TY - GEN
T1 - Estimation of Transmission Line Sequence Impedances using Real PMU Data
AU - Mansani, Prashanth Kumar
AU - Pal, Anamitra
AU - Rhodes, Matthew
AU - Keel, Brian
PY - 2019/1/2
Y1 - 2019/1/2
N2 - Accurate knowledge of transmission line parameters in general, and sequence impedances, in particular, plays an important role in state estimation, fault detection, and adjustment of relay settings. Line parameter estimation using online methods has attracted considerable interest with the widespread installation of phasor measurement units (PMUs). Although various methods have been proposed in the literature for line parameter estimation, most of them have been tested on purely synthetic datasets. A synthetic dataset does not capture the nuances of real data, such as measurement invariance and realistic field noise. Therefore, the algorithms developed using synthetic datasets may not be as effective when used in practice. In this paper, a three-stage test procedure is developed to compare the performance of two algorithms, namely, moving-window total least squares (MWTLS) recursive Kalman filter (RKF), on real PMU data. The results prove that RKF is better than MWTLS. This paper also proposes using ASPEN data as an initial estimate to RKF for further improving its performance. Finally, to circumvent the problems faced due to data dropouts, an auto regressive integrated moving average (ARIMA) model is implemented to predict the variations in sequence impedances.
AB - Accurate knowledge of transmission line parameters in general, and sequence impedances, in particular, plays an important role in state estimation, fault detection, and adjustment of relay settings. Line parameter estimation using online methods has attracted considerable interest with the widespread installation of phasor measurement units (PMUs). Although various methods have been proposed in the literature for line parameter estimation, most of them have been tested on purely synthetic datasets. A synthetic dataset does not capture the nuances of real data, such as measurement invariance and realistic field noise. Therefore, the algorithms developed using synthetic datasets may not be as effective when used in practice. In this paper, a three-stage test procedure is developed to compare the performance of two algorithms, namely, moving-window total least squares (MWTLS) recursive Kalman filter (RKF), on real PMU data. The results prove that RKF is better than MWTLS. This paper also proposes using ASPEN data as an initial estimate to RKF for further improving its performance. Finally, to circumvent the problems faced due to data dropouts, an auto regressive integrated moving average (ARIMA) model is implemented to predict the variations in sequence impedances.
KW - ARIMA
KW - ASPEN
KW - Parameter estimation
KW - Phasor measurement unit (PMU)
KW - Recursive Kalman filter (RKF)
KW - Total least squares (TLS)
UR - http://www.scopus.com/inward/record.url?scp=85061818338&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85061818338&partnerID=8YFLogxK
U2 - 10.1109/NAPS.2018.8600605
DO - 10.1109/NAPS.2018.8600605
M3 - Conference contribution
AN - SCOPUS:85061818338
T3 - 2018 North American Power Symposium, NAPS 2018
BT - 2018 North American Power Symposium, NAPS 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2018 North American Power Symposium, NAPS 2018
Y2 - 9 September 2018 through 11 September 2018
ER -