A partitioning around medoids (PAM) based sequential clustering approach for model order estimation of low-frequency oscillations in wide area measurement system

被引:1
作者
Sahoo, Manoranjan [1 ]
Rai, Shekha [1 ]
机构
[1] Natl Inst Technol Rourkela, Dept Elect Engn, Rourkela, Odisha, India
来源
SADHANA-ACADEMY PROCEEDINGS IN ENGINEERING SCIENCES | 2024年 / 49卷 / 01期
关键词
Low-frequency oscillation; model order; PAM; TLS-ESPRIT; SIGNAL STABILITY ANALYSIS; POWER; IDENTIFICATION; SUBSPACE;
D O I
10.1007/s12046-023-02408-5
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
The identification of poorly damped low-frequency oscillations (LFOs) is very crucial for maintaining small signal stability in power systems. The Total least square estimation of signal parameters via rotational invariance techniques (TLS-ESPRIT) is a model parameter estimation method, which works efficiently even in noisy conditions. However, the TLS-ESPRIT requires prior information about the number of modes present in the signal. Different techniques are proposed in the literature for model order (MO) estimation of signal, but they consider presumed tolerance value for the selection of probable separation boundary between two subspaces. As tolerance value fluctuates with noise level, these techniques become inefficient and harder to automate for online estimation. So, the main contribution of the proposed work is to design an efficacious MO estimation technique, which will efficiently eliminate the requirements of tolerance value to overcome the aforementioned limitations. So, in this proposed work, a sequential partitioning technique known as Partitioning Around Medoids (PAM) is adopted for the precise estimation of LFOs order of Phasor measurement unit (PMU) data. The MO estimation is done by separating the prominent eigen values of autocorrelation matrix (ACM) iteratively into two components i.e signal subspace and noise subspace. Since the PAM algorithm operates in two layers of clustering and robust towards high variance noise, hence the insignificant eigen values are hindered from getting clustered in the signal subspace. The robustness of the proposed technique is validated by conducting comparative study with Exact Model Order (EMO), modified sequential cumulative sum (modified-SCUSUM) and sequential eigen value contribution (SEVC) methods for synthetic signals, two area test system, practical probing data of Western Electricity Coordinating Council (WECC) system and oscillatory power data of WSCC 9 bus system simulated on Real Time Digital Simulator (RTDS).
引用
收藏
页数:17
相关论文
共 26 条
[1]   NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION [J].
AKAIKE, H .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :716-723
[2]   A tutorial on data-driven eigenvalue identification: Prony analysis, matrix pencil, and eigensystem realization algorithm [J].
Almunif, Anas ;
Fan, Lingling ;
Miao, Zhixin .
INTERNATIONAL TRANSACTIONS ON ELECTRICAL ENERGY SYSTEMS, 2020, 30 (04)
[3]  
Anderson P.M., 2008, Power system control and stability
[4]  
[Anonymous], Report and data of WECC
[5]   EIGENVALUE ANALYSIS OF SYNCHRONIZING POWER FLOW OSCILLATIONS IN LARGE ELECTRIC-POWER SYSTEMS [J].
BYERLY, RT ;
BENNON, RJ ;
SHERMAN, DE .
IEEE TRANSACTIONS ON POWER APPARATUS AND SYSTEMS, 1982, 101 (01) :235-243
[6]   Implementation of Spectral Clustering with Partitioning Around Medoids (PAM) Algorithm on Microarray Data of Carcinoma [J].
Cahyaningrum, Rosalia D. ;
Bustamam, Alhadi ;
Siswantining, Titin .
SYMPOSIUM ON BIOMATHEMATICS (SYMOMATH 2016), 2017, 1825
[7]   Sinusoidal Order Estimation Using Angles between Subspaces [J].
Christensen, Mads Graesboll ;
Jakobsson, Andreas ;
Jensen, Soren Holdt .
EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING, 2009,
[8]   The matrix pencil for power system modal extraction [J].
Crow, ML ;
Singh, A .
IEEE TRANSACTIONS ON POWER SYSTEMS, 2005, 20 (01) :501-502
[9]  
Djuric P. M., 1993, ICASSP-93. 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing (Cat. No.92CH3252-4), P53, DOI 10.1109/ICASSP.1993.319592
[10]  
Hai Y, 2012, PROCEEDINGS OF 2012 IEEE 14TH INTERNATIONAL CONFERENCE ON COMMUNICATION TECHNOLOGY, P1281, DOI 10.1109/ICCT.2012.6511395