Abstract
Transmission line parameters are required by various power system applications. Thus, it is important to develop accurate transmission line parameter estimation methods. This paper puts forward Kalman filter based methods to track the dynamically changing line parameters. The proposed methods can overcome the estimation inaccuracy problem caused by the effect of the noise in the measurements. The proposed methods are based upon the equivalent- circuit of long line and utilize voltage and current data recorded by the Phasor Measurement Units at both ends of the line. Non-compensated and differently configured series compensated lines have been considered in this paper. The case studies demonstrate that the proposed methods have the capability of dynamically tracking changing line parameters and yield accurate estimates with the presence of measurement noises.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Transmission line parameters are critical information for various power system applications including fault location and classification [1,2,3], relay protection [4,5,6], and other applications [7, 8]. Thus, the accuracy of line parameters is important, and accurate line parameter estimation methods are demanded. Basically, the line parameter estimation methods can be divided into two categories: online methods and offline methods. Offline estimation methods are based on factors such as conductor parameters and tower configurations [9, 10]. However, those parameters may not be the same as the assumed condition due to uncertain weather and loading conditions. Thus, the line parameter estimation results of offline methods may deviate from the actual values. Online estimation methods became more favored recently, which are based on the equivalent circuit of transmission lines and utilize the measurement data from phasor measurement units. Online methods are more likely to give the trustable estimation results since the measurements inherently reflect the prevailing operating and weather conditions.
Different online methods have been proposed in [11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27] to estimate line parameters considering the line length, specific line configurations, data availability, etc. The methods for estimating zero-sequence line parameters are proposed in [11,12,13]. The author of [14] proposed a method for identifying and estimating erroneous transmission line parameters. A non-linear weighted least-square error algorithm for three-terminal lines was proposed in [15]. Paper [16] proposed a method to estimate line parameters using a robust estimator to minimize the influence of outliers. Paper [17] presented a method to estimate the parameters of double circuit transmission lines. In [18], a Kalman filter based recursive regression method was proposed to estimate three-phase line parameters of non-transposed and non-symmetric medium length lines. A moving-window total least squares algorithm was proposed in [19] to estimate the parameters of a long transmission line. The method treats parameters as constant during a time window. In [20], the author proposed a method based on the Clarke’s transformation matrix using faulted voltage and current measurements obtained at the line terminals. The authors of [21] proposed a method for estimating line parameters using data during a grounding fault. In [22], an estimation method was proposed considering dynamic operating states.
Series compensation has been used in long transmission lines for improved operation. The estimation methods for series compensated lines have been proposed in [23, 24]. The author of [23] proposed methods for long lines where the series compensators are installed in the middle of the line. In [24], a method using traveling waves generated by disturbance was proposed to estimate the parameters for series compensated lines.
Most estimation methods are proposed under the assumption that line parameters are static. However, line parameters are continuously changing over time with varying loading and weather conditions. Thus, it is desirable to have an effective dynamic line parameter tracking method. The authors of [25] proposed a method that essentially counted on the regular Kalman filter for dynamic tracking, which might not be highly efficient. In [26], the author discussed dynamic tracking but required additional information (e.g., weather forecast) to build the prediction model. In [27], the author used the sliding window method to track the moving tendency of line parameters. However, this method requires complicated tuning of several parameters such as the window size.
This paper proposes Kalman filtering based methods that employ voltage and current measurements from line terminals and require no extra information to track parameters of long transmission lines dynamically. This paper proposes improvements over commonly used Kalman filtering to better track the parameter variations. In addition, the proposed methods are applicable to transmission lines with different series compensation configurations.
The rest of the paper is organized as follows. Section 2 presents the proposed line parameter estimation methods. Section 3 presents the evaluation studies and results, followed by the conclusion.
2 Proposed methods for line parameter estimation
This section presents the proposed line parameter estimation methods for non-compensated lines and series compensated lines. According to [28], the series compensator is typically installed at the middle or the end of a transmission line. The schematic diagram of the non-compensated and different series compensated transmission lines are shown in Fig. 1, where \(S\) is the sending end, and \(T\) is the receiving end. PMUs are installed at both ends for measuring the voltages and currents. \(V_{S}\) and \(I_{S}\) denote the sending end voltage and current and \(V_{T}\) and \(I_{T}\) represent the receiving end voltage and current.
The overall line parameter estimation method based on the Kalman filter technique is presented in Sect. 2.1, followed by specific methods for different line configurations.
2.1 Proposed methods based on Kalman filter
The Kalman filter method is defined by two steps: prediction and update. Define the state to be estimated as \(x\). In the first step, the priori estimate for the state at time \(k\) is predicted based on the state at time \(k - 1\) by using (1), and the priori error covariance at time \(k\) is calculated based on (2):
where the subscript \(k\) is the instant, the superscript − means priori, \(\hat{x}\) means the estimate for \(x\), \(P\) is the error covariance, and \(Q\) is the process noise covariance. Then the Kalman gain \(K\), the posteriori estimate \(\hat{x}\) and the posteriori error covariance \(P\) at instant \(k\) are calculated by (3)–(5),
where \(R\) is the observation noise covariance, \(I\) is the identity matrix, \(H\) is the observation matrix, and \(w\) is the observation vector.
To dynamically track the state without knowing any information between the states at instant \(k\) and \(k - 1\), it is critical to make the most of measurements. Regular Kalman filtering technique yields slow tracking performance as shown in case studies to be presented in Sect. 3. To improve tracking performance of the Kalman filter, the following two adjustment methods are proposed.
The first solution is to use the adjusted Kalman gain method, which increases the Kalman gain \(K\). While the priori estimate \(\hat{x}_{k}^{ - }\). gets pulled away from the actual value by the previous estimate \(\hat{x}_{k - 1}\), it is reasonable to add more weight to the current measurements. The concept of adjusting the Kalman gain has already been considered such as [32]. In our work, a factor \(D\) (\(D > 1\)) is introduced as follows:
The second solution is called the fast Kalman method, which is novel to the best knowledge of the authors. This new method considers the direction in which the line parameters are moving toward, i.e., taking \((\hat{x}_{k}^{ - } - \hat{x}_{k - n} )\) into account, where \(n \ge 1\). Thus,
where the variable \(U\) is used to adjust the weight of \((\hat{x}_{k}^{ - } - \hat{x}_{k - n} )\).
By rearranging the terms, \(\hat{x}_{k}\) can be expressed as
For the first \(n\) estimations (when \(k - n < 0\)), \(\hat{x}_{k - n}\) is set to zero. Note that while increasing the value of \(D\) and \(U\) can increase the ability of dynamic tracking, it also brings instability to the algorithms. In the case studies, the value of \(D\) and \(U\) are chosen as to balance the performance of dynamic tracking and stability. The choice of optimal process noise covariance \(Q\) and measurement noise covariance \(R\) are referred to [29,30,31].
In the following subsections, Kalman filter formulation is presented for different line configurations with and without series compensators, based on which the states and then the line parameters can be derived.
2.2 Non-compensated Line
The equivalent-π circuit of the positive-sequence network of the non-compensated transmission line is shown in Fig. 2. The following sections for different lines are all based on this model. \(Z\) is the series impedance and \(\frac{Y}{2}\) is the shunt admittance of this circuit at one end. The relationship between \(Z\) and \(Y\) and transmission line parameters is defined by the following equations,
where \(l\) is the length of the line, \(Z_{c}\) is the characteristic impedance of the line, \(\gamma\) is the propagation constant. \(z\) and \(y\) are the positive sequence series impedance and shunt admittance per unit length.
The voltages and currents in Fig. 2 can be related by the admittance matrix
where \(a = \frac{Y}{2} + \frac{1}{Z}\) and \(b = - \frac{1}{Z}\). Rewrite (13),
To use the Kalman filtering techniques presented in Sect. 2.1, define the state to be estimated as \(x = \left[ {\begin{array}{*{20}c} a \\ b \\ \end{array} } \right]\). The observation vector \(w = \left[ {\begin{array}{*{20}c} {I_{S} } \\ {I_{T} } \\ \end{array} } \right]\) and the observation matrix \(H = \left[ {\begin{array}{*{20}c} {V_{S} } & {V_{T} } \\ {V_{T} } & {V_{S} } \\ \end{array} } \right]\).
The state \(x\) can be estimated following the method shown in Sect. 2.1. After \(x\) is estimated, \(Z\) and \(Y\) can be calculated as
Multiplying \(Z\) and \(Y\)
Let \(d = e^{\gamma l}\), and (17) can be written as
Solving for \(d\), \(\gamma\) can be calculated as
At last, the line parameters are determined as
2.3 One-end compensated line
The configuration where the series compensator is installed at one end of a transmission line is discussed in this section. Assume that the impedance of the series compensator \(Z_{cap}\) is unknown. The equivalent-π circuit of this line configuration is shown in Fig. 3.
The voltage and current measured at each bus can be related by the admittance matrix
where \(a = \frac{Y}{2} + \frac{1}{Z}\) and \(b = - \frac{1}{Z}\). The unknown variables can be separated from measurements,
where \(g = bZ_{cap}\) and \(h = aZ_{cap}\). Since there are two additional unknown variables \(g\) and \(h\), one more set of measurements is needed to solve for the four variables. Thus, with two sets of measurements, (23) becomes
To use the Kalman filtering techniques presented in Sect. 2.1, define the state to be estimated as \(x = \left[ {\begin{array}{*{20}c} a & b & g & h \\ \end{array} } \right]^{T}\). The observation vector is \(w = \left[ {\begin{array}{*{20}c} {\begin{array}{*{20}c} {I_{{S_{1} }} } \\ {I_{{T_{1} }} } \\ \end{array} } \\ {I_{{S_{2} }} } \\ {I_{{T_{2} }} } \\ \end{array} } \right]\) and the observation matrix is \(H = \left[ {\begin{array}{*{20}c} {V_{{S_{1} }} } & {V_{{T_{1} }} } & { - I_{{S_{1} }} } & 0 \\ {V_{{T_{1} }} } & {V_{{S_{1} }} } & 0 & { - I_{{T_{1} }} } \\ {V_{{S_{2} }} } & {V_{{T_{2} }} } & { - I_{{T_{2} }} } & 0 \\ {V_{{T_{2} }} } & {V_{{S_{2} }} } & 0 & { - I_{{T_{2} }} } \\ \end{array} } \right]\). Once x is known, the line parameters can be solved using (15)–(21).
2.4 Mid-compensated line with known series compensator
The series compensator can also be installed at the middle of transmission line. The equivalent circuit is shown in Fig. 4. \(V_{R}\) and \(V_{G}\) are the voltage at each side of the series compensator. Assume that the admittance of the series compensator \(Y_{cap}\) is unknown.
Similarly, the relationship between the voltages and currents at each node can be built using the admittance matrix \(M\).
where \(M = \left[ {\begin{array}{*{20}c} a & 0 & b & 0 \\ 0 & a & 0 & b \\ b & 0 & {a + Y_{cap} } & { - Y_{cap} } \\ 0 & b & { - Y_{cap} } & {a + Y_{cap} } \\ \end{array} } \right]\), \(a = \frac{Y}{2} + \frac{1}{Z}\) and \(b = - \frac{1}{Z}\). The matrix can be reduced to (26) using node elimination technique
where \(A = a - \frac{{b^{2} }}{a} + \frac{{Y_{cap} b^{2} }}{{a^{2} + 2aY_{cap} }}\) and \(B = - \frac{{b^{2} Y_{cap} }}{{a^{2} + 2aY_{cap} }}\). Equation (26) can be rewritten as
To use the Kalman filtering techniques presented in Sect. 2.1, define the state to be estimated as \(x = \left[ {\begin{array}{*{20}c} A \\ B \\ \end{array} } \right]\). The observation vector \(w = \left[ {\begin{array}{*{20}c} {I_{S} } \\ {I_{T} } \\ \end{array} } \right]\) and the observation matrix is \(H = \left[ {\begin{array}{*{20}c} {V_{S} } & {V_{T} } \\ {V_{T} } & {V_{S} } \\ \end{array} } \right]\).
Once \(A\) and \(B\) are obtained, \(a\) and \(b\) can be calculated as
Then the line parameter can be derived using the method described in Sect. 2.2.
2.5 Mid-compensated line with known current through series compensator
For the mid-compensated line, another case is that the impedance of the series compensator is unknown, but the current through it is available. Figure 5 shows the equivalent circuit of the situation. \(I_{R}\) is the current flowing through the series compensator.
Define the unknown variables \(a = \frac{Y}{2} + \frac{1}{Z}\) and \(b = - \frac{1}{Z}\). Equations (30)–(33) can be written using Kirchhoff’s Current Law (KCL) for each node
Based on (30) and (31), \(V_{R}\) can be eliminated,
Based on (32) and (33), \(V_{G}\) can be eliminated,
Multiply \(V_{T}\) to (34) and \(V_{S}\) to (35) and then subtract the resultant equations. Then \(a\) can be expressed in terms of \(b\),
where \(C = \frac{{I_{R} \left( {V_{S} + V_{T} } \right)}}{{I_{S} V_{T} - I_{T} V_{S} }}\). The expression of \(b\) can be obtained by substituting (36) for \(a\) in (34),
Then based on (36) and (37), we can write
In this case, to use the Kalman filtering techniques presented in Sect. 2.1, define the state \(x = \left[ {\begin{array}{*{20}c} a \\ b \\ \end{array} } \right]\), the measurement vector \(w = \left[ {\begin{array}{*{20}c} {C\left( {CI_{S} - I_{R} } \right)} \\ {CI_{S} - I_{R} } \\ \end{array} } \right]\), and the measurement matrix \(H = \left[ {\begin{array}{*{20}c} {\left( {C^{2} - 1} \right)V_{S} } & 0 \\ 0 & {\left( {C^{2} - 1} \right)V_{S} } \\ \end{array} } \right]\). Then \(x\) can be used to solve for the line parameters.
2.6 Two-ends compensated line
The two-ends compensated line configuration is the case where two equivalent series compensators are installed at both ends of the line. The equivalent circuit is shown in Fig. 6. \(Y_{cap}\) represents the admittance of the series compensators.
The voltages and currents injected at each node can be related using the admittance matrix \(M\).
where \(M = \left[ {\begin{array}{*{20}c} {Y_{cap} } & 0 & { - Y_{cap} } & 0 \\ 0 & {Y_{cap} } & 0 & { - Y_{cap} } \\ { - Y_{cap} } & 0 & {a + Y_{cap} } & b \\ 0 & { - Y_{cap} } & b & {a + Y_{cap} } \\ \end{array} } \right]\), \(a = \frac{Y}{2} + \frac{1}{Z}\) and \(b = - \frac{1}{Z}\).
Equation (39) can be reduced to (40)
where
and
Equation (40) can be rewritten as
To use the Kalman filtering techniques presented in Sect. 2.1, define the state to be estimated as \(x = \left[ {\begin{array}{*{20}c} A \\ B \\ \end{array} } \right]\). The observation vector \(w = \left[ {\begin{array}{*{20}c} {I_{P} } \\ {I_{Q} } \\ \end{array} } \right]\) and the observation matrix is \(H = \left[ {\begin{array}{*{20}c} {V_{S} } & {V_{T} } \\ {V_{T} } & {V_{S} } \\ \end{array} } \right]\). Then the line parameters can be solved based on (41) and (42) using non-linear equation solving techniques.
3 Results and discussion
This section presents results and discussion based on evaluation studies utilizing simulated data. Different line configurations were built using Matlab Simscape Electrical, and the proposed algorithms were implemented in Matlab. The actual parameters of the transmission line are shown in Table 1. Simulation studies were run to generate voltage and current phasor measurements during varying operating conditions. The measurements were added with normally distributed random noises. Each random noise consists of two parts: the magnitude noise and the angle noise. Both noises have their mean set to zero. For the magnitude noise, its mean \(\mu_{mag} = 0\) and its standard deviation \(\sigma_{mag}\) varies for each case; for the angle noise, its mean \(\mu_{angle} = 0\) and its standard deviation \(\sigma_{angle} = 0.2^\circ\).
Since the series resistance is more susceptible to the change of loading and weather conditions when compared to the series reactance and shunt susceptance, we only changed the series resistance dynamically while the series reactance and shunt susceptance were kept static. In all the simulations, the series resistance was first set to its normal value and then changed to 120%, 150%, 100%, and 80% of its normal value during different time periods.
For each case study, the estimation results of the normal Kalman filter method, adjusted Kalman filter method, and fast Kalman filter method (the parameter \(n\) is set to 2 for all cases) were compared.
3.1 Case study for non-compensated line
The variables \(D\), \(U_{1}\) and \(U_{2}\) are set to 20, 0.4 and 0.3, which correspond to the adjusted Kalman gain method, fast KF1 and fast KF2 method, respectively. \(\sigma_{mag}\) is set to 0.5%.
The estimation results for the parameters of the non-compensated line are shown in Table 2 and Fig. 7. The formula of the Mean Absolute Percentage Error (MAPE) is given by (44), where \(n\) is the number of estimations, \(\hat{x}\) is the estimated value, and \(x\) is the actual value. It is shown that the adjusted Kalman gain method has the best performance among all the methods when tracking the dynamically changing series resistance.
3.2 Case study for one-end compensated line
The variables \(D\),\(U_{1}\) and \(U_{2}\) are set to 10, 0.4 and 0.3, which correspond to the adjusted Kalman gain method, fast KF1 and fast KF2 method, respectively. \(\sigma_{mag}\) is set to 0.1%.
The estimation results for the positive sequence series resistance, series reactance, shunt susceptance and the reactance of the series compensator of the one-end compensated line are shown in Table 3 and Fig. 8. From the results, we can see that the proposed methods are able to track the dynamically changing line parameters accurately and quickly.
3.3 Case study for mid-compensated line with known series compensator
The variables \(D\),\(U_{1}\) and \(U_{2}\) are set to 10, 0.4 and 0.3, which correspond to the adjusted Kalman gain method, fast KF1 and fast KF2 method, respectively. \(\sigma_{mag}\) is set to 0.5%.
The estimation results for the positive sequence series resistance, series reactance, and shunt susceptance of the mid-compensated line with known impedance of series compensator are shown in Table 4 and Fig. 9. The results show that all the proposed methods have the capability to accurately and quickly track the changing line parameters, and the adjusted Kalman gain method performs best.
3.4 Case study for mid-compensated line with known current through series compensator
The variables \(D\),\(U_{1}\) and \(U_{2}\) are set to 10, 0.3 and 0.2, which correspond to the adjusted Kalman gain method, fast KF1 and fast KF2 method, respectively. \(\sigma_{mag}\) is set to 0.1%.
The estimation results for the positive sequence series resistance, series reactance, and shunt susceptance of the one-end compensated line are shown in Table 5 and Fig. 10.
The proposed methods can successfully track the dynamic line parameters. It is seen that the adjusted Kalman gain method performs best, while Fast KF1 and KF2 have difficulty in estimating the shunt susceptance parameter.
3.5 Case study for two-ends compensated line
The variables \(D\),\(U_{1}\), and \(U_{2}\) are set to 10, 0.3 and 0.2, which correspond to the adjusted Kalman gain method, fast KF1 and fast KF2 method, respectively. \(\sigma_{mag}\) is set to 0.5%.
The estimation results for the positive sequence series resistance, series reactance, shunt susceptance and susceptance of the series compensator of the one-end compensated line are shown in Table 6 and Fig. 11. It is manifested that all three line parameters can be tracked successfully by the proposed methods and the adjusted Kalman method performs best.
4 Conclusion
Transmission line parameters are essential inputs to various power system applications. This paper puts forward several improved variants of Kalman filter based methods to dynamically and accurately track the positive transmission line parameters. The proposed methods utilize the voltage and current measurements obtained at both terminals of the line. Non-compensated and different types of series compensated lines have been discussed. The case study results have demonstrate that the proposed methods can accurately and quickly track the dynamically changing line parameters under the effect of measurement noises.
In general, the adjusted Kalman gain method outperforms other methods in the speed of tracking dynamically changing parameters and can be regarded as the first choice method, while other approaches including the Normal KF, Fast KF1 and Fast KF2 can be used for corroboration.
Availability of data and material
Not applicable.
Code availability
Not applicable.
References
Roostaee S, Thomas MS, Mehfuz S (2017) Experimental studies on impedance based fault location for long transmission lines. Prot Control Mod Power Syst 2(1):16. https://doi.org/10.1186/s41601-017-0048-y
Fan W, Liao Y (2019) Wide area measurements based fault detection and location method for transmission lines. Prot Control Mod Power Syst 4(1):7. https://doi.org/10.1186/s41601-019-0121-9
Rajaraman PNAS, Mallikarjuna B, Jaya Bharata Reddy M, Mohanta DK (2018) Robust fault analysis in transmission lines using Synchrophasor measurements. Prot Control Mod Power Syst 3(1):14. https://doi.org/10.1186/s41601-018-0082-4
Wang Q, Bo Z, Ma X, Zhang M, Zhao Y, Zhu Y, Wang L (2016) Integrated protection based on multi-frequency domain information for UHV half-wavelength AC transmission line. Prot Control Mod Power Syst 1(1):17. https://doi.org/10.1186/s41601-016-0027-8
Das S, Navalpakkam Ananthan S, Santoso S (2018) Relay performance verification using fault event records. Prot Control Mod Power Syst 3(1):22. https://doi.org/10.1186/s41601-018-0094-0
Mallikarjuna B, Shanmukesh P, Anmol D, Jaya Bharata Reddy M, Mohanta DK (2018) PMU based adaptive zone settings of distance relays for protection of multi-terminal transmission lines. Prot Control Mod Power Syst 3(1):12. https://doi.org/10.1186/s41601-018-0087-z
Ha H, Balakrishnan G, Zhang Z (2019) Dynamic line rating current calculation without ambient parameters inputs. In: 2019 72nd annual conference for protective relay engineers, college station, TX. https://doi.org/10.1109/CPRE.2019.8765875
Lin H, Zhuang J, Hu Y-C, Zhou H (2020) DefRec: establishing physical function virtualization to disrupt reconnaissance of power grids’ cyber-physical infrastructures. In: Annual network and distributed system security symposium, San Diego, California. https://doi.org/10.14722/ndss.2020.24365
Chan SM (1993) Computing overhead line parameters. IEEE Comput Appl Power 6(1):43–45
Dommel HW (1985) Overhead line parameters from handbook formulas and computer programs. IEEE Trans Power Appar Syst PAS 104(2):366–372
Momen A, Johnson BK, Chakhchoukh Y (2019) Parameters estimation for short line using the least trimmed squares (LTS). In: IEEE power and energy society innovative smart grid technologies conference, Washington, D.C. https://doi.org/10.1109/ISGT.2019.8791579
Das S, Navalpakkam Ananthan S, Santoso S (2018a). Estimating zero-sequence impedance of three-terminal transmission line and Thevenin impedance using relay measurement data. Prot Control Mod Power Syst 3(1):36. https://doi.org/10.1186/s41601-018-0108-y
Dasgupta K, Soman SA (2017) Estimation of zero sequence parameters of mutually coupled transmission lines from synchrophasor measurements. IET Gener Trans Distrib 11(14):3539–3547. https://doi.org/10.1049/iet-gtd.2017.0057
Asprou M, Kyriakides E (2017) Identification and estimation of erroneous transmission line parameters using PMU measurements. IEEE Trans Power Deliv 32(6):2510–2519. https://doi.org/10.1109/tpwrd.2017.2648881
Mousavi-Seyedi SS, Aminifar F, Afsharnia S (2015) Parameter estimation of multiterminal transmission lines using joint PMU and SCADA data. IEEE Trans Power Deliv 30(3):1077–1085. https://doi.org/10.1109/tpwrd.2014.2369500
Milojevic V, Calija S, Rietveld G, Acanski MV, Colangelo D (2018) Utilization of PMU measurements for three-phase line parameter estimation in power systems. IEEE Trans Instrum Meas 67(10):2453–2462. https://doi.org/10.1109/tim.2018.2843098
Unde SV, Dambhare SS (2016) Double circuit transmission line parameter estimation using PMU. In: 2016 IEEE 6th international conference on power systems, New Delhi, India. https://doi.org/10.1109/ICPES.2016.7584163
Mishra C, Centeno VA, Pal A (2015) Kalman-filter based recursive regression for three-phase line parameter estimation using synchrophasor measurements. In: 2015 IEEE power and energy society general meeting, Denver, CO. https://doi.org/10.1109/PESGM.2015.7285678
Ding L, Bi TS, Zhang DN (2011) Transmission line parameters identification based on moving-window TLS and PMU data. In: International conference on advanced power system automation and protection, pp 2187–2191. https://doi.org/10.1109/ISGT.2019.8791579
Pineda JTR, Kurokawa S (2018). Estimation of parameters for faulted transposed transmission lines. In: 2018 IEEE electrical power and energy conference (EPEC), Toronto, Canada. https://doi.org/10.1109/EPEC.2018.8598364
Hao J, Zhan G, Jianfeng Z (2020) Line parameter estimation of distribution network after grounding fault. In: 2020 IEEE power and energy conference at Illinois (PECI), Champaign, IL. https://doi.org/10.1109/PECI48348.2020.9064644
Jiang W, Tang H (2020) Distribution line parameter estimation considering dynamic operating states with a probabilistic graphical model. Int J Electr Power Energy Syst 121:106–133. https://doi.org/10.1016/j.ijepes.2020.106133
Zhang Y, Liao Y (2019) Optimal line parameter estimation method for mid-compensated transmission lines. In: 2019 North American power symposium (NAPS), Wichita, KS. https://doi.org/10.1109/NAPS46351.2019.9000274
Gajare S, Pradhan AK, Terzija V (2017) A method for accurate parameter estimation of series compensated transmission lines using synchronized data. IEEE Trans Power Syst 32(6):4843–4850. https://doi.org/10.1109/pesgm.2018.8586009
Ren P, Abur A, Lev-Ari H (2019) Tracking transmission line parameters in power grids observed by PMUs. 2019 IEEE Milan PowerTech, Milan, Italy. https://doi.org/10.1109/PTC.2019.8810838
Ren P, Lev-Ari H, Abur A (2018) Tracking three-phase untransposed transmission line parameters using synchronized measurements. IEEE Trans Power Syst 33(4):4155–4163. https://doi.org/10.1109/tpwrs.2017.2780225
Bian X, Li XR, Chen H, Gan D, Qiu J (2011) Joint estimation of state and parameter with synchrophasors-Part I: state tracking. IEEE Trans Power Syst 26(3):1196–1208. https://doi.org/10.1109/tpwrs.2010.2098422
Miller J, Brunet-watson M, Leighfield J (2014) Review of series compensation for transmission lines. https://www.spp.org/documents/22902/17_ju4715%20-%20spp%20-%20review%20of%20series%20compensation%20-%20final%20-%202.1.pdf. Accessed 4 Mar 2020
Huang Y, Zhang Y, Xu B, Wu Z, Chambers JA (2018) A new adaptive extended Kalman filter for cooperative localization. IEEE Trans Aerosp Electron Syst 54(1):353–368. https://doi.org/10.1109/taes.2017.2756763
Huang Y, Zhang Y, Wu Z, Li N, Chambers J (2018) A novel adaptive Kalman filter with inaccurate process and measurement noise covariance matrices. IEEE Trans Autom Control 63(2):594–601. https://doi.org/10.1109/tac.2017.2730480
Kaba A, Kıyak E (2020) Optimizing a Kalman filter with an evolutionary algorithm for nonlinear quadrotor attitude dynamics. J Comput Sci 39:101051. https://doi.org/10.1016/j.jocs.2019.101051
Huang Y, Zhang Y, Zhao Y, Shi P, Chambers J (2020) A novel outlier-robust Kalman filtering framework based on statistical similarity measure. IEEE Trans Autom Control. https://doi.org/10.1109/tac.2020.3011443
Funding
Not applicable.
Author information
Authors and Affiliations
Contributions
YZ implemented the proposed algorithms and prepared the paper draft. YL contributed to the algorithm formulation and advised on implementation and paper writing.
Corresponding author
Ethics declarations
Conflicts of interest
The authors declare that they have no conflicts of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Zhang, Y., Liao, Y. Kalman filter based method for tracking dynamic transmission line parameters. SN Appl. Sci. 3, 117 (2021). https://doi.org/10.1007/s42452-020-04036-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s42452-020-04036-8