequency oscillations. We evaluate polynomial transfer function models and state space model structures under different estimation methods are considered. The two-area four-generator power system model is used as test case, and the results of the four methods are compared to the analytical results in terms of both low-frequency oscillation modes and frequency response in the 0.1 - 2Hz interval. The combination of the Predictor Based Subspace IDentification (PBSID) method and the Prediction-Error Minimization (PEM) method showed the better performance.

AB - This paper compares four methods and their accuracy in the identification of a simplified power system model. The identified model is to be used for power system stabilizer (PSS) tuning. Therefore, the focus is on identifying a low-order linear model that is able to accurately capture the dynamics of the system in the frequency bandwidth in which the PSS has typically to act in order to damp low frequency oscillations. We evaluate polynomial transfer function models and state space model structures under different estimation methods are considered. The two-area four-generator power system model is used as test case, and the results of the four methods are compared to the analytical results in terms of both low-frequency oscillation modes and frequency response in the 0.1 - 2Hz interval. The combination of the Predictor Based Subspace IDentification (PBSID) method and the Prediction-Error Minimization (PEM) method showed the better performance.

U2 - 10.1109/PowerTech55446.2023.10202959

DO - 10.1109/PowerTech55446.2023.10202959

M3 - Conference Proceedings with Oral Presentation

SN - 9781665487788

BT - 2023 IEEE Belgrade PowerTech

CY - Belgrade

T2 - 2023 IEEE Belgrade PowerTech

Y2 - 25 June 2023 through 29 June 2023

ER -

Evaluation of Power System Identification Methods and Their Applicability for Online PSS Tuning

Abstract

Conference

Research Field

Access to Document

Fingerprint

Cite this