Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

An overview of power swing detection methods in distance relays and the factors involved

An overview of power swing detection methods in distance relays and the factors involved INTRODUCTIONIn current power systems, power transmission lines are considered a desirable conductor for transporting electrical energy. These lines transport electrical energy to remote locations with low power losses to reach the consumers through distribution networks [1]. Power transmission lines are one of the most essential components of a power system. In order to provide continuous and stable electrical energy to different consumers, it is necessary to quickly detect faults in power transmission lines using a suitable protection system. A proper protection system with timely and precise operation can reduce damages to the network equipment and even humans [2]. Among the various protection relays, distance relays are an integral part of protecting power systems. These relays have desirable features such as simplicity, use of local voltage and current for detecting a fault and its location [3]. The features of the distance relay make it suitable for protecting power transmission lines [4, 5]. However, a significant challenge that may affect the proper operation of distance relays is the occurrence of the power swing (PS) in power transmission lines [6, 7].PS is a transient phenomenon that occurs for a variety of reasons, including faults, line switching, and connecting or disconnecting of large blocks of loads in power systems [8]. PS in power systems can cause maloperation of distance relays due to entering the measuring impedance of the distance relay into the protection zones of this relay [9]. Undesirable transmission line tripping due to maloperation of distance relay against PS may destroy the power system stability and result in cascading outages of power transmission lines, resulting in global blackouts in the power system [10]. To prevent the maloperation of the distance relay during a PS, a power swing blocking (PSB) function is incorporated in distance relays [11, 12].PS detection can be challenging when the number of distributed generations like wind turbines is increased in the network. In fact, wind turbines can significantly change the PS characteristics and challenge power swing detection methods [13]. In addition, the presence of a thyristor‐controlled series capacitor (TCSC) in power systems can pose a fundamental problem to PS detection methods due to harmonic injection into voltage and current signals [14].Different researches have been presented to provide a method for detecting PS. Therefore, the main purpose of this paper is a comprehensive review of methods for detecting PS and the factors involved. Accordingly, the reviewed items include the types of PS, its detection methods, and different factors involved, such as affecting compensated lines and the presence of wind turbines on PS detection, determining threshold values ​in PS detection methods, and analyzing methods for proper testing of PSB function. Finally, suggestions for further research in this area are presented. It should be noted that there are some studies in this area that have already reviewed PS detection methods. Authors in [15] have investigated methods for detecting PS. However, this paper does not provide a comprehensive overview of the various methods and challenges of PS. In [16], methods for detecting PS have also been investigated. However, the effects of compensated lines and wind farms as well as PSB function testing method have not been investigated in this study.The present paper is structured as follows. In Section 2, different types of PS and the reasons for their occurrence are analysed. In Section 3, different types of PS detection methods are analysed. In Section 4, the proposed algorithms for determining the threshold values in the PS detection methods are analysed. In Section 5, the effects of compensated lines on PS parameters and PS detection methods are investigated. In Section 6, the effects of wind farms on PS are investigated. In Section 7, different methods for correct testing of the PSB function are analysed. Finally, in Section 8 additional suggestions for future researches are discussed. Conclusion is given in Section 9.DIFFERENT TYPES OF PSIn general, PS can be divided into three categories include stable, unstable, and multi‐mode power swing [11, 17]. These categories are explained in this section.Power swing equationIn order to express the power swing equations, consider Figure 1 [18, 19]. In this figure Eg${E_g}$represents a synchronous generator. This synchronous generator is connected to an infinite bus (Eb${E_b}$) via Xg${X_g}$and Xb${X_b}$ impedances. The electric power (Pe${P_e}$) which is transformed between the synchronous generator and the infinite bus can be calculated using Equation (1) [18, 19].1Pe=Eg.EbXg+Xesinδ,$$\begin{equation}{P_e} = \frac{{{E_g}.{E_b}}}{{{X_g} + {X_e}}}\sin \delta ,\end{equation}$$whereδ is the rotor angle.1FIGURESample transmission systemThe maximum transformed power (Pmax${P_{\max }}$) between the synchronous generator and the infinite bus can also be calculated from Equation (2) [19].2Pmax=Eg.EbXg+Xe.$$\begin{equation}{P_{\max }} = \frac{{{E_g}.{E_b}}}{{{X_g} + {X_e}}}.\end{equation}$$Figure 2 shows the power angle curve of a synchronous generator [19]. It is assumed that the synchronous generator transforms Pe${P_e}$ into the infinite bus. By creating a disturbance in the power system (such as connecting a load), the required power of the network suddenly increases and consequently, mechanical power (Pm${P_m}$) of the generator increases. This sudden increase in the mechanical power causes the operating point of the synchronous generator to be transferred toPm1${P_{m1}}$. Therefore, due to the inequality between electrical power and mechanical power, generator control systems operate and reduce mechanical power. In this reduction, the mechanical power is adjusted at the Pm2${P_{m2}}$point instead ofPe${P_e}$. At this stage, the generator control systems increasePm${P_m}$. This cycle continues until the mechanical power of the network balances with electrical power. Therefore, the rotor angle oscillation equation will be in accordance with Equation (3) [18–20].3Md2δdt2+Ddδdt=Pm−Pmaxsinδ,$$\begin{equation}M\frac{{{d^2}\delta }}{{d{t^2}}} + D\frac{{d\delta }}{{dt}} = {P_m} - {P_{\max }}\sin \delta ,\end{equation}$$where M is the inertia constant and D is the damping coefficient.2FIGUREP−δ$P - \delta \;$curve of the synchronous generatorStable PSStable PS typically occurs due to network disturbance, such as disconnecting a parallel line or reclosing lines in the power system [21]. Stable PS is the simplest and least intense type of PS that can occur in power systems. Figure 3 shows the changes in δduring this kind of PS [21]. Accordingly, the rotor angle has never been greater thanPmax${P_{\max }}$ (90 degrees) [18, 19]. This figure clearly shows that the rotor angle has stabilized after a short time, and therefore, this type of PS is called stable PS [19].3FIGUREChanges in the rotor angle of a synchronous generator during a stable PSUnstable PSUnstable PS is usually caused by a major power system disturbance. Compared to the stable type, the rotor angle in this type of PS will exceed Pmax${P_{\max }}$ (90 degrees), which makes the synchronous generator unstable [18, 19]. This instability of the synchronous generator is the reason why this PS is called unstable PS. As shown in Figure 4, the generator does not stabilize over time [19]. To eliminate this type of PS, the generator protection relays must detect the instability and disconnect the generator from the grid. In contrast, distance relays must maintain healthy transmission lines until instability in the network is eliminated and prevent further disturbances [19].4FIGUREChanges in the rotor angle of a synchronous generator during unstable PSMulti‐mode PSAs modern power systems become more sophisticated and more synchronous generators are implemented in networks with less distance, the probability of multi‐mode PS occurrence is significantly increased [11]. Multi‐mode PS occurs when two or more synchronous generators are involved in an oscillation [11]. In this case, the frequency of the PS is not constant and varies in each cycle. In addition, the current signal loses its sinusoidal waveform [22]. The occurrence of multi‐mode PS can pose serious challenges to existing methods of detecting PS [21].Figure 5 shows the changes in the rotor angle of a synchronous generator during a multi‐mode PS [21]. As can be seen from this figure, when the multi‐mode PS occurs, similar to the stable PS, the rotor angle of the synchronous generator does not exceed 90 degrees and the generator is not practically unstable [21].5FIGUREChanges in the rotor angle of a synchronous generator during a multi‐mode PSDETECTION OF POWER SWINGDetection of PS in power systems is very crucial. Failure to detect PS or simultaneous faults that occur during PS can cause instability in the network or even global blackout [23]. According to [16], a lack of accurate and timely diagnosis of PS caused various blackouts in Malaysia, Italy, Brazil, and India between 2003 and 2012. For this reason, there is a growing interest in research on the phenomenon of PS and its detection. It is worth mentioning that only the detection of PS and preventing the maloperation of the distance relay is not enough. PS detection algorithms must be able to detect a simultaneous fault (especially symmetric three‐phase fault) with PS very quickly [24]. In the following, the most important methods for detecting PS and its simultaneous faults are analysed.Impedance change methodImpedance change methods are one of the oldest methods for detecting PS. The basis of these methods is based on the impedance changes (ΔZ$\Delta Z$) calculated by the distance relay. Some of these methods also use an external zone with an impedance of 1Ω in relays with a 5 A current transformer (CT) or a 5 Ω impedance in relays with a 1 A CT [25]. In impedance‐based methods, an appropriate threshold value should be chosen in order to have a proper operation. This threshold value is determined according to the duration of the entrance and exit of impedance from the external zone. If the measured impedance remained in the external zone less than the threshold time, the relay would consider it as a fault. If this value exceeds the threshold time, the relay detects the PS and blocks the relay operation [26–28]. References [12, 29, 30] use the rate of impedance changes presented in reference [31] to detect PS. In this method, the three conditions of monotony, continuity, and smoothness are continuously checked, and in case of violation of these conditions, a fault is detected.Impedance change methods are generally reliable during slow PS (the frequency is less than 1 Hz) when impedance changes are very slow [32]. In contrast, during a fast PS (the frequency is more than 5 Hz), distance relays fail to block their operation and consider the PS as a fault [27, 28, 33‐35]. Furthermore, according to [17, 36], these methods cannot detect multi‐mode PS. In addition, the fault detection time during faults that simultaneously occurred with the PS in these methods is high [17, 36].Double blinder methodDouble blinder methods are generally developed to tackle the problems of impedance methods [37]. Since, during PS, the impedance path entered into the relay zones is approximately perpendicular to the line impedance, blinders are placed in parallel with the line impedance (Figure 6) [16]. Blinders methods perform much better than impedance methods [37–41]. The method of double blender is very similar to the impedance changes approach, with the difference that in the impedance method, an external zone is used to detect PS, however, in the blinder method two parallel blinders are used. If the impedance curve passes through blinder number 1, the counter will start, and at the time of passing through blinder number 2 it will stop. Afterward, acquired time is compared with the threshold value, and if it is more than the threshold value, the PS would be detected; otherwise, this condition will be considered as a fault. The most important problem with the double blinder method is the need for complex analysis and calculations for determining the threshold value. In addition, this method may cause relay maloperation during stable PS [16, 38].6FIGUREDouble blinder method operation in detecting PSDecreased resistance methodIn order to tackle the problems of the double blinder method, the decreased resistance method has been proposed. The principles of this method are based on continuous changes in impedance resistance (during PS the impedance resistance changes continuously but during a fault the resistance is constant) [26, 28]. The most important advantage of this method is the ability to distinguish between fault and PS. In addition, this method does not require extensive system analysis like the double blinder method [28]. On the other hand, when the network experiences a PS with low frequency, this method may detect PS as a fault at a power angle of 180 degrees which has significant changes in the impedance resistance [11]. This method will also fail to detect a three‐phase fault during slow PS [11].Swing centre voltage methodSwing centre voltage methods operate based on Vcosφ${V_{\cos \varphi }}$ variations. V is the amount of voltage at the relay location and ϕ is the angle between voltage and current measured at the relay location [39]. During a PS Vcosφ${V_{\cos \varphi }}$ continuously changes, but it remains almost constant during a fault. This method uses this criterion to distinguish between fault and PS. The most important advantage of this method is its independence from source and line impedance, which simplifies analysis. However, three‐phase fault detection in this method takes a long delay [26, 38]. This delay can cause the backup relays to operate, resulting in miscoordination. This method also needs to determine an appropriate threshold value for proper operation [16]. The threshold value for PS at a power angle close to 180° should be very low [38]. For this purpose, several threshold settings must be obtained for different conditions. This will complicate the operation of the method since it is very difficult to determine the network conditions to identify the threshold value. This method also has a problem with the detection of a high impedance fault occurrence so that detects the fault as a PS [26].Difference between three‐phase signalsIn recent years, researchers' interest in using the variation rate of three‐phase signals as a method for detecting PS has increased. The main basis of these methods is to study the variation rate of a signal obtained from the current and compare it for all three phases. Equation (4) shows the condition for detecting PS in these methods [24].where, di$di$ is the variation rate of the signal obtained from the current of each phase.4ifdiphase(a)≈diphase(b)≈diphase(b)⇒PSB=True.$$\begin{equation}if{\rm{ }}d{i^{phase(a)}} \approx d{i^{phase(b)}} \approx d{i^{phase(b)}} \Rightarrow PSB = True.\end{equation}$$Reference [24] uses the variation rate of the root mean square (RMS) of the current to distinguish PS from the fault. This method has the ability to detect various types of faults along with PS, especially high impedance faults [3]. Among the shortcomings of this method are the high fault detection time and non‐detection of asymmetric PS (PS occurs in one phase) [3]. Reference [42] uses Hilbert Transform to reduce fault detection time. This method can detect PS with a phase angle of 180 degrees. In addition, it has the ability to detect multi‐mode PS. This method is noise resistant and operates properly. Although this method has succeeded in reducing the fault detection time to 30 ms. However, this fault detection time is still high [11]. Authors in [10] compared the rate of instantaneous frequency variations to distinguish between fault and PS. This method can detect stable and unstable PS with a variety of faults that are simultaneous with PS. Among the most critical shortcomings of this method are high impedance fault detection time, non‐detection of multi‐mode PS, and inappropriate operation when the current signal is noisy [21].In [43], a method based on three‐phase power differences is presented to differentiate between PS and fault. During PS, the active and reactive powers of the three phases are constantly changing. At the time of a fault, the normalized values of active and reactive power are close to zero. One of the most important advantages of this method is its high sensitivity in detecting PS. Also, this method is not affected by the fault start time and other system parameters. Disadvantages of this method include the possibility of incorrect operation during a stable PS and correct operation only for symmetric faults.Signal processing methodsThese methods use various signal processing techniques to analyse the fault and PS characteristics. The most important signal processing technique that has received much attention over the years is the Fast Fourier Transform (FFT). Accordingly, [44] uses the value of the DC component obtained from FFT for differentiation PS from fault. Authors in [35] present a method based on the three‐phase active power frequency component using FFT approach [35, 44‐46]. Reference [36] provides a method based on creating a reference signal from a three‐phase current and obtaining the main component of the signal using FFT. This method has the ability to detect different types of PS and their simultaneous faults. In addition, low operating time is the most essential feature of this method [23]. It should be noted that all the methods reviewed in this section require a threshold value for proper performance [16, 35, 44].Since the frequency of a system varies around the nominal frequency during PS, the application of the wavelet transform (WT) to detect power swing has received a growing interest in recent years [47–49]. However, these conditions are not met during a permanent fault [47, 49‐53]. Accordingly, in [50] different sampling frequencies have been used to detect PS and faults. Authors in [49] present a high‐speed WT algorithm for detecting symmetric faults during PS. Reference [52] has used wavelet transform on impedance variations to distinguish between PS and fault. In [53], WT is used to analyse current changes to detect high impedance faults which occur simultaneously with PS. Authors in [54] have proposed a WT‐based high‐velocity PS detection method. Reference [55] also presents an approach based on a combination of improved discrete wavelet transform (IM‐DWT) and deep learning. It should be noted that despite the significant advantages of WT, the most critical drawback of these methods is their dependency on high sampling rates to achieve appropriate accuracy [21].In many studies, Prony method is used in power system protection [56, 57]. In [58], Prony has been proposed as a more appropriate method than FFT for detecting PS. In [57], an algorithm using the current waveform is proposed to detect a three‐phase fault during PS. In this study, only low impedance faults have been discussed. In [56], an algorithm based on the DC component of the current waveform is proposed.Various applications for empirical mode decomposition (EMD) in power systems have been proposed [59–61]. Accordingly, [62] presents a three‐step EMD‐based algorithm for differentiating PS from fault. In the first step, the intrinsic mode functions (IMF) values ​of the signal are extracted using EMD. In the second step, the instant amplitude and frequency of the first IMF are calculated using Hilbert–Huang transform (HHT). Finally, Teager energy is estimated using the instant amplitude and frequency. This method lacks the ability to detect multi‐mode PS. In addition, it may cause maloperation during high impedance fault [3]. Authors in [3] present an algorithm based on the rate of changes of the third IMF in each phase. In this study, first, an appropriate type of IMF for fault detection is defined according to the topology of the system. Then, a threshold value is obtained using different computer simulations. Whenever the IMF value exceeds the threshold value, the algorithm considers it as a fault. The most important drawback of the proposed method in [3] is its problem in detecting PS in different networks and its requirement for complex analysis to find the appropriate IMF [23].In [63, 64], S‐transform based method is proposed to differentiate fault from stable and unstable PS. In [65], active and reactive power is used as the input of the relay. In this study, it has been shown that during a fault, the S‐transform obtained from the active and reactive power is a value greater than zero, while in the event of a PS, this value is zero. Reference [66] uses the S‐transform to analyse voltage and current signals. However, the most important shortcoming of such methods is the lack of detection of high impedance faults, which are simultaneous with PS [67].Artificial intelligence methodWith the advancement of technology and the growing interest in artificial intelligence in various fields, researchers began to use artificial intelligence to protect power systems [68–70]. One of the most widely used of artificial intelligence methods is the adaptive neuro fuzzy inference system (ANFIS) method. ANFIS is an advanced version of the fuzzy logic system with generalizability, noise protection, reliability, and high fault tolerance [71–73]. Reference [73] uses positive sequence impedance and positive and negative current sequence as the input signals to ANFIS. This method has the ability to detect a variety of power swings and faults. Similarly, in [71], the rate of current variations and normalized active and reactive powers are used as ANFIS inputs. This method is not affected by system parameters including fault time, fault position, and pre‐loading conditions. However, the most important shortcoming of these kind of methods is the need for many simulations to train different fault types and power swings [74]. Reference [23] used GMDH neural network to differentiate PS and fault. In this reference, GMDH is trained to estimate the signal in the normal situation and during PS. During different faults, there is a significant difference between the estimated value and the actual value of the signal, which is implemented to detect different power swings and faults.In [75], the performance of the support vector machine (SVM) is considered to be better than the artificial neural network. Accordingly, authors in [76] have proposed a new method for differentiating fault from PS using the minimum square of the SVM model. This method uses the current and voltage signals at the relay location as the inputs. This method can differentiate fault from PS and also has the ability to detect faults that are simultaneous with PS. In [77], a WT‐based approach and SVM is proposed. This method can differentiate PS from faults and has the ability to detect faults that are simultaneous with PS.Reference [78] used the C‐MEAN classification method to detect faults which are simultaneous with PS. C‐MEAN is a fuzzy clustering method for automatic data classification. In contrast to ANFIS and SVM methods, this method is of the unsupervised type and does not require training with various fault types for performance. Authors in [79] have proposed a new method for differentiating fault from PS using the improved deep neural network (IDNN) with spider monkey optimization (SMO) algorithm.Other methodsReference [74] provides a method based on the moving window averaging (MWA) to detect PS. MWA is a low‐pass filter applied to a distance relay. This method is less sensitive to noise, and it also has the ability to detect faults that are simultaneous with PS. However, this method shows a maloperation during multi‐mode PS [11, 21].Authors in [80] have used transient monitoring (TM) to detect PS. This method compares TM with a predetermined threshold value. If the calculated TM is greater than the threshold value, a fault is detected; otherwise, PS is detected. In this method, phasor estimation plays an important role in determining the threshold value. Although there are different methods for estimating the phasor [81–86], this method uses dynamic phase estimation to improve the performance of the proposed method.Reference [87] provides a method based on the concept of the first zero crossing signal to detect PS. This method can block the third zone of the distance relay during unstable PS and prevent the relay maloperation. This method is not dependent on the network topology and, unlike the conventional blinder technique, does not require lots of stability studies to find the relay settings. An algorithm based on participatory factors acquired from the variance/covariance matrix is ​proposed in [88]. In this method, modal analysis is applied to obtain more properties for the transient response to the matrix, and a threshold for the difference between fault and PS is identified.In [17], a method based on the rate of current variance variation is proposed. In this reference, the variance of the sampled data from the current signal is calculated during a signal window, and PS and fault are differentiated by setting a threshold value. This method has the ability to detect different PS types, including stable, unstable, and multi‐mode. In addition, changes in angle and impedance of the fault do not affect the proposed method. Reference [89] has proposed a method for detecting PS based on chaos theory. The proposed method in this study is based on the logistic map, which is a simple mathematical model for chaos theory (according to the model presented in [90]). The method presented in [89] has the ability to detect different types of PS, including asymmetric types. In addition, this method has a precise operation when the network experiences a capacitor bank switching. Authors in [21] have proposed a new approach based on the compressed sensing theory (CST). In this approach, CST is implemented to distinguish PS and fault. This method does not require a high sampling rate and has a high fault detection speed. The proposed method also has the ability to detect various types of PS and faults and operates well on compensated lines. A Phaselet based method is proposed in [11], which uses Phaselet to estimate the current data and compares it with the actual data. A fault is detected if the difference between the estimated data and the real data exceeds a certain value. Researchers in [9] presented a method based on the signal energy variation rate to distinguish between PS and fault.Table 1 provides an overview of the available methods for detecting PS based on the studies analysed in this section.1TABLEComparison between different power swing detection methodsPS detectionDetecting faults simultaneously with PSMethodReferencesStableUnstableMulti‐modeSingle phaseTwo phasesThree phasesThreshold value requirementImpedance changes[12, 29, 30]✓✓×✓✓✓✓Double blinder[91]✓✓×✓✓×✓Decreased resistance[28]✓✓×✓✓✓✓Swing centre voltage[92]✓✓×✓✓✓×RMS[24]✓✓✓✓✓✓✓Hilbert transform[42]✓✓✓✓✓✓✓Instantaneous frequency[10]✓✓×✓✓✓✓Three phases power variation[43]✓✓×××✓×FFT[35, 44]✓✓×✓✓✓✓[36]✓✓✓✓✓✓✓WT[49, 50, 52]✓✓×✓✓✓✓Prony[57, 58]✓✓×✓✓✓✓EMD[62]✓✓×✓✓✓✓[3]✓✓✓✓✓✓✓S transform[63, 64]✓✓×✓✓✓✓ANFIS[71, 73]✓✓✓✓✓✓✓SVM[76, 77, 93]✓✓✓✓✓✓✓C‐MEAN[78]✓✓×✓✓✓×GMDH[23]✓✓✓✓✓✓✓MWA[74]✓✓×✓✓✓✓TM[80]✓✓×✓✓✓✓Variance[17]✓✓✓✓✓✓✓Chaos[89]✓✓✓✓✓✓✓CST[21]✓✓✓✓✓✓✓Phaselet[11]✓✓✓✓✓✓✓METHODS TO IDENTIFY THRESHOLD VALUE FOR PS DETECTIONBeing evidently presented in Table 1, most PS detection methods require optimal threshold values to differentiate PS from fault. This proves the importance of determining the correct threshold value. Studies have proposed different methods for determining threshold values. These researches are presented in the following section.Simulating different PS and fault ratesReferences [3, 21, 45, 89] use a threshold identification method based on numerical simulations of the different fault and PS rates. Therefore, to determine the threshold value for a network, different types of PS and faults are simulated based on the items mentioned in Table 2 [21]. Table 2 shows the used and unused items in each case with ✓ and ×, respectively. In the next step, the maximum points (peak value of the signal) obtained for each case in Table 2 are calculated. Finally, the threshold value is determined according to the recorded values for different types of PS and faults [21].2TABLESimulation cases to determine the threshold valuePower swingFaultDistance fault (%)CaseStableUnstableMulti‐modeSingle‐phaseThree‐phaseHigh impedance570901✓××✓××✓✓✓2✓×××✓×✓✓✓3✓××××✓✓✓✓4×✓×✓××✓✓✓5×✓××✓×✓✓✓6×✓×××✓✓✓✓7××✓✓××✓✓✓8××✓×✓×✓✓✓9××✓××✓✓✓✓Although the threshold value determining method, which is presented in [17], is based on numerical simulation, it has some differences from the previously stated method. This method is based on the principle that in a power system the highest amplitude among the different types of power swings is related to the unstable power swing. Also, the lowest amplitude between the different types of faults is related to the high impedance fault. Therefore, to obtain the threshold value in the power system, it is only required to simulate these two cases. On this basis, this reference uses Equation (5) to determine the threshold value (K). As it is clear from this equation, the threshold value can be obtained only by obtaining the maximum and minimum peak values expressed in the two cases.5K=mean([max(UnstablePS)max(Highimpedancefault]).$$\begin{equation}K = mean([\max (Unstable{\rm{ }}PS){\rm{ }}\max (High{\rm{ }}impedance{\rm{ }}fault]).\end{equation}$$Adaptive method based on permissible faultUsing the numerical simulation method to define a suitable threshold value for different network types increases study time and therefore increases design costs [19]. It also increases human errors in designing protection systems and setting distance relays [19]. For this reason, in [19], a method to determine the adaptive threshold value based on the acceptable error of relays, is proposed. The threshold value is proportional to the tolerance value in most relays and is equal to 5% and 50 mA. The reason for using two values ​is that for currents with a small amplitude (for example 200 mA) the value of 5% of the current will be very small. Therefore, the algorithm selects 50 mA. In contrast, in networks with high current levels, the value of 50 mA is minimal, and the algorithm selects 5% as the threshold value. Figure 7 shows the algorithm for calculating the threshold value using the adaptive method [19].7FIGUREThreshold value algorithm by an adaptive methodSwarm intelligence methodIn [94], a new method for determining the threshold value is proposed based on swarm intelligence. In this method, the mechanism of fault parameters, PS, and the minimum values ​of detection indices are identified using particle swarm optimization (PSO) techniques. In this method, the optimal threshold value is determined by extensive case studies and using the PSO algorithm (examining 150,000 simulation cases in an iterative process). Figure 8 shows the threshold value detection algorithm by the swarm intelligence method [94].8FIGUREThreshold value algorithm by swarm intelligence methodEFFECTS OF COMPENSATED LINES ON PS DETECTION METHODSThe application of a flexible AC transmission system (FACTS) in transmission lines can have a significant impact on the impedance characteristic of PS and the performance of protection designs [95]. For this reason, this section examines the available research on PS detection in compensated lines.Impacts of a unified power flow controllerA unified power flow controller (UPFC) is a FACTS device used to control the current of transmission lines and to improve system stability [96, 97]. In transmission lines equipped with UPFC, ordinary distance relays are overreached or underreached during a fault [97]. For this reason, various studies have been performed to investigate the effect of FACTS compensator devices on distance relays during fault.The effect of UPFC on oscillating properties (radius and centre point changes) is investigated in [98]. According to this study, transmission line parameters change in the presence of UPFC. In [97], the impacts of UPFC on methods of the power variations rate, the main frequency of instantaneous three‐phase active power, the rate of impedance changes, and the transient monitor are investigated. In this reference, it is analytically proven that since the index value of the power variations rate method decreases during power swing, UPFC affects the performance of this method. On this basis, the presence of UPFC prevents the application of power variations rate method for PS detection [97]. Also, the studies performed in [97] show that the method's reliability of the main frequency estimation of three‐phase instantaneous active power decreases under compensated conditions. The transient monitor method is also influenced by the UPFC of the harmonic influence of multi‐level converters [97]. Therefore, it is clear that the performance of UPFC can have a negative effect on different PS detection methods and lead to incorrect performance of these methods.Effects of TCSCA thyristor‐controlled series capacitor (TCSC) is used to improve the transmission capabilities of the power line. However, due to the inversion of voltage and current as well as the nonlinear nature of the metal oxide varistor, many protection approaches are not applicable in the TCSC environment [99]. The presence of TCSC can have many effects on transmission line protection, including the impact on the calculated apparent impedance, the impact on the PS frequency, and also the impact on the phase angles measured in a bus [14, 100, 101]. Therefore, the application of TCSC in the transmission network increases the protection challenges. In addition, the operation of protection relays such as fault detection, PS detection, and detection of faulty section faces serious challenges [102–104]. For this reason, the presence of TCSC requires advanced protection techniques to detect all types of faults and power swings. Accordingly, in [101], the Teager Kaiser Energy Operators (TKEOs) of negative sequence technique of voltage and current have been used to detect PS in TCSC compensated lines. In [103], a wide‐area backup protection approach is proposed to protect TCSC compensated lines. In this method, negative and zero sequences of voltage values are implemented for detecting asymmetric faults, and positive sequence voltages are used for symmetric faults. This method may fail to detect simultaneous faults with PS [99]. The authors in [100] have used the Clark transform to detect PS in transmission lines equipped with TCSC. In [104], a high‐speed fault detection method is proposed based on mathematical morphology in TCSC‐compensated lines.THE EFFECT OF WIND FARMS ON PS DETECTION METHODSEnvironmental problems along with increasing energy demand, have drawn the attention to the application of wind turbines at different network levels [18, 105]. However, the existence of wind turbines in the network can significantly change the PS characteristics and pose major challenges for existing PS detection methods [106]. Maloperation of PS detection methods can increase with increasing wind production. The presence of a wind turbine can also change the PS impedance path and affect the performance of the out‐of‐step protection [13].To investigate the effects of the wind turbines on PS characteristics as well as their impact on several PS detection methods, the IEEE standard 39‐bus network has been simulated using DIgSILNET and MATLAB software. Figure 9 shows the single line diagram of this network. More information on this network is available in [11, 17]. In order to simulate wind farms, two wind farms are connected to buses 26 and 29. The protection relay is placed on bus number 26 to protect the line connecting bus 26 to bus 29.9FIGURESingle‐line diagram of IEEE 39‐bus networkInstalling a wind turbine in the network may affect the PS by changing the rotor angle of the synchronous generator. Figure 10 shows the rotor angle variations of synchronous generator No. 9, which is connected to bus 38 in the presence and absence of a wind turbine. As demonstrated in this figure, when the wind turbine is added to the network, stable PS becomes unstable.10FIGUREChanges in the rotor angle of a synchronous generatorInstalling a wind turbine on a network may also affect the impedance trajectory. Figure 11 shows the impedance trajectory in two network conditions, when the wind turbine is connected or disconnected from the network. As can be seen from this figure, when the wind turbine is connected, the impedance trajectory is very different. In addition, as presented in Figure 12, in the presence of a wind turbine, the current waveform may not be sinusoidal anymore and become very noisy.11FIGUREImpedance trajectory12FIGURECurrent waveform during PSIn the following, the effect of adding wind turbines to the network on four PS detection methods (including Taylor Series [107], Phaselet [11], chaos theory [89], and CST [21]) is investigated. The simulation results for each method are depicted in Figures 13–16, respectively. As it is clear from these figures, all the methods have exceeded the considered threshold value and cannot differentiate fault from PS.13FIGUREPerformance of the Taylor Series method14FIGUREPerformance of the Phaselet method15FIGUREPerformance of the Chaos method16FIGUREPerformance of the CST methodPS FUNCTION TEST METHODSProtection relays always require various tests during the development, commissioning, maintenance, configuration, and troubleshooting stages [108]. Similar to other algorithms, PS algorithms in protection relays require different tests. Accordingly, various test methods for the PS function are discussed in this section.Z‐Tracking methodZ‐tracking is a conventional and common method of PS function testing [109]. It provides the ability to change the impedance at any angle, speed, and in any direction of the distance relay zones. As Figure 17 presents, in this method, it is only required to specify the starting and ending points of entering the zone, the exit point and movement speed, and the amount of voltage and current. This method also has the ability to detect fault that is simultaneous with PS. However, since the PS waveforms are not accurate, the reliability of this method is low [109].17FIGUREPS testing via Z‐Tracking methodTransient simulation of PS in transient analysis softwareUnlike classic tests, network simulations in software such as EMTP, PSCAD, and DIGSILENT lead to waveforms that are very similar to the actual waveforms of the network. Accordingly, PS functions can be tested with high reliability via this method. However, these methods require a long analysis and testing time for execution [110]. Figure 18 depicts an example of the PS generated from transient analysis software. This type of testing has drawn the attraction of many researchers, including references [3, 11, 17, 42] in order to evaluate the performance of their proposed method.18FIGUREPS generated from simulation software in relay test softwareTesting via the information received from Comtrade fileDigital relays have the ability to record events of the network and give Comtrade file output. Since these outputs are actual waveforms from transmission lines, they are an appropriate reference for testing relays. This type of test provides the most realistic test possible for PS function testing. A big challenge with this type of test is the limited number of Comtrade files [110]. In addition, these files cannot be used to test different relays with different settings. Figure 19 shows the entering of the impedance seen by the distance relay into its protection zones during PS as a result of testing with a relay Comtrade file.19FIGUREEntering of the impedance seen by distance relay into its protection zones during PSDISCUSSIONDiscussing the existing research in the PS field proves the requirement for further studies in this field. Being previously mentioned, existing non‐intelligent methods have shortcomings such as inability to detect multi‐mode PS, inability to detect faults that are simultaneous with PS, failure to operate during high impedance faults, dependency on the high sampling rate, and maloperation when the signal is under a noisy condition. Although intelligent methods are able to solve the challenges of non‐intelligent methods, their most important challenge is the necessity for many offline simulations to train the algorithm for different faults and PS conditions.These simulations can significantly increase design duration and cost. In addition, intelligent methods require far more powerful hardware to perform. This can be simply demonstrated using the Python programming language and the psutil library. The psutil library has the ability to record the amount of CPU usage for each software using Process ID (PID). In this section, two different power swing detection algorithms (a non‐intelligent algorithm and an intelligent algorithm) are run separately by MATLAB software, and CPU consumption is recorded by code written in Python. Figure 20 shows CPU usage in each mode. According to this figure, non‐intelligent methods can be implemented on a conventional microprocessor. However intelligent methods need powerful hardware.20FIGURECPU usage in non‐intelligent and intelligent PS detection algorithmsOn the other hand, as previously shown, the presence of compensators on transmission lines and the connection of wind farms to these lines can significantly impact on protection algorithms. Although many researches have been carried out in this field, various challenges such as voltage and current reversal increase the requirement for further research in this field. Therefore, the following subjects are suggested for additional researches in the PS protection.Provide methods for detecting PS without the need to determine the threshold value.Reduce the required sampling rate.Introducing new methods to determine the appropriate threshold values for different methods.Provide unsupervised artificial intelligence methods to reduce the necessary simulations.Investigate the effect of other types of compensators on PS.Provide methods for detecting PS in lines connected to the wind turbines.In addition to the mentioned issues, detecting power swings in conventional electromechanical and electrostatic distance relays is still considered a fundamental challenge. In fact, these types of relays do not have the possibility of detecting power swings due to the lack of using microprocessors. It is also not possible to implement new methods of power swing detection methods in these types of relays. For this reason, a suggestion for future work in the field of power swing detection can be the design and construction of suitable hardware to use in these types of relays. This hardware must be in line with the operational performance of these types of relays and also have the ability to simply implement new methods.CONCLUSIONSThe present paper reviews different types of power swing and their detection methods. The focus of this study is on power swing types and various factors that can affect the methods of detecting power swings. As studies have shown, power swings can be divided into three general categories of stable, unstable, and multi‐mode power swings. As the result of this review shows, multi‐mode power swing is a serious challenge for power swing detection methods due to complexities such as non‐sinusoidal current waveform and variable frequency that may occur during this type of power swing. After introducing different types of power swing, the common methods for detection of power swing were investigated. Among the available methods, the methods using signal processing are more typical. However, the main challenge with these methods is the necessity of determining a threshold value for proper operation. On this basis, the methods of determining the threshold value were investigated in different studies. The effects of compensated lines and wind turbines on the power swing detection methods have been investigated. Despite the necessity of an actual test of the power swing detection function, this issue has been considered in a few studies. Various approaches to power swing test methods and studies conducted in this field were also analysed in this article.AUTHOR CONTRIBUTIONBehrooz Taheri (BT): Conceptualization, Methodology, Software, Visualization, Investigation. Seyed Amir Hosseini (SAH): Conceptualization, Data curation, Writing‐ review and editing, Supervision, Software. Sirus Salehimehr (SS): Writing‐ original draft, Resources, Validation.CONFLICT OF INTERESTThe authors declare no conflict of interest.FUNDING INFORMATIONNone.DATA AVAILABILITY STATEMENTData sharing is not applicable to this article as no new data were created or analysed in this study. cd_value_code=textREFERENCESJannati, M., Mohammadi, M.: A novel fast power swing blocking strategy for distance relay based on ADALINE and moving window averaging technique. IET Gener. Transm. Distrib. 15(1), 97–107 (2021)Doria‐García, J., Orozco‐Henao, C., Leborgne, R., Montoya, O.D., Gil‐González, W.: High impedance fault modeling and location for transmission line. Electr. Power Syst. Res. 196, 107202 (2021)Taheri, B., Hosseini, S.A., Askarian‐Abyaneh, H., Razavi, F.: Power swing detection and blocking of the third zone of distance relays by the combined use of empirical‐mode decomposition and Hilbert transform. IET Gener. Transm. Distrib. 14(6), 1062–1076 (2020)Taheri, B., Hosseini, S.A., Askarian‐Abyaneh, H., Razavi, F.: A new method for remote testing distance relay using internet protocol version 4. IEEE International Conference on Protection and Automation of Power System (IPAPS), pp. 31–37. IEEE, Piscataway (2019)Taheri, B., Hosseini, S.A., Sedighizadeh, M., Khatibi, M.: A moving window numerical distance protection based on flat‐top signal windowing. Arab J. Sci. Eng. 47, 14249–14266 (2022)Khodaparast, J., Khederzadeh, M., da Silva, F.F., Bak, C.L.: A novel approach to detect faults occurring during power swings by abrupt change of impedance trajectory. Electr. Power Comp. Syst. 45(15), 1638–1652 (2017)Moazzami, M., Hooshmand, R., Khodabakhshian, A., Yazdanpanah, M.: Blackout prevention in power system using flexible AC transmission system devices and combined corrective actions. Electr. Power Comp. Syst. 41(15), 1433–1455 (2013)Devi, R., et al.: Development of symmetrical fault detection during power swing based on entropy. J. Electr. Eng. Technol. 17, 1533–1542 (2022)Taheri, B., Salehimehr, S., Razavi, F.: A new method for fast power swing detection using the rate of change of energy in the current signal. In: IEEE 27th Iranian Conference on Electrical Engineering (ICEE), pp. 464–469. IEEE, Piscataway (2019)Taheri, B., Salehimehr, S., Razavi, F., Parpaei, M.: Detection of power swing and fault occurring simultaneously with power swing using instantaneous frequency. Energy Syst. 11(2), 491–514 (2020)Ghalesefidi, M.M., Ghaffarzadeh, N.: A new phaselet‐based method for detecting the power swing in order to prevent the malfunction of distance relays in transmission lines. Energy Syst. 12(2), 491–515 (2021)Ziegler, G.: Numerical Distance Protection: Principles and Applications. John Wiley & Sons, Boca Raton (2011)Haddadi, A., Kocar, I., Karaagac, U., Gras, H., Farantatos, E.: Impact of wind generation on power swing protection. IEEE Trans. Power Delivery 34(3), 1118–1128 (2019)Sai Kumar, M., Kumar, J., Mahanty, R.: Fault classification in a TCSC compensated transmission line during power swing using Wigner Ville transform. IETE J. Res. 1–22 (2021)Hida, S., Pradhan, V., Naidu, O., Cheriyan, E.P.: A critical review of distance relaying techniques under power swing condition. In: IEEE PES GTD Grand International Conference and Exposition Asia (GTD Asia), pp. 500–505. IEEE, Piscataway (2019)Gunasegaran, M., Tan, C., Bakar, A.H.A., Mokhlis, H., Illias, H.A.: Progress on power swing blocking schemes and the impact of renewable energy on power swing characteristics: A review. Renew. Sustain. Energy Rev. 52, 280–288 (2015)Salehimehr, S., Taheri, B., Faghihlou, M.: Detection of power swing and blocking the distance relay using the variance calculation of the current sampled data. Electr. Eng. 104, 913–927 (2022)Machowski, J., Lubosny, Z., Bialek, J.W., Bumby, J.R.: Power System Dynamics: Stability and Control. John Wiley & Sons, Hoboken, NJ (2020)Taheri, B., Faghihlou, M., Salehimehr, S., Razavi, F.: Symmetrical fault detection during power swing using mean value of sampled data from the current signal. IETE J. Res. 1–13 (2020)Kumar, J., Jena, P.: Fault detection during asymmetrical power swing using superimposed negative sequence current. Arab J. Sci. Eng. 44(8), 7033–7046 (2019)Taheri, B., Sedighizadeh, M.: Detection of power swing and prevention of mal‐operation of distance relay using compressed sensing theory. IET Gener. Transm. Distrib. 14(23), 5558–5570 (2020)Hashemi, S.M., Sanaye‐Pasand, M., Shahidehpour, M.: Fault detection during power swings using the properties of fundamental frequency phasors. IEEE Trans. Smart Grid. 10(2), 1385–1394 (2017)Hosseini, S.A., Taheri, B., Abyaneh, H.A., Razavi, F.: Comprehensive power swing detection by current signal modeling and prediction using the GMDH method. Protect. Control Modern Power Syst. 6(1), 1–11 (2021)Taheri, B., Razavi, F.: Power swing detection using rms current measurements. J. Electr. Eng. Technol. 13(5), 1831–1840 (2018)Blumschein, J., Yelgin, Y., Kereit, M.: Proper detection and treatment of power swing to reduce the risk of blackouts. In: Third International Conference on Electric Utility Deregulation and Restructuring and Power Technologies, pp. 2440–2446. IEEE, Piscataway (2008)Nayak, P.K., Rao, J.G., Kundu, P., Pradhan, A., Bajpai, P.: A comparative assessment of power swing detection techniques. In: IEEE Joint International Conference on Power Electronics, Drives and Energy Systems & 2010 Power India, pp. 1–4. IEEE, Piscataway (2010)Lin, X., Li, Z., Ke, S., Gao, Y.: Theoretical fundamentals and implementation of novel self‐adaptive distance protection resistant to power swings. IEEE Trans. Power Delivery 25(3), 1372–1383 (2010)Khoradshadi‐Zadeh, H.: Evaluation and performance comparison of power swing detection algorithms. In: IEEE Power Engineering Society General Meeting, pp. 1842–1848. IEEE, Piscataway (2005)SIPROTEC, S.: Distance Protection 7SA522 V4. 70, Instruction Manual (2011)Ganeshan, I.S., Jena, M.K., Panigrahi, B.K.: An improved setting‐free power swing blocking scheme using local end information. In: IEEMA Engineer Infinite Conference (eTechNxT), pp. 1–6. IEEE, Piscataway (2018)Jurisch, A., Schwenke, M.: Method of deriving a signal indicating an oscillation in an electric power supply system. Google Patents (2000)Corsi, S., Sabelli, C.: General blackout in Italy Sunday September 28, 2003, h. 03: 28: 00. In: IEEE Power Engineering Society General Meeting, pp. 1691–1702. IEEE, Piscataway (2004)Martuscello, L., Krizauskas, E., Holbach, J., Lu, Y.: Tests of distance relay performance on stable and unstable power swings reported using simulated data of the August 14 th 2003 system disturbance. In: IEEE Power Systems Conference, pp. 1–21. IEEE, Piscataway (2009)Mohamad, N.Z., Abidin, A.F., Musirin, I.: Adaptive distance protection to prevent false relay tripping during power swing. Appl. Mech. Mater. 785, 358–362 (2015)Mahamedi, B.: A very fast unblocking scheme for distance protection to detect symmetrical faults during power swings. In: IEEE Conference Proceedings IPEC, pp. 378–383. IEEE, Piscataway (2010)Taheri, B., Faghihlou, M., Salehimehr, S., Razavi, F.: A fast Fourier transform‐based method for power swing detection and distance relay malfunction prevention. J. Control Autom. Electr. Syst. 31(6), 1458–1468 (2020)Zadeh, H.K., Li, Z.: Artificial neural network based load blinder for distance protection. In: IEEE Power and Energy Society General Meeting‐Conversion and Delivery of Electrical Energy in the 21st Century, pp. 1–6. IEEE, Piscataway (2008)Ambekar, V.A., Dambhare, S.S.: Comparative evaluation of out of step detection schemes for distance relays. In: IEEE Fifth Power India Conference, pp. 1–6. IEEE, Piscataway (2012)Fischer, N., Benmouyal, G., Hou, D., Tziouvaras, D., Byrne‐Finley, J., Smyth, B.: Do system impedances really affect power swings—Applying power swing protection elements without complex system studies. In: IEEE 65th Annual Conference for Protective Relay Engineers, pp. 108–119. IEEE, Piscataway (2012)Holbach, J.: New out of step blocking algorithm for detecting fast power swing frequencies. In: IEEE Power Systems Conference: Advanced Metering, Protection, Control, Communication, and Distributed Resources, pp. 182–199. IEEE, Piscataway (2006)Tziouvaras, D.A., Hou, D.: Out‐of‐step protection fundamentals and advancements. In: IEEE 57th Annual Conference for Protective Relay Engineers, pp. 282–307. IEEE, Piscataway (2004)Salehimehr, S., Taheri, B., Hosseini, S.A., Abyaneh, H.A., Razavi, F.: A new power swing detection method based on Hilbert Transform. Int. J. Indus. Electron. Control Optim. 3(2), 103–114 (2020)Lin, X., Gao, Y., Liu, P.: A novel scheme to identify symmetrical faults occurring during power swings. IEEE Trans. Power Delivery 23(1), 73–78 (2007)Karegar, H.K., Mohamedi, B.: A new method for fault detection during power swing in distance protection. In: IEEE 6th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, vol. 1, pp. 230–233. IEEE, Piscataway (2009)Mahamedi, B., Zhu, J.G.: A novel approach to detect symmetrical faults occurring during power swings by using frequency components of instantaneous three‐phase active power. IEEE Trans. Power Delivery 27(3), 1368–1376 (2012)D'Apuzzo, M., D'Arco, M.: A time‐domain approach for the analysis of nonstationary signals in power systems. IEEE Trans. Instrum. Meas. 57(9), 1969–1977 (2008)Fu, L., He, Z., Bo, Z.: Wavelet transform and approximate entropy based identification of faults in power swings. In: 2008 IET 9th International Conference on Developments in Power System Protection (DPSP 2008), pp. 590–594. IET, Stevenage (2008)Parameswariah, C., Cox, M.: Frequency characteristics of wavelets. IEEE Trans. Power Delivery 17(3), 800–804 (2002)Pang, C., Kezunovic, M.: Fast distance relay scheme for detecting symmetrical fault during power swing. IEEE Trans. Power Delivery 25(4), 2205–2212 (2010)Brahma, S.: Use of wavelets for out of step blocking function of distance relays. In: IEEE Power Engineering Society General Meeting, p. 5. IEEE, Piscataway (2006)Dubey, R., Samantaray, S.R.: Wavelet singular entropy‐based symmetrical fault‐detection and out‐of‐step protection during power swing. IET Gener. Transm. Distrib. 7(10), 1123–1134 (2013)Mahamedi, B.: A new power swing blocking function based on wavelet transform. In: IEEE 2nd International Conference on Electric Power and Energy Conversion Systems (EPECS), pp. 1–6. IEEE, Piscataway (2011)Dubey, R., Samantaray, S., Tripathy, A., Babu, B.C., Ehtesham, M.: Wavelet based energy function for symmetrical fault detection during power swing. In: IEEE Students Conference on Engineering and Systems, pp. 1–6. IEEE, Piscataway (2012)Alsyoufi, Y.R., Hajjar, A.A.: A high‐speed algorithm to discriminate between power swing and faults in distance relays based on a fast wavelet. Electr. Power Syst. Res. 172, 269–276 (2019)Sriram, C., Somlal, J.: IM‐DWT with DNN based blocking scheme of third zone distance relay in power swing condition. Smart Sci. 1–20 (2022)Thakallapelli, A., Mehra, R., Mangalvedekar, H.: Differentiation of faults from power swings and detection of high impedance faults by distance relays. In: IEEE 1st International Conference on Condition Assessment Techniques in Electrical Systems (CATCON), pp. 374–377. IEEE, Piscataway (2013)Lotfifard, S., Faiz, J., Kezunovic, M.: Detection of symmetrical faults by distance relays during power swings. IEEE Trans. Power Delivery 25(1), 81–87 (2009)Leonowicz, Z.: Application of the Prony method for compensation of errors in distance relays. In: IEEE 12th International Conference on Environment and Electrical Engineering, pp. 568–572. IEEE, Piscataway (2013)Wang, Y., Wang, J., Meng, Q., Yu, H.: Power system osciliation modes identification based on eigen system realization algorithm via empirical mode decomposition. Energy Procedia. 17, 189–195 (2012)Lala, H., Karmakar, S.: Detection and experimental validation of high impedance arc fault in distribution system using empirical mode decomposition. IEEE Syst. J. 14(3), 3494–3505 (2020)Dedović, M.M., Avdaković, S.: A new approach for df/dt and active power imbalance in power system estimation using Huang's Empirical Mode decomposition. Int. J. Electr. Power Energy Syst. 110, 62–71 (2019)Biswal, S., Biswal, M.: Fault‐swing discrimination using Hilbert–Huang transform integrated discrete teager energy operator. IET Sci. Meas. Technol. 12(7), 829–837 (2018)Abidin, A.F., Mohamed, A., Shareef, H.: Power swing detection for correct distance relay operation using S‐transform and neural networks. In: IEEE 5th International Power Engineering and Optimization Conference, pp. 279–284. IEEE, Piscataway (2011)Abidin, A.F., Mohamed, A., Shareef, H.: Intelligent detection of unstable power swing for correct distance relay operation using S‐transform and neural networks. Expert Syst. Appl. 38(12), 14969–14975 (2011)Mohamad, N.Z., Abidin, A.F., Munim, W.N.W.A.: A new tracking method of symmetrical fault during Power Swing Based on S‐Transform. In: IEEE International Power Engineering and Optimization Conference, pp. 141–146. IEEE, Piscataway (2012)Mohamad, N.Z., Abidin, A.F., Munim, W.N.W.A., Yahya, M.: Symmetrical fault detection technique during power swing based on S‐Transform. In: IEEE Student Conference on Research and Development (SCOReD), pp. 7–10. IEEE, Piscataway (2012)Mohamad, N.Z., Abidin, A.F.: S‐Transform based technique to detect high resistance symmetrical fault during power swing. In: IEEE 8th International Colloquium on Signal Processing and its Applications, pp. 282–286. IEEE, Piscataway (2012)Cho, K., Kang, Y., Kim, S., Park, J., Kang, S., Kim, K.: An ANN based approach to improve the speed of a differential equation based distance relaying algorithm. IEEE Trans. Power Delivery 14(2), 349–357 (1999)Vaidya, A., Venikar, P.A.: ANN based distance protection of long transmission lines by considering the effect of fault resistance. In: IEEE‐International Conference on Advances In Engineering, Science and Management (ICAESM‐2012), pp. 590–594. IEEE, Piscataway (2012)Feilat, E., Al‐Tallaq, K.: A new approach for distance protection using artificial neural network. In: IEEE 39th International Universities Power Engineering Conference, vol. 1, pp. 473–477. IEEE, Piscataway (2004)Esmaeilian, A., Astinfeshan, S.: A novel power swing detection algorithm using adaptive neuro fuzzy technique. In: Proceedings of the IEEE International Conference on Electrical Engineering and Informatics, pp. 1–6. IEEE, Piscataway (2011)Sanaye‐Pasand, M., Jafarian, P.: An adaptive decision logic to enhance distance protection of transmission lines. IEEE Trans. Power Delivery 26(4), 2134–2144 (2011)Zadeh, H.K., Li, Z.: A novel power swing blocking scheme using adaptive neuro‐fuzzy inference system. Electr. Power Syst. Res. 78(7), 1138–1146 (2008)Rao, J.G., Pradhan, A.K.: Power‐swing detection using moving window averaging of current signals. IEEE Trans. Power Delivery 30(1), 368–376 (2014)Vapnik, V.: The Nature of Statistical Learning Theory. Springer Science & Business Media, Dordrecht (1999)Kampeerawat, W., Buangam, W., Chusanapiputt, S.: New swing‐blocking methods for digital distance protection using Support Vector Machine. In: IEEE International Conference on Power System Technology, pp. 1–6. IEEE, Piscataway (2010)Kampeerawat, W., Buangam, W., Chusanapiputt, S.: The swing‐blocking methods for digital distance protection based on wavelet packet transform and support vector machine. In: IEEE Asia‐Pacific Power and Energy Engineering Conference, pp. 1–4. IEEE, Piscataway (2012)Jena, P., Gupta, R.K.: Detection of fault in transmission line during power swing using fuzzy c‐means clustering transmission line protection. In: IEEE International Conference on Information, Communication, Instrumentation and Control (ICICIC), pp. 1–4. IEEE, Piscataway (2017)Sriram, C., Somlal, J., Goud, B.S., Bajaj, M., Elnaggar, M.F., Kamel, S.: Improved deep neural network (IDNN) with SMO algorithm for enhancement of third zone distance relay under power swing condition. Mathematics 10(11), 1944 (2022)Khodaparast, J., Khederzadeh, M.: Three‐phase fault detection during power swing by transient monitor. IEEE Trans. Power Syst. 30(5), 2558–2565 (2014)Platas‐Garza, M.A., Platas‐Garza, J., de la O Serna, J.A.: Dynamic phasor and frequency estimates through maximally flat differentiators. IEEE Trans. Instrum. Meas. 59(7), 1803–1811 (2009)Mai, R., He, Z., Fu, L., Kirby, B., Bo, Z.: A dynamic synchrophasor estimation algorithm for online application. IEEE Trans. Power Delivery 25(2), 570–578 (2010)Serna, J.A.O.: Phasor estimation from phasorlets. IEEE Trans. Instrum. Meas. 54(1), 134–143 (2005)de la O Serna, J.A., Rodriguez‐Maldonado, J.: Instantaneous oscillating phasor estimates with Taylor ^ K ‐Kalman filters. IEEE Trans. Power Syst. 26(4), 2336–2344 (2011)De la O Serna, J.A., Rodríguez‐Maldonado, J.: Taylor–Kalman–Fourier filters for instantaneous oscillating phasor and harmonic estimates. IEEE Trans. Instrum. Meas. 61(4), 941–951 (2012)Munoz, A.T., d l O Serna, J.A.: Shanks' method for dynamic phasor estimation. IEEE Trans. Instrum. Meas. 57(4), 813–819 (2008)Kang, D., Gokaraju, R.: A new method for blocking third‐zone distance relays during stable power swings. IEEE Trans. Power Delivery 31(4), 1836–1843 (2016)Torres, S., et al.: Unblocking function of distance relay during power swing based on modal analysis. In: IEEE PES Transmission & Distribution Conference and Exposition‐Latin America (PES T&D‐LA), pp. 1–6. IEEE, Piscataway (2016)Salehimehr, S., Taheri, B., Razavi, F., Parpaei, M., Faghihlou, M.: A new power swing detection method based on chaos theory. Electrical Eng. 102(2), 663–681 (2020)Robert, M.: Simple mathematical models with complicated dynamics. Nature 261, 459–467 (1976)Khan, U.N., Yan, L.: Power swing phenomena and its detection and prevention. In: 7th EEEIC International Workshop on Environment and Electrical Engineering, pp. 69–72. IEEE, Piscataway (2008)Benmouyal, G., Hou, D., Tziouvaras, D.: Zero‐setting power‐swing blocking protection. In: 31st Annual Western Protective Relay Conference. pp. 19–21. Washington State University, Pullman, WA (2004)Seethalekshmi, K., Singh, S., Srivastava, S.: SVM based power swing identification scheme for distance relays. In: IEEE PES General Meeting, pp. 1–8. IEEE, Piscataway (2010)Prasad, C.D., Biswal, M.: Swarm evaluated threshold elimination approach for symmetrical fault detection during power swing. IETE J. Res. 1–15 (2021)Das, S., Panigrahi, B.K., Jaiswal, P.K.: Qualitative assessment of power swing for enhancing security of distance relay in a TCSC‐compensated line. IEEE Trans. Power Delivery 36(1), 223–234 (2020)Hingorani, N.G., Gyugyi, L., El‐Hawary, M.: Understanding FACTS: Concepts and Technology of Flexible AC Transmission Systems. IEEE Press, New York (2000)Khodaparast, J., Khederzadeh, M., Faria da Silva, F., Back, C.L.: Performance of power swing blocking methods in UPFC‐compensated line. Int. Trans. Electr. Energy Syst. 27(11), e2382 (2017)Moravej, Z., Pazoki, M., Khederzadeh, M.: Impact of UPFC on power swing characteristic and distance relay behavior. IEEE Trans. Power Delivery 29(1), 261–268 (2013)Shekhar, H., Kumar, J.: Recognition of fault section during power swing in the presence of TCSC environment. Arab J. Sci. Eng. 47, 14229–14247 (2022)Sai Kumar, M., Kumar, J., Mahanty, R.: Fault detection during power swing in a TCSC‐compensated transmission line based on Clark's transform and Teager–Kaiser energy operator. Iran. J. Sci. Technol. 45(1), 47–60 (2021)Shekhar, H., Kumar, J., Nayak, A.: A fault detection technique during power swing in a TCSC‐compensated line using teager kaiser energy operator. IETE J. Res. 1–21 (2020)Nayak, P.K., Pradhan, A.K., Bajpai, P.: A fault detection technique for the series‐compensated line during power swing. IEEE Trans. Power Delivery 28(2), 714–722 (2013)Nayak, P.K., Pradhan, A.K., Bajpai, P.: Wide‐area measurement‐based backup protection for power network with series compensation. IEEE Trans. Power Delivery 29(4), 1970–1977 (2014)Alizadeh, B.A.M., Khederzadeh, M., Razzaghi, R.: Fault detection during power swing in thyristor‐controlled series capacitor‐compensated transmission lines. Electr. Power Syst. Res. 187, 106481 (2020)Juttu, T.R., Bhalja, B.R., Andreev, M., Malik, O.P.: Synchrophasor assisted power swing detection scheme for wind integrated transmission network. IEEE Trans. Power Delivery 37(3), 1952–1962 (2021)Niknezhad, M., Sadeh, J.: Analysis of wind farm penetration on power swing detection in distance relays. IET Renew. Power Gener. 16(2), 375–388 (2022)Tekdemir, I.G., Alboyaci, B.: A novel approach for improvement of power swing blocking and deblocking functions in distance relays. IEEE Trans. Power Delivery 32(4), 1986–1994 (2016)Musaruddin, M., Zaporoshenko, M., Zivanovic, R.: Remote protective relay testing. In: IEEE Australasian Universities Power Engineering Conference, pp. 1–4. IEEE, Piscataway (2008)Vebko Amirkabir Co.: https://vebko.org/en/Default.aspx Accessed 12 June 2018Verzosa, Q.: Realistic testing of power swing blocking and out‐of‐step tripping functions. In: IEEE 66th Annual Conference for Protective Relay Engineers, pp. 420–449. IEEE, Piscataway (2013) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png IET Generation Transmission & Distribution Wiley

An overview of power swing detection methods in distance relays and the factors involved

Loading next page...
 
/lp/wiley/an-overview-of-power-swing-detection-methods-in-distance-relays-and-OvlAxmFCim

References (62)

Publisher
Wiley
Copyright
© 2023 The Institution of Engineering and Technology.
eISSN
1751-8695
DOI
10.1049/gtd2.12711
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTIONIn current power systems, power transmission lines are considered a desirable conductor for transporting electrical energy. These lines transport electrical energy to remote locations with low power losses to reach the consumers through distribution networks [1]. Power transmission lines are one of the most essential components of a power system. In order to provide continuous and stable electrical energy to different consumers, it is necessary to quickly detect faults in power transmission lines using a suitable protection system. A proper protection system with timely and precise operation can reduce damages to the network equipment and even humans [2]. Among the various protection relays, distance relays are an integral part of protecting power systems. These relays have desirable features such as simplicity, use of local voltage and current for detecting a fault and its location [3]. The features of the distance relay make it suitable for protecting power transmission lines [4, 5]. However, a significant challenge that may affect the proper operation of distance relays is the occurrence of the power swing (PS) in power transmission lines [6, 7].PS is a transient phenomenon that occurs for a variety of reasons, including faults, line switching, and connecting or disconnecting of large blocks of loads in power systems [8]. PS in power systems can cause maloperation of distance relays due to entering the measuring impedance of the distance relay into the protection zones of this relay [9]. Undesirable transmission line tripping due to maloperation of distance relay against PS may destroy the power system stability and result in cascading outages of power transmission lines, resulting in global blackouts in the power system [10]. To prevent the maloperation of the distance relay during a PS, a power swing blocking (PSB) function is incorporated in distance relays [11, 12].PS detection can be challenging when the number of distributed generations like wind turbines is increased in the network. In fact, wind turbines can significantly change the PS characteristics and challenge power swing detection methods [13]. In addition, the presence of a thyristor‐controlled series capacitor (TCSC) in power systems can pose a fundamental problem to PS detection methods due to harmonic injection into voltage and current signals [14].Different researches have been presented to provide a method for detecting PS. Therefore, the main purpose of this paper is a comprehensive review of methods for detecting PS and the factors involved. Accordingly, the reviewed items include the types of PS, its detection methods, and different factors involved, such as affecting compensated lines and the presence of wind turbines on PS detection, determining threshold values ​in PS detection methods, and analyzing methods for proper testing of PSB function. Finally, suggestions for further research in this area are presented. It should be noted that there are some studies in this area that have already reviewed PS detection methods. Authors in [15] have investigated methods for detecting PS. However, this paper does not provide a comprehensive overview of the various methods and challenges of PS. In [16], methods for detecting PS have also been investigated. However, the effects of compensated lines and wind farms as well as PSB function testing method have not been investigated in this study.The present paper is structured as follows. In Section 2, different types of PS and the reasons for their occurrence are analysed. In Section 3, different types of PS detection methods are analysed. In Section 4, the proposed algorithms for determining the threshold values in the PS detection methods are analysed. In Section 5, the effects of compensated lines on PS parameters and PS detection methods are investigated. In Section 6, the effects of wind farms on PS are investigated. In Section 7, different methods for correct testing of the PSB function are analysed. Finally, in Section 8 additional suggestions for future researches are discussed. Conclusion is given in Section 9.DIFFERENT TYPES OF PSIn general, PS can be divided into three categories include stable, unstable, and multi‐mode power swing [11, 17]. These categories are explained in this section.Power swing equationIn order to express the power swing equations, consider Figure 1 [18, 19]. In this figure Eg${E_g}$represents a synchronous generator. This synchronous generator is connected to an infinite bus (Eb${E_b}$) via Xg${X_g}$and Xb${X_b}$ impedances. The electric power (Pe${P_e}$) which is transformed between the synchronous generator and the infinite bus can be calculated using Equation (1) [18, 19].1Pe=Eg.EbXg+Xesinδ,$$\begin{equation}{P_e} = \frac{{{E_g}.{E_b}}}{{{X_g} + {X_e}}}\sin \delta ,\end{equation}$$whereδ is the rotor angle.1FIGURESample transmission systemThe maximum transformed power (Pmax${P_{\max }}$) between the synchronous generator and the infinite bus can also be calculated from Equation (2) [19].2Pmax=Eg.EbXg+Xe.$$\begin{equation}{P_{\max }} = \frac{{{E_g}.{E_b}}}{{{X_g} + {X_e}}}.\end{equation}$$Figure 2 shows the power angle curve of a synchronous generator [19]. It is assumed that the synchronous generator transforms Pe${P_e}$ into the infinite bus. By creating a disturbance in the power system (such as connecting a load), the required power of the network suddenly increases and consequently, mechanical power (Pm${P_m}$) of the generator increases. This sudden increase in the mechanical power causes the operating point of the synchronous generator to be transferred toPm1${P_{m1}}$. Therefore, due to the inequality between electrical power and mechanical power, generator control systems operate and reduce mechanical power. In this reduction, the mechanical power is adjusted at the Pm2${P_{m2}}$point instead ofPe${P_e}$. At this stage, the generator control systems increasePm${P_m}$. This cycle continues until the mechanical power of the network balances with electrical power. Therefore, the rotor angle oscillation equation will be in accordance with Equation (3) [18–20].3Md2δdt2+Ddδdt=Pm−Pmaxsinδ,$$\begin{equation}M\frac{{{d^2}\delta }}{{d{t^2}}} + D\frac{{d\delta }}{{dt}} = {P_m} - {P_{\max }}\sin \delta ,\end{equation}$$where M is the inertia constant and D is the damping coefficient.2FIGUREP−δ$P - \delta \;$curve of the synchronous generatorStable PSStable PS typically occurs due to network disturbance, such as disconnecting a parallel line or reclosing lines in the power system [21]. Stable PS is the simplest and least intense type of PS that can occur in power systems. Figure 3 shows the changes in δduring this kind of PS [21]. Accordingly, the rotor angle has never been greater thanPmax${P_{\max }}$ (90 degrees) [18, 19]. This figure clearly shows that the rotor angle has stabilized after a short time, and therefore, this type of PS is called stable PS [19].3FIGUREChanges in the rotor angle of a synchronous generator during a stable PSUnstable PSUnstable PS is usually caused by a major power system disturbance. Compared to the stable type, the rotor angle in this type of PS will exceed Pmax${P_{\max }}$ (90 degrees), which makes the synchronous generator unstable [18, 19]. This instability of the synchronous generator is the reason why this PS is called unstable PS. As shown in Figure 4, the generator does not stabilize over time [19]. To eliminate this type of PS, the generator protection relays must detect the instability and disconnect the generator from the grid. In contrast, distance relays must maintain healthy transmission lines until instability in the network is eliminated and prevent further disturbances [19].4FIGUREChanges in the rotor angle of a synchronous generator during unstable PSMulti‐mode PSAs modern power systems become more sophisticated and more synchronous generators are implemented in networks with less distance, the probability of multi‐mode PS occurrence is significantly increased [11]. Multi‐mode PS occurs when two or more synchronous generators are involved in an oscillation [11]. In this case, the frequency of the PS is not constant and varies in each cycle. In addition, the current signal loses its sinusoidal waveform [22]. The occurrence of multi‐mode PS can pose serious challenges to existing methods of detecting PS [21].Figure 5 shows the changes in the rotor angle of a synchronous generator during a multi‐mode PS [21]. As can be seen from this figure, when the multi‐mode PS occurs, similar to the stable PS, the rotor angle of the synchronous generator does not exceed 90 degrees and the generator is not practically unstable [21].5FIGUREChanges in the rotor angle of a synchronous generator during a multi‐mode PSDETECTION OF POWER SWINGDetection of PS in power systems is very crucial. Failure to detect PS or simultaneous faults that occur during PS can cause instability in the network or even global blackout [23]. According to [16], a lack of accurate and timely diagnosis of PS caused various blackouts in Malaysia, Italy, Brazil, and India between 2003 and 2012. For this reason, there is a growing interest in research on the phenomenon of PS and its detection. It is worth mentioning that only the detection of PS and preventing the maloperation of the distance relay is not enough. PS detection algorithms must be able to detect a simultaneous fault (especially symmetric three‐phase fault) with PS very quickly [24]. In the following, the most important methods for detecting PS and its simultaneous faults are analysed.Impedance change methodImpedance change methods are one of the oldest methods for detecting PS. The basis of these methods is based on the impedance changes (ΔZ$\Delta Z$) calculated by the distance relay. Some of these methods also use an external zone with an impedance of 1Ω in relays with a 5 A current transformer (CT) or a 5 Ω impedance in relays with a 1 A CT [25]. In impedance‐based methods, an appropriate threshold value should be chosen in order to have a proper operation. This threshold value is determined according to the duration of the entrance and exit of impedance from the external zone. If the measured impedance remained in the external zone less than the threshold time, the relay would consider it as a fault. If this value exceeds the threshold time, the relay detects the PS and blocks the relay operation [26–28]. References [12, 29, 30] use the rate of impedance changes presented in reference [31] to detect PS. In this method, the three conditions of monotony, continuity, and smoothness are continuously checked, and in case of violation of these conditions, a fault is detected.Impedance change methods are generally reliable during slow PS (the frequency is less than 1 Hz) when impedance changes are very slow [32]. In contrast, during a fast PS (the frequency is more than 5 Hz), distance relays fail to block their operation and consider the PS as a fault [27, 28, 33‐35]. Furthermore, according to [17, 36], these methods cannot detect multi‐mode PS. In addition, the fault detection time during faults that simultaneously occurred with the PS in these methods is high [17, 36].Double blinder methodDouble blinder methods are generally developed to tackle the problems of impedance methods [37]. Since, during PS, the impedance path entered into the relay zones is approximately perpendicular to the line impedance, blinders are placed in parallel with the line impedance (Figure 6) [16]. Blinders methods perform much better than impedance methods [37–41]. The method of double blender is very similar to the impedance changes approach, with the difference that in the impedance method, an external zone is used to detect PS, however, in the blinder method two parallel blinders are used. If the impedance curve passes through blinder number 1, the counter will start, and at the time of passing through blinder number 2 it will stop. Afterward, acquired time is compared with the threshold value, and if it is more than the threshold value, the PS would be detected; otherwise, this condition will be considered as a fault. The most important problem with the double blinder method is the need for complex analysis and calculations for determining the threshold value. In addition, this method may cause relay maloperation during stable PS [16, 38].6FIGUREDouble blinder method operation in detecting PSDecreased resistance methodIn order to tackle the problems of the double blinder method, the decreased resistance method has been proposed. The principles of this method are based on continuous changes in impedance resistance (during PS the impedance resistance changes continuously but during a fault the resistance is constant) [26, 28]. The most important advantage of this method is the ability to distinguish between fault and PS. In addition, this method does not require extensive system analysis like the double blinder method [28]. On the other hand, when the network experiences a PS with low frequency, this method may detect PS as a fault at a power angle of 180 degrees which has significant changes in the impedance resistance [11]. This method will also fail to detect a three‐phase fault during slow PS [11].Swing centre voltage methodSwing centre voltage methods operate based on Vcosφ${V_{\cos \varphi }}$ variations. V is the amount of voltage at the relay location and ϕ is the angle between voltage and current measured at the relay location [39]. During a PS Vcosφ${V_{\cos \varphi }}$ continuously changes, but it remains almost constant during a fault. This method uses this criterion to distinguish between fault and PS. The most important advantage of this method is its independence from source and line impedance, which simplifies analysis. However, three‐phase fault detection in this method takes a long delay [26, 38]. This delay can cause the backup relays to operate, resulting in miscoordination. This method also needs to determine an appropriate threshold value for proper operation [16]. The threshold value for PS at a power angle close to 180° should be very low [38]. For this purpose, several threshold settings must be obtained for different conditions. This will complicate the operation of the method since it is very difficult to determine the network conditions to identify the threshold value. This method also has a problem with the detection of a high impedance fault occurrence so that detects the fault as a PS [26].Difference between three‐phase signalsIn recent years, researchers' interest in using the variation rate of three‐phase signals as a method for detecting PS has increased. The main basis of these methods is to study the variation rate of a signal obtained from the current and compare it for all three phases. Equation (4) shows the condition for detecting PS in these methods [24].where, di$di$ is the variation rate of the signal obtained from the current of each phase.4ifdiphase(a)≈diphase(b)≈diphase(b)⇒PSB=True.$$\begin{equation}if{\rm{ }}d{i^{phase(a)}} \approx d{i^{phase(b)}} \approx d{i^{phase(b)}} \Rightarrow PSB = True.\end{equation}$$Reference [24] uses the variation rate of the root mean square (RMS) of the current to distinguish PS from the fault. This method has the ability to detect various types of faults along with PS, especially high impedance faults [3]. Among the shortcomings of this method are the high fault detection time and non‐detection of asymmetric PS (PS occurs in one phase) [3]. Reference [42] uses Hilbert Transform to reduce fault detection time. This method can detect PS with a phase angle of 180 degrees. In addition, it has the ability to detect multi‐mode PS. This method is noise resistant and operates properly. Although this method has succeeded in reducing the fault detection time to 30 ms. However, this fault detection time is still high [11]. Authors in [10] compared the rate of instantaneous frequency variations to distinguish between fault and PS. This method can detect stable and unstable PS with a variety of faults that are simultaneous with PS. Among the most critical shortcomings of this method are high impedance fault detection time, non‐detection of multi‐mode PS, and inappropriate operation when the current signal is noisy [21].In [43], a method based on three‐phase power differences is presented to differentiate between PS and fault. During PS, the active and reactive powers of the three phases are constantly changing. At the time of a fault, the normalized values of active and reactive power are close to zero. One of the most important advantages of this method is its high sensitivity in detecting PS. Also, this method is not affected by the fault start time and other system parameters. Disadvantages of this method include the possibility of incorrect operation during a stable PS and correct operation only for symmetric faults.Signal processing methodsThese methods use various signal processing techniques to analyse the fault and PS characteristics. The most important signal processing technique that has received much attention over the years is the Fast Fourier Transform (FFT). Accordingly, [44] uses the value of the DC component obtained from FFT for differentiation PS from fault. Authors in [35] present a method based on the three‐phase active power frequency component using FFT approach [35, 44‐46]. Reference [36] provides a method based on creating a reference signal from a three‐phase current and obtaining the main component of the signal using FFT. This method has the ability to detect different types of PS and their simultaneous faults. In addition, low operating time is the most essential feature of this method [23]. It should be noted that all the methods reviewed in this section require a threshold value for proper performance [16, 35, 44].Since the frequency of a system varies around the nominal frequency during PS, the application of the wavelet transform (WT) to detect power swing has received a growing interest in recent years [47–49]. However, these conditions are not met during a permanent fault [47, 49‐53]. Accordingly, in [50] different sampling frequencies have been used to detect PS and faults. Authors in [49] present a high‐speed WT algorithm for detecting symmetric faults during PS. Reference [52] has used wavelet transform on impedance variations to distinguish between PS and fault. In [53], WT is used to analyse current changes to detect high impedance faults which occur simultaneously with PS. Authors in [54] have proposed a WT‐based high‐velocity PS detection method. Reference [55] also presents an approach based on a combination of improved discrete wavelet transform (IM‐DWT) and deep learning. It should be noted that despite the significant advantages of WT, the most critical drawback of these methods is their dependency on high sampling rates to achieve appropriate accuracy [21].In many studies, Prony method is used in power system protection [56, 57]. In [58], Prony has been proposed as a more appropriate method than FFT for detecting PS. In [57], an algorithm using the current waveform is proposed to detect a three‐phase fault during PS. In this study, only low impedance faults have been discussed. In [56], an algorithm based on the DC component of the current waveform is proposed.Various applications for empirical mode decomposition (EMD) in power systems have been proposed [59–61]. Accordingly, [62] presents a three‐step EMD‐based algorithm for differentiating PS from fault. In the first step, the intrinsic mode functions (IMF) values ​of the signal are extracted using EMD. In the second step, the instant amplitude and frequency of the first IMF are calculated using Hilbert–Huang transform (HHT). Finally, Teager energy is estimated using the instant amplitude and frequency. This method lacks the ability to detect multi‐mode PS. In addition, it may cause maloperation during high impedance fault [3]. Authors in [3] present an algorithm based on the rate of changes of the third IMF in each phase. In this study, first, an appropriate type of IMF for fault detection is defined according to the topology of the system. Then, a threshold value is obtained using different computer simulations. Whenever the IMF value exceeds the threshold value, the algorithm considers it as a fault. The most important drawback of the proposed method in [3] is its problem in detecting PS in different networks and its requirement for complex analysis to find the appropriate IMF [23].In [63, 64], S‐transform based method is proposed to differentiate fault from stable and unstable PS. In [65], active and reactive power is used as the input of the relay. In this study, it has been shown that during a fault, the S‐transform obtained from the active and reactive power is a value greater than zero, while in the event of a PS, this value is zero. Reference [66] uses the S‐transform to analyse voltage and current signals. However, the most important shortcoming of such methods is the lack of detection of high impedance faults, which are simultaneous with PS [67].Artificial intelligence methodWith the advancement of technology and the growing interest in artificial intelligence in various fields, researchers began to use artificial intelligence to protect power systems [68–70]. One of the most widely used of artificial intelligence methods is the adaptive neuro fuzzy inference system (ANFIS) method. ANFIS is an advanced version of the fuzzy logic system with generalizability, noise protection, reliability, and high fault tolerance [71–73]. Reference [73] uses positive sequence impedance and positive and negative current sequence as the input signals to ANFIS. This method has the ability to detect a variety of power swings and faults. Similarly, in [71], the rate of current variations and normalized active and reactive powers are used as ANFIS inputs. This method is not affected by system parameters including fault time, fault position, and pre‐loading conditions. However, the most important shortcoming of these kind of methods is the need for many simulations to train different fault types and power swings [74]. Reference [23] used GMDH neural network to differentiate PS and fault. In this reference, GMDH is trained to estimate the signal in the normal situation and during PS. During different faults, there is a significant difference between the estimated value and the actual value of the signal, which is implemented to detect different power swings and faults.In [75], the performance of the support vector machine (SVM) is considered to be better than the artificial neural network. Accordingly, authors in [76] have proposed a new method for differentiating fault from PS using the minimum square of the SVM model. This method uses the current and voltage signals at the relay location as the inputs. This method can differentiate fault from PS and also has the ability to detect faults that are simultaneous with PS. In [77], a WT‐based approach and SVM is proposed. This method can differentiate PS from faults and has the ability to detect faults that are simultaneous with PS.Reference [78] used the C‐MEAN classification method to detect faults which are simultaneous with PS. C‐MEAN is a fuzzy clustering method for automatic data classification. In contrast to ANFIS and SVM methods, this method is of the unsupervised type and does not require training with various fault types for performance. Authors in [79] have proposed a new method for differentiating fault from PS using the improved deep neural network (IDNN) with spider monkey optimization (SMO) algorithm.Other methodsReference [74] provides a method based on the moving window averaging (MWA) to detect PS. MWA is a low‐pass filter applied to a distance relay. This method is less sensitive to noise, and it also has the ability to detect faults that are simultaneous with PS. However, this method shows a maloperation during multi‐mode PS [11, 21].Authors in [80] have used transient monitoring (TM) to detect PS. This method compares TM with a predetermined threshold value. If the calculated TM is greater than the threshold value, a fault is detected; otherwise, PS is detected. In this method, phasor estimation plays an important role in determining the threshold value. Although there are different methods for estimating the phasor [81–86], this method uses dynamic phase estimation to improve the performance of the proposed method.Reference [87] provides a method based on the concept of the first zero crossing signal to detect PS. This method can block the third zone of the distance relay during unstable PS and prevent the relay maloperation. This method is not dependent on the network topology and, unlike the conventional blinder technique, does not require lots of stability studies to find the relay settings. An algorithm based on participatory factors acquired from the variance/covariance matrix is ​proposed in [88]. In this method, modal analysis is applied to obtain more properties for the transient response to the matrix, and a threshold for the difference between fault and PS is identified.In [17], a method based on the rate of current variance variation is proposed. In this reference, the variance of the sampled data from the current signal is calculated during a signal window, and PS and fault are differentiated by setting a threshold value. This method has the ability to detect different PS types, including stable, unstable, and multi‐mode. In addition, changes in angle and impedance of the fault do not affect the proposed method. Reference [89] has proposed a method for detecting PS based on chaos theory. The proposed method in this study is based on the logistic map, which is a simple mathematical model for chaos theory (according to the model presented in [90]). The method presented in [89] has the ability to detect different types of PS, including asymmetric types. In addition, this method has a precise operation when the network experiences a capacitor bank switching. Authors in [21] have proposed a new approach based on the compressed sensing theory (CST). In this approach, CST is implemented to distinguish PS and fault. This method does not require a high sampling rate and has a high fault detection speed. The proposed method also has the ability to detect various types of PS and faults and operates well on compensated lines. A Phaselet based method is proposed in [11], which uses Phaselet to estimate the current data and compares it with the actual data. A fault is detected if the difference between the estimated data and the real data exceeds a certain value. Researchers in [9] presented a method based on the signal energy variation rate to distinguish between PS and fault.Table 1 provides an overview of the available methods for detecting PS based on the studies analysed in this section.1TABLEComparison between different power swing detection methodsPS detectionDetecting faults simultaneously with PSMethodReferencesStableUnstableMulti‐modeSingle phaseTwo phasesThree phasesThreshold value requirementImpedance changes[12, 29, 30]✓✓×✓✓✓✓Double blinder[91]✓✓×✓✓×✓Decreased resistance[28]✓✓×✓✓✓✓Swing centre voltage[92]✓✓×✓✓✓×RMS[24]✓✓✓✓✓✓✓Hilbert transform[42]✓✓✓✓✓✓✓Instantaneous frequency[10]✓✓×✓✓✓✓Three phases power variation[43]✓✓×××✓×FFT[35, 44]✓✓×✓✓✓✓[36]✓✓✓✓✓✓✓WT[49, 50, 52]✓✓×✓✓✓✓Prony[57, 58]✓✓×✓✓✓✓EMD[62]✓✓×✓✓✓✓[3]✓✓✓✓✓✓✓S transform[63, 64]✓✓×✓✓✓✓ANFIS[71, 73]✓✓✓✓✓✓✓SVM[76, 77, 93]✓✓✓✓✓✓✓C‐MEAN[78]✓✓×✓✓✓×GMDH[23]✓✓✓✓✓✓✓MWA[74]✓✓×✓✓✓✓TM[80]✓✓×✓✓✓✓Variance[17]✓✓✓✓✓✓✓Chaos[89]✓✓✓✓✓✓✓CST[21]✓✓✓✓✓✓✓Phaselet[11]✓✓✓✓✓✓✓METHODS TO IDENTIFY THRESHOLD VALUE FOR PS DETECTIONBeing evidently presented in Table 1, most PS detection methods require optimal threshold values to differentiate PS from fault. This proves the importance of determining the correct threshold value. Studies have proposed different methods for determining threshold values. These researches are presented in the following section.Simulating different PS and fault ratesReferences [3, 21, 45, 89] use a threshold identification method based on numerical simulations of the different fault and PS rates. Therefore, to determine the threshold value for a network, different types of PS and faults are simulated based on the items mentioned in Table 2 [21]. Table 2 shows the used and unused items in each case with ✓ and ×, respectively. In the next step, the maximum points (peak value of the signal) obtained for each case in Table 2 are calculated. Finally, the threshold value is determined according to the recorded values for different types of PS and faults [21].2TABLESimulation cases to determine the threshold valuePower swingFaultDistance fault (%)CaseStableUnstableMulti‐modeSingle‐phaseThree‐phaseHigh impedance570901✓××✓××✓✓✓2✓×××✓×✓✓✓3✓××××✓✓✓✓4×✓×✓××✓✓✓5×✓××✓×✓✓✓6×✓×××✓✓✓✓7××✓✓××✓✓✓8××✓×✓×✓✓✓9××✓××✓✓✓✓Although the threshold value determining method, which is presented in [17], is based on numerical simulation, it has some differences from the previously stated method. This method is based on the principle that in a power system the highest amplitude among the different types of power swings is related to the unstable power swing. Also, the lowest amplitude between the different types of faults is related to the high impedance fault. Therefore, to obtain the threshold value in the power system, it is only required to simulate these two cases. On this basis, this reference uses Equation (5) to determine the threshold value (K). As it is clear from this equation, the threshold value can be obtained only by obtaining the maximum and minimum peak values expressed in the two cases.5K=mean([max(UnstablePS)max(Highimpedancefault]).$$\begin{equation}K = mean([\max (Unstable{\rm{ }}PS){\rm{ }}\max (High{\rm{ }}impedance{\rm{ }}fault]).\end{equation}$$Adaptive method based on permissible faultUsing the numerical simulation method to define a suitable threshold value for different network types increases study time and therefore increases design costs [19]. It also increases human errors in designing protection systems and setting distance relays [19]. For this reason, in [19], a method to determine the adaptive threshold value based on the acceptable error of relays, is proposed. The threshold value is proportional to the tolerance value in most relays and is equal to 5% and 50 mA. The reason for using two values ​is that for currents with a small amplitude (for example 200 mA) the value of 5% of the current will be very small. Therefore, the algorithm selects 50 mA. In contrast, in networks with high current levels, the value of 50 mA is minimal, and the algorithm selects 5% as the threshold value. Figure 7 shows the algorithm for calculating the threshold value using the adaptive method [19].7FIGUREThreshold value algorithm by an adaptive methodSwarm intelligence methodIn [94], a new method for determining the threshold value is proposed based on swarm intelligence. In this method, the mechanism of fault parameters, PS, and the minimum values ​of detection indices are identified using particle swarm optimization (PSO) techniques. In this method, the optimal threshold value is determined by extensive case studies and using the PSO algorithm (examining 150,000 simulation cases in an iterative process). Figure 8 shows the threshold value detection algorithm by the swarm intelligence method [94].8FIGUREThreshold value algorithm by swarm intelligence methodEFFECTS OF COMPENSATED LINES ON PS DETECTION METHODSThe application of a flexible AC transmission system (FACTS) in transmission lines can have a significant impact on the impedance characteristic of PS and the performance of protection designs [95]. For this reason, this section examines the available research on PS detection in compensated lines.Impacts of a unified power flow controllerA unified power flow controller (UPFC) is a FACTS device used to control the current of transmission lines and to improve system stability [96, 97]. In transmission lines equipped with UPFC, ordinary distance relays are overreached or underreached during a fault [97]. For this reason, various studies have been performed to investigate the effect of FACTS compensator devices on distance relays during fault.The effect of UPFC on oscillating properties (radius and centre point changes) is investigated in [98]. According to this study, transmission line parameters change in the presence of UPFC. In [97], the impacts of UPFC on methods of the power variations rate, the main frequency of instantaneous three‐phase active power, the rate of impedance changes, and the transient monitor are investigated. In this reference, it is analytically proven that since the index value of the power variations rate method decreases during power swing, UPFC affects the performance of this method. On this basis, the presence of UPFC prevents the application of power variations rate method for PS detection [97]. Also, the studies performed in [97] show that the method's reliability of the main frequency estimation of three‐phase instantaneous active power decreases under compensated conditions. The transient monitor method is also influenced by the UPFC of the harmonic influence of multi‐level converters [97]. Therefore, it is clear that the performance of UPFC can have a negative effect on different PS detection methods and lead to incorrect performance of these methods.Effects of TCSCA thyristor‐controlled series capacitor (TCSC) is used to improve the transmission capabilities of the power line. However, due to the inversion of voltage and current as well as the nonlinear nature of the metal oxide varistor, many protection approaches are not applicable in the TCSC environment [99]. The presence of TCSC can have many effects on transmission line protection, including the impact on the calculated apparent impedance, the impact on the PS frequency, and also the impact on the phase angles measured in a bus [14, 100, 101]. Therefore, the application of TCSC in the transmission network increases the protection challenges. In addition, the operation of protection relays such as fault detection, PS detection, and detection of faulty section faces serious challenges [102–104]. For this reason, the presence of TCSC requires advanced protection techniques to detect all types of faults and power swings. Accordingly, in [101], the Teager Kaiser Energy Operators (TKEOs) of negative sequence technique of voltage and current have been used to detect PS in TCSC compensated lines. In [103], a wide‐area backup protection approach is proposed to protect TCSC compensated lines. In this method, negative and zero sequences of voltage values are implemented for detecting asymmetric faults, and positive sequence voltages are used for symmetric faults. This method may fail to detect simultaneous faults with PS [99]. The authors in [100] have used the Clark transform to detect PS in transmission lines equipped with TCSC. In [104], a high‐speed fault detection method is proposed based on mathematical morphology in TCSC‐compensated lines.THE EFFECT OF WIND FARMS ON PS DETECTION METHODSEnvironmental problems along with increasing energy demand, have drawn the attention to the application of wind turbines at different network levels [18, 105]. However, the existence of wind turbines in the network can significantly change the PS characteristics and pose major challenges for existing PS detection methods [106]. Maloperation of PS detection methods can increase with increasing wind production. The presence of a wind turbine can also change the PS impedance path and affect the performance of the out‐of‐step protection [13].To investigate the effects of the wind turbines on PS characteristics as well as their impact on several PS detection methods, the IEEE standard 39‐bus network has been simulated using DIgSILNET and MATLAB software. Figure 9 shows the single line diagram of this network. More information on this network is available in [11, 17]. In order to simulate wind farms, two wind farms are connected to buses 26 and 29. The protection relay is placed on bus number 26 to protect the line connecting bus 26 to bus 29.9FIGURESingle‐line diagram of IEEE 39‐bus networkInstalling a wind turbine in the network may affect the PS by changing the rotor angle of the synchronous generator. Figure 10 shows the rotor angle variations of synchronous generator No. 9, which is connected to bus 38 in the presence and absence of a wind turbine. As demonstrated in this figure, when the wind turbine is added to the network, stable PS becomes unstable.10FIGUREChanges in the rotor angle of a synchronous generatorInstalling a wind turbine on a network may also affect the impedance trajectory. Figure 11 shows the impedance trajectory in two network conditions, when the wind turbine is connected or disconnected from the network. As can be seen from this figure, when the wind turbine is connected, the impedance trajectory is very different. In addition, as presented in Figure 12, in the presence of a wind turbine, the current waveform may not be sinusoidal anymore and become very noisy.11FIGUREImpedance trajectory12FIGURECurrent waveform during PSIn the following, the effect of adding wind turbines to the network on four PS detection methods (including Taylor Series [107], Phaselet [11], chaos theory [89], and CST [21]) is investigated. The simulation results for each method are depicted in Figures 13–16, respectively. As it is clear from these figures, all the methods have exceeded the considered threshold value and cannot differentiate fault from PS.13FIGUREPerformance of the Taylor Series method14FIGUREPerformance of the Phaselet method15FIGUREPerformance of the Chaos method16FIGUREPerformance of the CST methodPS FUNCTION TEST METHODSProtection relays always require various tests during the development, commissioning, maintenance, configuration, and troubleshooting stages [108]. Similar to other algorithms, PS algorithms in protection relays require different tests. Accordingly, various test methods for the PS function are discussed in this section.Z‐Tracking methodZ‐tracking is a conventional and common method of PS function testing [109]. It provides the ability to change the impedance at any angle, speed, and in any direction of the distance relay zones. As Figure 17 presents, in this method, it is only required to specify the starting and ending points of entering the zone, the exit point and movement speed, and the amount of voltage and current. This method also has the ability to detect fault that is simultaneous with PS. However, since the PS waveforms are not accurate, the reliability of this method is low [109].17FIGUREPS testing via Z‐Tracking methodTransient simulation of PS in transient analysis softwareUnlike classic tests, network simulations in software such as EMTP, PSCAD, and DIGSILENT lead to waveforms that are very similar to the actual waveforms of the network. Accordingly, PS functions can be tested with high reliability via this method. However, these methods require a long analysis and testing time for execution [110]. Figure 18 depicts an example of the PS generated from transient analysis software. This type of testing has drawn the attraction of many researchers, including references [3, 11, 17, 42] in order to evaluate the performance of their proposed method.18FIGUREPS generated from simulation software in relay test softwareTesting via the information received from Comtrade fileDigital relays have the ability to record events of the network and give Comtrade file output. Since these outputs are actual waveforms from transmission lines, they are an appropriate reference for testing relays. This type of test provides the most realistic test possible for PS function testing. A big challenge with this type of test is the limited number of Comtrade files [110]. In addition, these files cannot be used to test different relays with different settings. Figure 19 shows the entering of the impedance seen by the distance relay into its protection zones during PS as a result of testing with a relay Comtrade file.19FIGUREEntering of the impedance seen by distance relay into its protection zones during PSDISCUSSIONDiscussing the existing research in the PS field proves the requirement for further studies in this field. Being previously mentioned, existing non‐intelligent methods have shortcomings such as inability to detect multi‐mode PS, inability to detect faults that are simultaneous with PS, failure to operate during high impedance faults, dependency on the high sampling rate, and maloperation when the signal is under a noisy condition. Although intelligent methods are able to solve the challenges of non‐intelligent methods, their most important challenge is the necessity for many offline simulations to train the algorithm for different faults and PS conditions.These simulations can significantly increase design duration and cost. In addition, intelligent methods require far more powerful hardware to perform. This can be simply demonstrated using the Python programming language and the psutil library. The psutil library has the ability to record the amount of CPU usage for each software using Process ID (PID). In this section, two different power swing detection algorithms (a non‐intelligent algorithm and an intelligent algorithm) are run separately by MATLAB software, and CPU consumption is recorded by code written in Python. Figure 20 shows CPU usage in each mode. According to this figure, non‐intelligent methods can be implemented on a conventional microprocessor. However intelligent methods need powerful hardware.20FIGURECPU usage in non‐intelligent and intelligent PS detection algorithmsOn the other hand, as previously shown, the presence of compensators on transmission lines and the connection of wind farms to these lines can significantly impact on protection algorithms. Although many researches have been carried out in this field, various challenges such as voltage and current reversal increase the requirement for further research in this field. Therefore, the following subjects are suggested for additional researches in the PS protection.Provide methods for detecting PS without the need to determine the threshold value.Reduce the required sampling rate.Introducing new methods to determine the appropriate threshold values for different methods.Provide unsupervised artificial intelligence methods to reduce the necessary simulations.Investigate the effect of other types of compensators on PS.Provide methods for detecting PS in lines connected to the wind turbines.In addition to the mentioned issues, detecting power swings in conventional electromechanical and electrostatic distance relays is still considered a fundamental challenge. In fact, these types of relays do not have the possibility of detecting power swings due to the lack of using microprocessors. It is also not possible to implement new methods of power swing detection methods in these types of relays. For this reason, a suggestion for future work in the field of power swing detection can be the design and construction of suitable hardware to use in these types of relays. This hardware must be in line with the operational performance of these types of relays and also have the ability to simply implement new methods.CONCLUSIONSThe present paper reviews different types of power swing and their detection methods. The focus of this study is on power swing types and various factors that can affect the methods of detecting power swings. As studies have shown, power swings can be divided into three general categories of stable, unstable, and multi‐mode power swings. As the result of this review shows, multi‐mode power swing is a serious challenge for power swing detection methods due to complexities such as non‐sinusoidal current waveform and variable frequency that may occur during this type of power swing. After introducing different types of power swing, the common methods for detection of power swing were investigated. Among the available methods, the methods using signal processing are more typical. However, the main challenge with these methods is the necessity of determining a threshold value for proper operation. On this basis, the methods of determining the threshold value were investigated in different studies. The effects of compensated lines and wind turbines on the power swing detection methods have been investigated. Despite the necessity of an actual test of the power swing detection function, this issue has been considered in a few studies. Various approaches to power swing test methods and studies conducted in this field were also analysed in this article.AUTHOR CONTRIBUTIONBehrooz Taheri (BT): Conceptualization, Methodology, Software, Visualization, Investigation. Seyed Amir Hosseini (SAH): Conceptualization, Data curation, Writing‐ review and editing, Supervision, Software. Sirus Salehimehr (SS): Writing‐ original draft, Resources, Validation.CONFLICT OF INTERESTThe authors declare no conflict of interest.FUNDING INFORMATIONNone.DATA AVAILABILITY STATEMENTData sharing is not applicable to this article as no new data were created or analysed in this study. cd_value_code=textREFERENCESJannati, M., Mohammadi, M.: A novel fast power swing blocking strategy for distance relay based on ADALINE and moving window averaging technique. IET Gener. Transm. Distrib. 15(1), 97–107 (2021)Doria‐García, J., Orozco‐Henao, C., Leborgne, R., Montoya, O.D., Gil‐González, W.: High impedance fault modeling and location for transmission line. Electr. Power Syst. Res. 196, 107202 (2021)Taheri, B., Hosseini, S.A., Askarian‐Abyaneh, H., Razavi, F.: Power swing detection and blocking of the third zone of distance relays by the combined use of empirical‐mode decomposition and Hilbert transform. IET Gener. Transm. Distrib. 14(6), 1062–1076 (2020)Taheri, B., Hosseini, S.A., Askarian‐Abyaneh, H., Razavi, F.: A new method for remote testing distance relay using internet protocol version 4. IEEE International Conference on Protection and Automation of Power System (IPAPS), pp. 31–37. IEEE, Piscataway (2019)Taheri, B., Hosseini, S.A., Sedighizadeh, M., Khatibi, M.: A moving window numerical distance protection based on flat‐top signal windowing. Arab J. Sci. Eng. 47, 14249–14266 (2022)Khodaparast, J., Khederzadeh, M., da Silva, F.F., Bak, C.L.: A novel approach to detect faults occurring during power swings by abrupt change of impedance trajectory. Electr. Power Comp. Syst. 45(15), 1638–1652 (2017)Moazzami, M., Hooshmand, R., Khodabakhshian, A., Yazdanpanah, M.: Blackout prevention in power system using flexible AC transmission system devices and combined corrective actions. Electr. Power Comp. Syst. 41(15), 1433–1455 (2013)Devi, R., et al.: Development of symmetrical fault detection during power swing based on entropy. J. Electr. Eng. Technol. 17, 1533–1542 (2022)Taheri, B., Salehimehr, S., Razavi, F.: A new method for fast power swing detection using the rate of change of energy in the current signal. In: IEEE 27th Iranian Conference on Electrical Engineering (ICEE), pp. 464–469. IEEE, Piscataway (2019)Taheri, B., Salehimehr, S., Razavi, F., Parpaei, M.: Detection of power swing and fault occurring simultaneously with power swing using instantaneous frequency. Energy Syst. 11(2), 491–514 (2020)Ghalesefidi, M.M., Ghaffarzadeh, N.: A new phaselet‐based method for detecting the power swing in order to prevent the malfunction of distance relays in transmission lines. Energy Syst. 12(2), 491–515 (2021)Ziegler, G.: Numerical Distance Protection: Principles and Applications. John Wiley & Sons, Boca Raton (2011)Haddadi, A., Kocar, I., Karaagac, U., Gras, H., Farantatos, E.: Impact of wind generation on power swing protection. IEEE Trans. Power Delivery 34(3), 1118–1128 (2019)Sai Kumar, M., Kumar, J., Mahanty, R.: Fault classification in a TCSC compensated transmission line during power swing using Wigner Ville transform. IETE J. Res. 1–22 (2021)Hida, S., Pradhan, V., Naidu, O., Cheriyan, E.P.: A critical review of distance relaying techniques under power swing condition. In: IEEE PES GTD Grand International Conference and Exposition Asia (GTD Asia), pp. 500–505. IEEE, Piscataway (2019)Gunasegaran, M., Tan, C., Bakar, A.H.A., Mokhlis, H., Illias, H.A.: Progress on power swing blocking schemes and the impact of renewable energy on power swing characteristics: A review. Renew. Sustain. Energy Rev. 52, 280–288 (2015)Salehimehr, S., Taheri, B., Faghihlou, M.: Detection of power swing and blocking the distance relay using the variance calculation of the current sampled data. Electr. Eng. 104, 913–927 (2022)Machowski, J., Lubosny, Z., Bialek, J.W., Bumby, J.R.: Power System Dynamics: Stability and Control. John Wiley & Sons, Hoboken, NJ (2020)Taheri, B., Faghihlou, M., Salehimehr, S., Razavi, F.: Symmetrical fault detection during power swing using mean value of sampled data from the current signal. IETE J. Res. 1–13 (2020)Kumar, J., Jena, P.: Fault detection during asymmetrical power swing using superimposed negative sequence current. Arab J. Sci. Eng. 44(8), 7033–7046 (2019)Taheri, B., Sedighizadeh, M.: Detection of power swing and prevention of mal‐operation of distance relay using compressed sensing theory. IET Gener. Transm. Distrib. 14(23), 5558–5570 (2020)Hashemi, S.M., Sanaye‐Pasand, M., Shahidehpour, M.: Fault detection during power swings using the properties of fundamental frequency phasors. IEEE Trans. Smart Grid. 10(2), 1385–1394 (2017)Hosseini, S.A., Taheri, B., Abyaneh, H.A., Razavi, F.: Comprehensive power swing detection by current signal modeling and prediction using the GMDH method. Protect. Control Modern Power Syst. 6(1), 1–11 (2021)Taheri, B., Razavi, F.: Power swing detection using rms current measurements. J. Electr. Eng. Technol. 13(5), 1831–1840 (2018)Blumschein, J., Yelgin, Y., Kereit, M.: Proper detection and treatment of power swing to reduce the risk of blackouts. In: Third International Conference on Electric Utility Deregulation and Restructuring and Power Technologies, pp. 2440–2446. IEEE, Piscataway (2008)Nayak, P.K., Rao, J.G., Kundu, P., Pradhan, A., Bajpai, P.: A comparative assessment of power swing detection techniques. In: IEEE Joint International Conference on Power Electronics, Drives and Energy Systems & 2010 Power India, pp. 1–4. IEEE, Piscataway (2010)Lin, X., Li, Z., Ke, S., Gao, Y.: Theoretical fundamentals and implementation of novel self‐adaptive distance protection resistant to power swings. IEEE Trans. Power Delivery 25(3), 1372–1383 (2010)Khoradshadi‐Zadeh, H.: Evaluation and performance comparison of power swing detection algorithms. In: IEEE Power Engineering Society General Meeting, pp. 1842–1848. IEEE, Piscataway (2005)SIPROTEC, S.: Distance Protection 7SA522 V4. 70, Instruction Manual (2011)Ganeshan, I.S., Jena, M.K., Panigrahi, B.K.: An improved setting‐free power swing blocking scheme using local end information. In: IEEMA Engineer Infinite Conference (eTechNxT), pp. 1–6. IEEE, Piscataway (2018)Jurisch, A., Schwenke, M.: Method of deriving a signal indicating an oscillation in an electric power supply system. Google Patents (2000)Corsi, S., Sabelli, C.: General blackout in Italy Sunday September 28, 2003, h. 03: 28: 00. In: IEEE Power Engineering Society General Meeting, pp. 1691–1702. IEEE, Piscataway (2004)Martuscello, L., Krizauskas, E., Holbach, J., Lu, Y.: Tests of distance relay performance on stable and unstable power swings reported using simulated data of the August 14 th 2003 system disturbance. In: IEEE Power Systems Conference, pp. 1–21. IEEE, Piscataway (2009)Mohamad, N.Z., Abidin, A.F., Musirin, I.: Adaptive distance protection to prevent false relay tripping during power swing. Appl. Mech. Mater. 785, 358–362 (2015)Mahamedi, B.: A very fast unblocking scheme for distance protection to detect symmetrical faults during power swings. In: IEEE Conference Proceedings IPEC, pp. 378–383. IEEE, Piscataway (2010)Taheri, B., Faghihlou, M., Salehimehr, S., Razavi, F.: A fast Fourier transform‐based method for power swing detection and distance relay malfunction prevention. J. Control Autom. Electr. Syst. 31(6), 1458–1468 (2020)Zadeh, H.K., Li, Z.: Artificial neural network based load blinder for distance protection. In: IEEE Power and Energy Society General Meeting‐Conversion and Delivery of Electrical Energy in the 21st Century, pp. 1–6. IEEE, Piscataway (2008)Ambekar, V.A., Dambhare, S.S.: Comparative evaluation of out of step detection schemes for distance relays. In: IEEE Fifth Power India Conference, pp. 1–6. IEEE, Piscataway (2012)Fischer, N., Benmouyal, G., Hou, D., Tziouvaras, D., Byrne‐Finley, J., Smyth, B.: Do system impedances really affect power swings—Applying power swing protection elements without complex system studies. In: IEEE 65th Annual Conference for Protective Relay Engineers, pp. 108–119. IEEE, Piscataway (2012)Holbach, J.: New out of step blocking algorithm for detecting fast power swing frequencies. In: IEEE Power Systems Conference: Advanced Metering, Protection, Control, Communication, and Distributed Resources, pp. 182–199. IEEE, Piscataway (2006)Tziouvaras, D.A., Hou, D.: Out‐of‐step protection fundamentals and advancements. In: IEEE 57th Annual Conference for Protective Relay Engineers, pp. 282–307. IEEE, Piscataway (2004)Salehimehr, S., Taheri, B., Hosseini, S.A., Abyaneh, H.A., Razavi, F.: A new power swing detection method based on Hilbert Transform. Int. J. Indus. Electron. Control Optim. 3(2), 103–114 (2020)Lin, X., Gao, Y., Liu, P.: A novel scheme to identify symmetrical faults occurring during power swings. IEEE Trans. Power Delivery 23(1), 73–78 (2007)Karegar, H.K., Mohamedi, B.: A new method for fault detection during power swing in distance protection. In: IEEE 6th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, vol. 1, pp. 230–233. IEEE, Piscataway (2009)Mahamedi, B., Zhu, J.G.: A novel approach to detect symmetrical faults occurring during power swings by using frequency components of instantaneous three‐phase active power. IEEE Trans. Power Delivery 27(3), 1368–1376 (2012)D'Apuzzo, M., D'Arco, M.: A time‐domain approach for the analysis of nonstationary signals in power systems. IEEE Trans. Instrum. Meas. 57(9), 1969–1977 (2008)Fu, L., He, Z., Bo, Z.: Wavelet transform and approximate entropy based identification of faults in power swings. In: 2008 IET 9th International Conference on Developments in Power System Protection (DPSP 2008), pp. 590–594. IET, Stevenage (2008)Parameswariah, C., Cox, M.: Frequency characteristics of wavelets. IEEE Trans. Power Delivery 17(3), 800–804 (2002)Pang, C., Kezunovic, M.: Fast distance relay scheme for detecting symmetrical fault during power swing. IEEE Trans. Power Delivery 25(4), 2205–2212 (2010)Brahma, S.: Use of wavelets for out of step blocking function of distance relays. In: IEEE Power Engineering Society General Meeting, p. 5. IEEE, Piscataway (2006)Dubey, R., Samantaray, S.R.: Wavelet singular entropy‐based symmetrical fault‐detection and out‐of‐step protection during power swing. IET Gener. Transm. Distrib. 7(10), 1123–1134 (2013)Mahamedi, B.: A new power swing blocking function based on wavelet transform. In: IEEE 2nd International Conference on Electric Power and Energy Conversion Systems (EPECS), pp. 1–6. IEEE, Piscataway (2011)Dubey, R., Samantaray, S., Tripathy, A., Babu, B.C., Ehtesham, M.: Wavelet based energy function for symmetrical fault detection during power swing. In: IEEE Students Conference on Engineering and Systems, pp. 1–6. IEEE, Piscataway (2012)Alsyoufi, Y.R., Hajjar, A.A.: A high‐speed algorithm to discriminate between power swing and faults in distance relays based on a fast wavelet. Electr. Power Syst. Res. 172, 269–276 (2019)Sriram, C., Somlal, J.: IM‐DWT with DNN based blocking scheme of third zone distance relay in power swing condition. Smart Sci. 1–20 (2022)Thakallapelli, A., Mehra, R., Mangalvedekar, H.: Differentiation of faults from power swings and detection of high impedance faults by distance relays. In: IEEE 1st International Conference on Condition Assessment Techniques in Electrical Systems (CATCON), pp. 374–377. IEEE, Piscataway (2013)Lotfifard, S., Faiz, J., Kezunovic, M.: Detection of symmetrical faults by distance relays during power swings. IEEE Trans. Power Delivery 25(1), 81–87 (2009)Leonowicz, Z.: Application of the Prony method for compensation of errors in distance relays. In: IEEE 12th International Conference on Environment and Electrical Engineering, pp. 568–572. IEEE, Piscataway (2013)Wang, Y., Wang, J., Meng, Q., Yu, H.: Power system osciliation modes identification based on eigen system realization algorithm via empirical mode decomposition. Energy Procedia. 17, 189–195 (2012)Lala, H., Karmakar, S.: Detection and experimental validation of high impedance arc fault in distribution system using empirical mode decomposition. IEEE Syst. J. 14(3), 3494–3505 (2020)Dedović, M.M., Avdaković, S.: A new approach for df/dt and active power imbalance in power system estimation using Huang's Empirical Mode decomposition. Int. J. Electr. Power Energy Syst. 110, 62–71 (2019)Biswal, S., Biswal, M.: Fault‐swing discrimination using Hilbert–Huang transform integrated discrete teager energy operator. IET Sci. Meas. Technol. 12(7), 829–837 (2018)Abidin, A.F., Mohamed, A., Shareef, H.: Power swing detection for correct distance relay operation using S‐transform and neural networks. In: IEEE 5th International Power Engineering and Optimization Conference, pp. 279–284. IEEE, Piscataway (2011)Abidin, A.F., Mohamed, A., Shareef, H.: Intelligent detection of unstable power swing for correct distance relay operation using S‐transform and neural networks. Expert Syst. Appl. 38(12), 14969–14975 (2011)Mohamad, N.Z., Abidin, A.F., Munim, W.N.W.A.: A new tracking method of symmetrical fault during Power Swing Based on S‐Transform. In: IEEE International Power Engineering and Optimization Conference, pp. 141–146. IEEE, Piscataway (2012)Mohamad, N.Z., Abidin, A.F., Munim, W.N.W.A., Yahya, M.: Symmetrical fault detection technique during power swing based on S‐Transform. In: IEEE Student Conference on Research and Development (SCOReD), pp. 7–10. IEEE, Piscataway (2012)Mohamad, N.Z., Abidin, A.F.: S‐Transform based technique to detect high resistance symmetrical fault during power swing. In: IEEE 8th International Colloquium on Signal Processing and its Applications, pp. 282–286. IEEE, Piscataway (2012)Cho, K., Kang, Y., Kim, S., Park, J., Kang, S., Kim, K.: An ANN based approach to improve the speed of a differential equation based distance relaying algorithm. IEEE Trans. Power Delivery 14(2), 349–357 (1999)Vaidya, A., Venikar, P.A.: ANN based distance protection of long transmission lines by considering the effect of fault resistance. In: IEEE‐International Conference on Advances In Engineering, Science and Management (ICAESM‐2012), pp. 590–594. IEEE, Piscataway (2012)Feilat, E., Al‐Tallaq, K.: A new approach for distance protection using artificial neural network. In: IEEE 39th International Universities Power Engineering Conference, vol. 1, pp. 473–477. IEEE, Piscataway (2004)Esmaeilian, A., Astinfeshan, S.: A novel power swing detection algorithm using adaptive neuro fuzzy technique. In: Proceedings of the IEEE International Conference on Electrical Engineering and Informatics, pp. 1–6. IEEE, Piscataway (2011)Sanaye‐Pasand, M., Jafarian, P.: An adaptive decision logic to enhance distance protection of transmission lines. IEEE Trans. Power Delivery 26(4), 2134–2144 (2011)Zadeh, H.K., Li, Z.: A novel power swing blocking scheme using adaptive neuro‐fuzzy inference system. Electr. Power Syst. Res. 78(7), 1138–1146 (2008)Rao, J.G., Pradhan, A.K.: Power‐swing detection using moving window averaging of current signals. IEEE Trans. Power Delivery 30(1), 368–376 (2014)Vapnik, V.: The Nature of Statistical Learning Theory. Springer Science & Business Media, Dordrecht (1999)Kampeerawat, W., Buangam, W., Chusanapiputt, S.: New swing‐blocking methods for digital distance protection using Support Vector Machine. In: IEEE International Conference on Power System Technology, pp. 1–6. IEEE, Piscataway (2010)Kampeerawat, W., Buangam, W., Chusanapiputt, S.: The swing‐blocking methods for digital distance protection based on wavelet packet transform and support vector machine. In: IEEE Asia‐Pacific Power and Energy Engineering Conference, pp. 1–4. IEEE, Piscataway (2012)Jena, P., Gupta, R.K.: Detection of fault in transmission line during power swing using fuzzy c‐means clustering transmission line protection. In: IEEE International Conference on Information, Communication, Instrumentation and Control (ICICIC), pp. 1–4. IEEE, Piscataway (2017)Sriram, C., Somlal, J., Goud, B.S., Bajaj, M., Elnaggar, M.F., Kamel, S.: Improved deep neural network (IDNN) with SMO algorithm for enhancement of third zone distance relay under power swing condition. Mathematics 10(11), 1944 (2022)Khodaparast, J., Khederzadeh, M.: Three‐phase fault detection during power swing by transient monitor. IEEE Trans. Power Syst. 30(5), 2558–2565 (2014)Platas‐Garza, M.A., Platas‐Garza, J., de la O Serna, J.A.: Dynamic phasor and frequency estimates through maximally flat differentiators. IEEE Trans. Instrum. Meas. 59(7), 1803–1811 (2009)Mai, R., He, Z., Fu, L., Kirby, B., Bo, Z.: A dynamic synchrophasor estimation algorithm for online application. IEEE Trans. Power Delivery 25(2), 570–578 (2010)Serna, J.A.O.: Phasor estimation from phasorlets. IEEE Trans. Instrum. Meas. 54(1), 134–143 (2005)de la O Serna, J.A., Rodriguez‐Maldonado, J.: Instantaneous oscillating phasor estimates with Taylor ^ K ‐Kalman filters. IEEE Trans. Power Syst. 26(4), 2336–2344 (2011)De la O Serna, J.A., Rodríguez‐Maldonado, J.: Taylor–Kalman–Fourier filters for instantaneous oscillating phasor and harmonic estimates. IEEE Trans. Instrum. Meas. 61(4), 941–951 (2012)Munoz, A.T., d l O Serna, J.A.: Shanks' method for dynamic phasor estimation. IEEE Trans. Instrum. Meas. 57(4), 813–819 (2008)Kang, D., Gokaraju, R.: A new method for blocking third‐zone distance relays during stable power swings. IEEE Trans. Power Delivery 31(4), 1836–1843 (2016)Torres, S., et al.: Unblocking function of distance relay during power swing based on modal analysis. In: IEEE PES Transmission & Distribution Conference and Exposition‐Latin America (PES T&D‐LA), pp. 1–6. IEEE, Piscataway (2016)Salehimehr, S., Taheri, B., Razavi, F., Parpaei, M., Faghihlou, M.: A new power swing detection method based on chaos theory. Electrical Eng. 102(2), 663–681 (2020)Robert, M.: Simple mathematical models with complicated dynamics. Nature 261, 459–467 (1976)Khan, U.N., Yan, L.: Power swing phenomena and its detection and prevention. In: 7th EEEIC International Workshop on Environment and Electrical Engineering, pp. 69–72. IEEE, Piscataway (2008)Benmouyal, G., Hou, D., Tziouvaras, D.: Zero‐setting power‐swing blocking protection. In: 31st Annual Western Protective Relay Conference. pp. 19–21. Washington State University, Pullman, WA (2004)Seethalekshmi, K., Singh, S., Srivastava, S.: SVM based power swing identification scheme for distance relays. In: IEEE PES General Meeting, pp. 1–8. IEEE, Piscataway (2010)Prasad, C.D., Biswal, M.: Swarm evaluated threshold elimination approach for symmetrical fault detection during power swing. IETE J. Res. 1–15 (2021)Das, S., Panigrahi, B.K., Jaiswal, P.K.: Qualitative assessment of power swing for enhancing security of distance relay in a TCSC‐compensated line. IEEE Trans. Power Delivery 36(1), 223–234 (2020)Hingorani, N.G., Gyugyi, L., El‐Hawary, M.: Understanding FACTS: Concepts and Technology of Flexible AC Transmission Systems. IEEE Press, New York (2000)Khodaparast, J., Khederzadeh, M., Faria da Silva, F., Back, C.L.: Performance of power swing blocking methods in UPFC‐compensated line. Int. Trans. Electr. Energy Syst. 27(11), e2382 (2017)Moravej, Z., Pazoki, M., Khederzadeh, M.: Impact of UPFC on power swing characteristic and distance relay behavior. IEEE Trans. Power Delivery 29(1), 261–268 (2013)Shekhar, H., Kumar, J.: Recognition of fault section during power swing in the presence of TCSC environment. Arab J. Sci. Eng. 47, 14229–14247 (2022)Sai Kumar, M., Kumar, J., Mahanty, R.: Fault detection during power swing in a TCSC‐compensated transmission line based on Clark's transform and Teager–Kaiser energy operator. Iran. J. Sci. Technol. 45(1), 47–60 (2021)Shekhar, H., Kumar, J., Nayak, A.: A fault detection technique during power swing in a TCSC‐compensated line using teager kaiser energy operator. IETE J. Res. 1–21 (2020)Nayak, P.K., Pradhan, A.K., Bajpai, P.: A fault detection technique for the series‐compensated line during power swing. IEEE Trans. Power Delivery 28(2), 714–722 (2013)Nayak, P.K., Pradhan, A.K., Bajpai, P.: Wide‐area measurement‐based backup protection for power network with series compensation. IEEE Trans. Power Delivery 29(4), 1970–1977 (2014)Alizadeh, B.A.M., Khederzadeh, M., Razzaghi, R.: Fault detection during power swing in thyristor‐controlled series capacitor‐compensated transmission lines. Electr. Power Syst. Res. 187, 106481 (2020)Juttu, T.R., Bhalja, B.R., Andreev, M., Malik, O.P.: Synchrophasor assisted power swing detection scheme for wind integrated transmission network. IEEE Trans. Power Delivery 37(3), 1952–1962 (2021)Niknezhad, M., Sadeh, J.: Analysis of wind farm penetration on power swing detection in distance relays. IET Renew. Power Gener. 16(2), 375–388 (2022)Tekdemir, I.G., Alboyaci, B.: A novel approach for improvement of power swing blocking and deblocking functions in distance relays. IEEE Trans. Power Delivery 32(4), 1986–1994 (2016)Musaruddin, M., Zaporoshenko, M., Zivanovic, R.: Remote protective relay testing. In: IEEE Australasian Universities Power Engineering Conference, pp. 1–4. IEEE, Piscataway (2008)Vebko Amirkabir Co.: https://vebko.org/en/Default.aspx Accessed 12 June 2018Verzosa, Q.: Realistic testing of power swing blocking and out‐of‐step tripping functions. In: IEEE 66th Annual Conference for Protective Relay Engineers, pp. 420–449. IEEE, Piscataway (2013)

Journal

IET Generation Transmission & DistributionWiley

Published: Feb 1, 2023

Keywords: Compensated lines; distance relay; power swing; power systems protection; wind turbine

There are no references for this article.