Fault Classification and Zone Identification in Power Transmission Systems

Abstract

Fault classification and zone identification in an electrical power transmission system is described. Voltage or current measurements are obtained at the terminals of a transmission line in each of three phases measured during a fault. Mode conversion is performed on the voltage or current measurements to obtain traveling wave signals for each of the three phases. Based on the magnitude of the traveling wave signals, the fault is classified.

Description

[Technical field]

[0001] Technical Field The present subject matter relates generally to fault classification and zone identification in electrical power transmission systems. [Background technology]

[0002] background Electric power transmission systems are large and complex networks comprising transmission lines with numerous electrical components such as generators, transformers, shunt reactors, etc. Power systems are often subject to system disturbances such as transmission line faults, loss of generating units, switching operations in highly loaded transmission lines, changes in load magnitude and direction, etc. In general, a fault may be defined as an abnormal condition in an electrical system that causes a disturbance in the normal flow of current. This deviant current flow causes a change in voltage and / or current flow, interrupting the transmission.

[0003] As the power system faces a paradigm shift, power grids around the world are undergoing a significant change in their generation mix, from synchronous AC rotating machines to inverter-based resource (IBR) technologies. A natural consequence of this trend is the deployment of more renewable sources of power in both transmission and distribution grids. In addition to the well-known environmental benefits, the introduction of these non-conventional technologies changes the operation of the power system, for example, by reducing inertia, reducing stability margins, and increasing unpredictability. Since most of these renewable technologies are interfaced with inverters, their behavior under fault conditions differs from that of conventional rotating machines. This creates new challenges for power system protection. Furthermore, one issue due to the high penetration of inverter-based resources, such as wind and photovoltaic (PV) generation, is the reduction in fault currents and short-circuit strength of the power grid. Summary of the Invention [Means for solving the problem]

[0004] overview SUMMARY OF THE DISCLOSURE Embodiments of the present invention provide a method for fault classification in an electrical power transmission system, a method for fault zone identification, an apparatus for fault classification and zone identification, and a computer-readable storage medium for fault classification and zone identification.

[0005] An objective of embodiments of the present invention may be to quickly and accurately clear faults in a power transmission system, regardless of the type of generating source connected to the power transmission system, particularly a power transmission system including an IBR. For example, a fault may be cleared in less than a quarter cycle in a system with a strong power source. In another example, a fault may be cleared in less than 5-7 ms in a power transmission system including an IBR.

[0006] According to a first aspect, a method for fault classification is provided. The method includes obtaining voltage or current measurements at terminals of a transmission line in each of three phases measured during a fault. Mode conversion is performed on the voltage or current measurements to obtain traveling wave signals for each of the three phases. Further, the fault is classified based on the magnitude of the traveling wave signals.

[0007] According to one embodiment, fault classification is performed without using incremental voltage or current measurements with respect to time.

[0008] According to one embodiment, the mode transformation is a Clarke transformation. According to one embodiment, the method generates a trip signal to cause a single phase trip or a three phase trip based on a classification of the fault.

[0009] According to one embodiment, the traveling wave signal includes a zero mode traveling wave signal for each of three phases, and the method includes comparing a magnitude of the zero mode traveling wave signal to a zero mode threshold value to distinguish earth faults from non-earth faults.

[0010] According to one embodiment, the traveling wave signals further include a beta mode traveling wave signal for each of the three phases, and the method includes, if the fault is a fault to ground, classifying the fault as involving a first of the three phases and ground when a first condition is met, The first condition is met when a magnitude of the beta traveling wave signal for the first phase is the minimum of the magnitudes of the beta traveling wave signals for each of the three phases and is less than a beta mode threshold.

[0011] According to one embodiment, the traveling wave signals further include an alpha mode traveling wave signal for each of the three phases, and the method includes, if the fault is a fault to earth, classifying the fault as involving two of the three phases and earth when a second condition is met. The second condition is met when the magnitude of the alpha traveling wave signal for the third of the three phases is the minimum of the magnitudes of the alpha traveling wave signals for each of the three phases. Furthermore, the fault is classified as involving the three phases and earth when the fault is a fault to earth and the second condition is not met. Furthermore, if the fault is a non-fault to earth, the method includes classifying the fault as involving two of the three phases when a third condition is met. The third condition is met when the magnitude of the alpha traveling wave signal for the third of the three phases is the minimum of the magnitudes of the alpha traveling wave signals for each of the three phases and is less than the alpha mode threshold. Furthermore, the fault is classified as involving three phases when the third condition is not met and the fault is non-fault to earth.

[0012] According to one embodiment, the terminals are connected to a synchronous generation resource or a grid-connected transmission line or an inverter-based generation resource.

[0013] According to a second aspect, a method for fault zone identification is provided. A zone of the transmission line where a fault exists is identified based on a traveling wave signal if the difference between the calculated line length and the actual line length is less than a threshold line length. A zone of the transmission line where a fault exists is identified based on a rate of change of incremental terminal current with respect to time if the difference between the calculated line length and the actual line length is greater than the threshold line length.

[0014] According to one embodiment, a method for fault zone identification includes receiving an identification of a transmission line having a fault and calculating a line length of the transmission line based on a traveling wave signal obtained from a mode conversion performed on voltage or current measurements determined at the terminals of the transmission line during the fault to obtain a calculated line length.

[0015] According to one embodiment, the zone identified based on the traveling wave signal is within one-third to two-thirds of the actual line length from the terminal, and the zone identified based on the rate of change of the incremental terminal current is within one-third of the actual line length from the terminal.

[0016] According to a third aspect, a method for fault classification and zone identification includes a method for fault classification as disclosed herein and a method for zone identification as disclosed herein.

[0017] According to a fourth aspect, an intelligent electronic device is disclosed for performing a method for fault classification and / or zone identification.

[0018] According to a fifth aspect, a non-transitory computer readable medium is disclosed that includes program instructions that, when executed, cause a processor to perform a method for fault classification and / or fault zone identification.

[0019] BRIEF DESCRIPTION OF THE DRAWINGS The features, aspects, and advantages of the present subject matter will become better understood with regard to the following description and the accompanying drawings, in which: The use of the same reference numbers in different figures indicates similar or identical features and components. [Brief description of the drawings]

[0020] [Figure 1(a)] FIG. 1 illustrates the monitoring of traveling wave signals for an A-phase to ground (Ag) fault for the A-phase in alpha mode, in accordance with an embodiment of the present subject matter. [Figure 1(b)]FIG. 1 illustrates the monitoring of traveling wave signals for an A-phase ground (Ag) fault for the A-phase in beta mode, in accordance with an embodiment of the present subject matter. [Figure 2(a)] FIG. 1 illustrates monitoring of traveling wave signals for a phase B to phase C (BC) fault on phase A in alpha mode, according to an embodiment of the present subject matter. [Figure 2(b)] FIG. 1 illustrates monitoring of traveling wave signals for a phase B-to-phase C (BC) fault on phase A in beta mode, according to an embodiment of the present subject matter. [Figure 3(a)] FIG. 1 illustrates the monitoring of traveling wave signals for a phase C to phase A ground (CA-g) fault in alpha mode in accordance with an embodiment of the present subject matter. [Figure 3(b)] FIG. 1 illustrates the monitoring of traveling wave signals for a phase C to phase A ground (CA-g) fault on phase A in beta mode in accordance with an embodiment of the present subject matter. [Figure 4(a)] FIG. 1 illustrates the monitoring of traveling wave signals for a three-phase-to-ground (ABC-g) fault case for phase A in alpha mode, in accordance with an embodiment of the present subject matter. [Figure 4(b)] FIG. 1 illustrates the monitoring of traveling wave signals for a three-phase-to-ground (ABC-g) fault case for the A phase in beta mode, in accordance with an embodiment of the present subject matter. [Figure 5(a)] FIG. 2 is a block diagram of a two-source equivalent electrical network for classifying faults, according to an embodiment of the present subject matter. [Figure 5(b)] FIG. 13 shows three-phase currents for an A-phase earth fault 50 km from bus M of a 150 km transmission line. [Figure 5(c)] FIG. 13 illustrates the alpha component of the current signal for an A-phase earth fault 50 km from bus M of a 150 km transmission line. [Figure 5(d)] FIG. 13 shows extracted traveling waves for an A-phase to earth fault 50 km from bus M of a 150 km transmission line. [Figure 5(e)] FIG. 2 is a block diagram of a two-source equivalent electrical network for identifying a zone in which a fault exists, according to one embodiment of the present subject matter. [Figure 6]FIG. 2 illustrates a technique for identifying whether a detected fault is within a defined zone, according to one embodiment of the present subject matter. [Figure 7] FIG. 1 is a grid diagram of a traveling wave based approach for line length calculation, in accordance with an embodiment of the present subject matter. [Figure 8(a)] FIG. 1 illustrates a method for classifying faults, according to one embodiment of the present subject matter. [Figure 8(b)] FIG. 2 illustrates a method for identifying a zone in which a fault exists, according to one embodiment of the present subject matter. [Figure 9] FIG. 1 illustrates a method for fault phase selection for fault classification, according to an embodiment of the present subject matter. [Figure 10] FIG. 2 illustrates a method for identifying a zone in which a detected fault exists, according to one embodiment of the present subject matter. [Figure 11] FIG. 13 illustrates a plot of incremental current thresholds and actual rates of change for fault zone identification in a first example of a first scenario. [Figure 12(a)] FIG. 13 illustrates a first peak of a beta traveling wave signal recorded at bus M in a first example of a first scenario, according to one embodiment of the present subject matter. [Figure 12(b)] FIG. 2 illustrates a first peak of a zero mode traveling wave signal recorded on bus M in a first example of a first scenario, according to one embodiment of the present subject matter. [Figure 13] FIG. 2 illustrates a traveling wave signal recorded on bus M showing the first three peaks and their arrival times in a first example of a first scenario, in accordance with an embodiment of the present subject matter. [Figure 14] FIG. 13 illustrates a plot of incremental current thresholds and actual rates of change for fault zone identification in a second example of a first scenario. [Figure 15(a)] FIG. 13 illustrates a first peak of an alpha traveling wave signal recorded at bus M in a second example of a first scenario, according to one embodiment of the present subject matter. [Figure 15(b)]FIG. 13 illustrates a first peak of a zero mode traveling wave signal recorded on bus M in a second example of a first scenario, according to an embodiment of the present subject matter. [Figure 16] FIG. 2 illustrates a traveling wave signal recorded on bus M showing the first three peaks and their arrival times in a second example of the first scenario, in accordance with an embodiment of the present subject matter. [Figure 17] FIG. 13 illustrates a plot of incremental current thresholds and actual rates of change for fault zone identification in a first example of a second scenario. [Figure 18(a)] FIG. 13 shows a first peak of a beta traveling wave signal recorded at bus M in a first example of the second scenario, according to one embodiment of the present subject matter. [Figure 18(b)] FIG. 13 illustrates a first peak of a zero mode traveling wave signal recorded on bus M in a first example of a second scenario, according to an embodiment of the present subject matter. [Figure 19] FIG. 13 illustrates a plot of incremental current thresholds and actual rates of change for fault zone identification in a second example of a second scenario. [Figure 20(a)] FIG. 13 shows a first peak of an alpha traveling wave signal recorded at bus M in a second example of the second scenario, according to one embodiment of the present subject matter. [Figure 20(b)] FIG. 13 illustrates a first peak of a zero mode traveling wave signal recorded on bus M in a second example of a second scenario, according to an embodiment of the present subject matter. [Figure 21] FIG. 13 illustrates a traveling wave signal recorded on bus M showing the first three peaks and their arrival times in a second example of the second scenario, in accordance with an embodiment of the present subject matter. [Figure 22(a)] FIG. 1 illustrates the operating time of a system having weak inverter-based resources for determining system reliability according to a conventional approach and a hybrid approach, in one example, in accordance with an embodiment of the present subject matter. [Figure 22(b)]FIG. 1 illustrates the operating time of a system having weak inverter-based resources for determining system reliability according to a conventional approach and a hybrid approach, in one example, in accordance with an embodiment of the present subject matter. [Figure 23(a)] FIG. 13 illustrates, in another example, the operating time of a system having weak inverter-based resources for determining system reliability according to a conventional approach and a hybrid approach, in accordance with an embodiment of the present subject matter. [Figure 23(b)] FIG. 13 illustrates, in another example, the operating time of a system having weak inverter-based resources for determining system reliability according to a conventional approach and a hybrid approach, in accordance with an embodiment of the present subject matter. [Figure 24(a)] FIG. 13 illustrates, in yet another example, a typical transmission line operation time for determining system reliability according to a conventional approach and a hybrid approach, in accordance with an embodiment of the present subject matter. [Figure 24(b)] FIG. 13 illustrates, in yet another example, a typical transmission line operation time for determining system reliability according to a conventional approach and a hybrid approach, in accordance with an embodiment of the present subject matter. [Figure 25(a)] FIG. 13 illustrates, in yet another example, the operation time of a dual circuit transmission line for determining system reliability according to a conventional approach and a hybrid approach, in accordance with an embodiment of the present subject matter. [Figure 25(b)] FIG. 13 illustrates, in yet another example, the operation time of a dual circuit transmission line for determining system reliability according to a conventional approach and a hybrid approach, in accordance with an embodiment of the present subject matter. [Figure 26(a)] FIG. 13 illustrates the operation time of a series compensation line for determining system reliability according to a conventional approach and a hybrid approach in yet another example, in accordance with an embodiment of the present subject matter. [Figure 26(b)] FIG. 13 illustrates the operation time of a series compensation line for determining system reliability according to a conventional approach and a hybrid approach in yet another example, in accordance with an embodiment of the present subject matter. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] Detailed Description The present subject matter relates to fault classification and zone identification in a power transmission system. In the following, fault classification and zone identification are described for a power system including inverter-based resource technology as an example. However, the present subject matter is not limited to fault classification and zone identification in a power system having an IBR.

[0022] Furthermore, the stability margin of a power system depends on the inertia of the power system, which determines the desired speed of the protection scheme used to identify and clear faults in the power system. Power systems that include IBRs have lower inertia and therefore lower stability margins. Lower stability margins imply lower critical clearing times. Faults must be cleared faster than the critical clearing time, otherwise the system may lose transient stability, which may lead to power system blackouts. Therefore, the integration of renewable sources into the power grid requires high-speed protection to maintain the limits of system stability.

[0023] Traditionally, time domain protection principles are used for high speed line protection. High sampling rates, processing power and communication capabilities of new relay hardware platforms are used to improve line protection operation speed. However, these solutions have drawbacks in terms of speed. Furthermore, the increasing addition of renewable sources to the primary power grid is expected to reduce the overall inertia, especially due to the high integration of photovoltaic power (PV). This may reduce the inertia constant and transient stability margin of the power system, and therefore the critical clearing time to maintain stable operation of the grid.

[0024] In one prior art technique, a time domain based algorithm based on the rate of change of incremental terminal current is used for fault zone identification. A threshold rate of change of incremental terminal current is calculated by assuming the fault is at the set Zone-1 boundary. The actual rate of change of incremental terminal current is calculated from the terminal current measurements. If the actual rate is greater than the threshold rate, the fault is identified as being within the set Zone-1 boundary. Alternatively, if the actual rate is less than the threshold rate, the fault is identified as being beyond the set Zone-1 boundary and no trip signal is generated.

[0025] Time domain based zone identification as described above generates trip signals with 100% certainty within <5 ms for faults within 50% of typical transmission line systems with conventional generation sources, as shown in Table I.

[0026] [Table 1]

[0027] However, for transmission line systems with inverter-based sources, the certainty of time domain based zone identification drops to 86% even for faults on 50% of the lines. The mean time to trip also increases undesirably as shown in Table II below.

[0028] [Table 2]

[0029] Therefore, the time domain based zone identification based on conventional techniques may be insufficient for reliable and fast protection of transmission lines having inverter-based power sources. Therefore, there is a need for a fast distance protection technique, i.e., fault classification and zone identification, that can protect the power system.

[0030] The subject matter provides a method for rapid fault classification and fast and accurate zone identification. Fault classification (phase selection) is achieved based on traveling wave signals obtained from mode conversion of current or voltage measurements. Furthermore, after fault classification, the fault area or zone is identified based on a combination of incremental magnitude and traveling wave principles. Thus, a hybrid approach is provided that combines incremental magnitude in the time domain with a traveling wave based distance protection approach to achieve better reliability and speed without compromising safety. The approach is suitable for all power systems, especially inverter-based resource (IBR) systems, regardless of the generation source used.

[0031] The method and apparatus disclosed herein can clear faults in less than a quarter cycle for systems with conventional generation sources. The method can clear faults in less than 5-7 ms for IBR systems. Furthermore, the method can be applied to series compensated lines, which do not malfunction during power swings and therefore may not require power swing blocking capabilities.

[0032] The above and other features, aspects, and advantages of the present subject matter will be better described with reference to the following description and accompanying drawings. Wherever possible, the same reference numbers are used in the drawings and the following description to refer to the same or like parts. Several examples are described, although modifications, adaptations, and other implementations are possible.

[0033] The traveling wave signals obtained for the three phases during the simulation of different types of faults in a power transmission line are first described to establish the principles by which a rapid fault classification can be performed. For this, as an example, a 220 kV, 150 km long transmission line is considered and the alpha and beta components of the traveling wave signals for four types of faults are monitored. In this example, for illustration purposes, the following faults are considered: Ag (phase-to-ground fault), BC (phase-to-phase fault), CA-g (phase-to-ground fault), and ABC-g (three-phase-to-ground fault). However, the alpha and beta components of the traveling wave signals of other phase-to-ground, phase-to-phase, and phase-to-ground faults also give similar results.

[0034] 1(a) and 1(b) show the monitoring of traveling wave signals for an A-phase ground (Ag) fault in (a) alpha mode and (b) beta mode, according to one embodiment of the present subject matter. In one example, a phase-to-ground (Ag) fault is simulated at 25 km of a 150 km transmission line, with a fault resistance of 20 Ω and an initiation angle of 60 degrees. FIG. 1(a) shows the monitoring of traveling wave signals for an A-phase ground (Ag) fault in alpha mode on A phase, and FIG. 1(b) shows the monitoring of traveling wave signals for an A-phase ground (Ag) fault in beta mode on A phase. As can be observed, for an A-phase ground fault, only an alpha mode signal for A phase exists, while the beta mode signal for A phase is negligible. Furthermore, the beta mode signals for B and C phases have significant magnitude when an A-phase-g fault occurs (not shown in the figures). Similarly, for a B-phase to earth (Bg) fault, the beta mode signal referenced to phase B is negligible, and for a C-phase to earth (Cg) fault, the beta mode signal referenced to phase C is negligible. Furthermore, since earth is involved in the fault, the magnitude of the zero mode signal cannot be neglected for these faults.

[0035] 2(a) and 2(b) show the monitoring of traveling wave signals for a phase B-C (BC) fault on phase A in (a) alpha mode and (b) beta mode according to one embodiment of the present subject matter. In one example, a phase-to-phase (phase B-C) ​​fault is simulated on 25 km of a 150 km transmission line with a fault resistance of 1 Ω and a starting angle of 60 degrees. FIG. 2(a) shows the monitoring of traveling wave signals for a phase B-C (BC) fault on phase A in alpha mode, and FIG. 2(b) shows the monitoring of traveling wave signals for a phase B-C (BC) fault on phase A in beta mode. It is observed that for a phase B-C fault, the alpha mode signal estimated on phase A is negligible, which can be used for fault type identification. Similarly, for phase A-B and phase C-A faults, the magnitude of the alpha mode signals on phase C and phase B, respectively, is close to zero. Also, since the ground is not involved in the fault, the magnitude of the zero mode signal is negligible for these faults.

[0036] 3(a) and 3(b) show the monitoring of traveling wave signals for a phase C to A ground (CA-g) fault on phase A in (a) alpha mode and (b) beta mode according to one embodiment of the present subject matter. In one example, a phase-to-phase ground (phase C to A ground) fault is simulated on 25 km of a 150 km transmission line with a fault resistance of 20 Ω and a starting angle of 60 degrees. FIG. 3(a) shows the monitoring of traveling wave signals for a phase C to A ground (CA-g) fault on phase A in alpha mode, and FIG. 3(b) shows the monitoring of traveling wave signals for a phase C to A ground (CA-g) fault on phase A in beta mode. For the CA-g fault, both alpha and beta mode signals are available. However, the magnitude of the alpha mode signal estimated with reference to phase B is the smallest of the alpha mode signals calculated with reference to all three phases. Also, since the ground is involved in the fault, the magnitude of the zero mode signal cannot be neglected for these faults.

[0037] 4(a) and 4(b) show the monitoring of traveling wave signals for a three-phase-to-ground (ABC-g) fault case for phase A in (a) alpha mode and (b) beta mode, according to one embodiment of the present subject matter. In one example, a three-phase-to-ground (ABC-g) fault is simulated at 25 km of a 150 km transmission line, with a fault resistance of 10 Ω and a starting angle of 60 degrees. FIG. 4(a) shows the monitoring of traveling wave signals for a three-phase-to-ground (ABC-g) fault in alpha mode for phase A, and FIG. 4(b) shows the monitoring of traveling wave signals for a three-phase-to-ground (ABC-g) fault in beta mode for phase A. As can be seen from FIG. 4(a) and FIG. 4(b), for a three-phase-to-ground fault, both alpha and beta mode signals are available.

[0038] Based on these observations, the present subject matter proposes a phase selection method based on traveling wave signal information for fast and reliable classification of faults, as described below.

[0039] FIG. 5(a) illustrates a block diagram of a two-source equivalent electrical network for classifying faults, according to one embodiment of the present subject matter. The two-source equivalent electrical network 500 comprises a power transmission line 512 of length L connected between two terminals, bus M 501 and bus N 502. Power sources 503 and 504 supply power to bus M 501 and bus N 502, respectively. In one example, power sources 503 and 504 may be connected to generators, such as synchronous generation resources or grid-connected transmission lines or inverter-based generation resources. The electrical network 500 may transmit power at high voltages, such as in the kilovolt range, and over long distances, such as tens or hundreds of kilometers.

[0040] It will be appreciated that the two-terminal system for fault classification 500 can include multiple additional components or devices for monitoring, sensing, and controlling various parameters that may be associated with the transmission line but are not shown for the sake of brevity. For example, components such as circuit breakers, sensors, current transformers, voltage transformers, loads connected to the transmission line, shunt reactors, intelligent electronic devices (IEDs), protective relays, etc. may be connected to the transmission line.

[0041] Additionally, a device 508 may be associated with the transmission line 512 for fault classification. For purposes of explanation, techniques for classifying faults are described with reference to a device 508 implemented at terminal M 501. However, similar techniques may be applied by a device at terminal N 502, as can be appreciated.

[0042] In one example, the device 508 may be an intelligent electronic device (IED). In another example, the device 508 may be any computing device, such as a server, desktop device, laptop, etc., capable of receiving measurements from an IED. In one example, the subject matter may be implemented by one or more modules. The modules may be implemented as instructions executable by one or more processors. For example, in an example where the device 508 performs a method, the modules are executed by the processor of the device 508. If the method is partially performed by the device 508 and partially performed by a server, the modules (depending on the steps) are distributed accordingly between the device 508 and the server.

[0043] In operation, the device 508 may be configured to receive input measurement signals associated with the bus M501, also referred to as the first terminal, from various measurement devices connected to the transmission line 512, such as current transformers, potential transformers, Rogowski coils, or other measurement sensors. For example, the input measurements may be obtained using a voltage transformer, shown as a VT, and / or a current transformer, shown as a CT. In one example, the device 508 may obtain voltage or current measurements, or both voltage and current measurements, associated with the bus M501. The device 508 may be configured to detect faults in the power transmission line using techniques known in the art. In response, the device 508 may be configured to classify the fault as further described below, and may generate a trip signal accordingly to isolate the fault.

[0044] To classify the fault, the device 508 can process the acquired measurements with the aid of a processor 520. The processor 520 may be implemented as a dedicated processor, a shared processor, or multiple individual processors, some of which may be shared. The device 508 can include a memory 526 that can be communicatively connected to the processor 520. Among other things, the processor 520 can fetch and execute computer-readable instructions stored in the memory 526. In one example, the memory 526 can store a fault classification module 522. In other examples, the fault classification module 522 can be external to the memory 526. The memory 526 can include any non-transitory computer-readable medium, including, for example, a volatile memory, such as a RAM, or a non-volatile memory, such as an EPROM, a flash memory, etc.

[0045] Additionally, the device 508 may include an output interface 524 for communicating results obtained from the fault classification module 522, for example to a server. In one example, if the method is implemented in a server, the device 508 may communicate the current and voltage measurements to the server via the output interface 524. The output interface 524 may include various computer-readable instruction-based interfaces and hardware interfaces that enable interaction with other communication, storage, and computing devices, such as network entities, web servers, databases, and external repositories, as well as peripherals. In one example, the fault classification, current and voltage measurements, etc. may be viewed on a display connected to the output interface 524 or integrated with the device 508.

[0046] In one example, upon detecting a fault in the transmission line, the processor 520 of the device 508 may execute a fault classification module 522 to classify the fault. The fault classification module 522 may obtain measurements of voltage or current at the terminals of the transmission in each of the three phases measured during the fault by measurement equipment associated with the transmission line. The voltage or current measurements may be obtained from a repository within the device 508 or from an external device or repository.

[0047] The fault classification module 522 may then perform mode conversion on the voltage or current measurements to obtain a traveling wave signal for each of the three phases. The traveling wave signal includes one or more of a beta mode traveling wave signal, an alpha mode traveling wave signal, and a zero mode traveling wave signal for each of the three phases. In one example, a mode conversion that may be performed is a Clarke transformation, and a mode conversion may be performed on each phase, i.e., phase A, phase B, and phase C, to obtain respective alpha, beta, and zero mode traveling wave signals.

[0048] In one example, a voltage or current measurement can be decomposed into two air mode signals (alpha mode signal and beta mode signal) and an earth mode signal. The decomposition of phase voltages or currents into air mode signal and earth mode signal can be performed by Clarke transformation. The air mode signal and earth mode signal can then be band-pass filtered to extract the traveling waves. For example, consider an A-phase earth fault at 50 km of a 150 km transmission line. Figure 5(b) shows the three-phase currents for an A-phase earth fault at 50 km from bus M of the 150 km transmission line. Figure 5(c) shows the alpha component of the current signal in A-phase for an A-phase earth fault at 50 km from bus M of the 150 km transmission line. Figure 5(d) shows the extracted alpha mode traveling wave in A-phase for an A-phase earth fault at 50 km from bus M of the 150 km transmission line. The high frequency transient component (traveling wave) is represented in FIG. 5(b) at 550, and the high frequency transient component in the alpha mode signal is represented in FIG. 5(c) at 552, which is extracted by passing the alpha mode signal shown in FIG. 5(c) through a band pass filter to obtain the traveling wave shown in FIG. 5(d). Based on the magnitude of the traveling wave signal, the fault may be classified. In one example, the fault classification may be performed without using the incremental amount of voltage or current measurements with respect to time. Furthermore, based on the classification of the fault, a single phase trip signal or a three phase trip signal may be generated.

[0049] To classify the fault, the fault classification module 522 can compare the magnitude of the zero mode traveling wave signal to a zero mode threshold to distinguish ground faults from non-ground faults. In one example, the zero mode threshold can be 0.01. Further, to determine whether the fault is a ground fault, the fault can be classified as a single-phase fault, a single-phase-to-ground fault, a phase-to-phase fault, a phase-to-ground fault, a three-phase fault, or a three-phase-to-ground fault based on the magnitude of the alpha and beta mode traveling wave signals of the three phases A, B, and C.

[0050] In one example, if the fault is a ground fault, the fault classification module 522 can classify the fault as involving a first of the three phases and ground when a first condition is met. The first condition can be met when the magnitude of the beta traveling wave signal for the first phase is the minimum of the magnitudes of the beta traveling wave signals of each of the three phases and is less than a beta mode threshold.

[0051] Furthermore, if the fault is a fault to ground, the fault classification module 522 can classify the fault as involving two of the three phases and ground when a second condition is met. The second condition can be met when the magnitude of the alpha traveling wave signal for the third of the three phases is the minimum of the magnitude of the alpha traveling wave signal for each of the three phases. When the second condition is not met but is a fault to ground, the fault classification module 522 can classify the fault as involving three phases and ground.

[0052] On the other hand, if the fault is not a fault to ground, i.e., the magnitude of the zero mode traveling wave signal is less than the zero mode threshold, the fault classification module 522 can classify the fault as involving two of the three phases when a third condition is met. The third condition can be met when the magnitude of the alpha traveling wave signal for the third of the three phases is the minimum of the magnitudes of the alpha traveling wave signals of each of the three phases and is less than the alpha mode threshold. When the third condition is not met and there is no fault to ground, the fault classification module 522 can classify the fault as involving three phases.

[0053] The phase selection process for classifying the fault is described in further detail with reference to Figure 9. Furthermore, based on the classification, the fault classification module can generate a single phase trip signal or a three phase trip signal.

[0054] FIG. 5(e) illustrates a block diagram of a two-source equivalent electrical network for identifying a zone in which a fault exists, according to one embodiment of the present subject matter. In one example, once a fault is detected and classified, a segment in which the fault may exist or a zone in which the fault may exist can be identified. In one example, the fault may be classified based on the classification technique described with respect to FIG. 5(a), or any other fault classification technique may be used. In one example, a two-terminal system 580 for fault zone identification may be similar to the two-terminal system 500 used for fault classification described with respect to FIG. 5(a).

[0055] As shown in FIG. 5(e), a device 540 may be associated with the transmission line 512 for zone identification. Similar to device 508, device 540 may be any computing device, such as a server, desktop device, laptop, etc., capable of receiving measurements from an intelligent electronic device or IED. In one example, device 508 and device 540 may be integrated into a single device to perform both fault classification and zone identification. Similar to device 508, device 540 may receive measurements of voltages and currents associated with bus M501 from various measurement devices connected to the transmission line 512, such as current transformers, potential transformers, Rogowski coils, or other measurement sensors. Furthermore, device 540 may be configured to identify a zone in which a fault may be located. For zone identification, device 540 may process the acquired measurements with the help of a processor 542, similar to processor 520.

[0056] In one example, the subject matter may be implemented by one or more modules executed by processor 542. The modules may be implemented as instructions executable by one or more processors. For example, in examples where device 540 executes the method, the modules are executed by the processor of device 540. If the method is partially executed by device 540 and partially executed by a server, the modules (depending on the steps) are distributed accordingly between device 540 and the server.

[0057] The apparatus 540 may include a memory 546 that may be communicatively coupled to the processor 542. Among other things, the processor 542 may fetch and execute computer-readable instructions stored in the memory 546. In one example, the memory 546 may store a fault zone identification module 544. In other examples, the fault zone identification module 544 may be external to the memory 546. The memory 546 may include any non-transitory computer-readable medium including, for example, a volatile memory such as a RAM, or a non-volatile memory such as an EPROM, a flash memory, or the like.

[0058] In one example, upon detecting a fault in the transmission line, a method for identifying a zone in which the fault may exist may be executed by the device 540. Furthermore, the device 540 may comprise an output interface 548 for communicating the results obtained from the fault zone identification module 544, for example to a server. In one example, if the method is implemented in a server, the device 540 may communicate the current and voltage measurements to the server via the output interface 548. The output interface 548 may include various computer-readable instruction-based interfaces and hardware interfaces that enable interaction with other communication, storage, and computing devices, such as network entities, web servers, databases, and external repositories, as well as peripherals. In one example, the fault zone identification, the obtained current and voltage measurements, etc. may be viewed on a display connected to the output interface 548 or integrated with the device 540.

[0059] To identify the zone in which the fault exists, the device 540 may be configured to receive an identification of the transmission line having the fault. For purposes of illustration, the method of identifying the fault zone, alternatively referred to as the fault segment, will be described with respect to the device 540 implemented at terminal M501. However, a similar method may be performed by the device at terminal N502, as can be appreciated.

[0060] During operation, the apparatus 540 may receive an indication that a fault exists in the transmission line 512. Upon receiving this indication, the line length of the transmission line 512 may be calculated. In one example, the line length may be calculated based on a traveling wave signal obtained from a mode conversion performed on voltage or current measurements determined at a terminal, such as terminal M501, of the transmission line 512 during the fault. Further, a comparison of the actual line length to the calculated line length may be used to identify the zone in which the fault exists.

[0061] In one example, the apparatus 540 can identify a zone of the transmission line where a fault exists based on the traveling wave signal if the difference between the calculated line length and the actual line length is less than the threshold line length. The zone identified based on the traveling wave signal can be within one-third to two-thirds of the actual line length from the terminal. In one example, the apparatus 540 can identify a zone of the transmission line where a fault exists based on the rate of change of the incremental terminal current with respect to time if the difference between the calculated line length and the actual line length is equal to or greater than the threshold line length. The zone identified based on the rate of change of the incremental terminal current can be within one-third of the actual line length from the terminal. Aspects of fault segment or fault zone identification are described in further detail below with reference to FIGS. 6 and 7.

[0062] FIG. 6 illustrates a technique for identifying whether a detected fault is within a defined zone, according to one embodiment of the present subject matter. It is observed that for faults very close to a local terminal, such as terminal 501 or 502, the time domain approach can be implemented as it is fast and reliable. However, for those faults occurring close to the terminal, the traveling wave approach may be unreliable due to the identification of multiple reflections and refractions. On the other hand, for faults occurring away from the local terminal, the time domain approach may be slower and less reliable. In particular, for faults over 50% of a transmission line having an inverter-based power supply, the time domain approach may be less efficient compared to the traveling wave-based approach. The traveling wave-based approach has been found to work very fast and accurately for portions of the line between 33.3% and 66.67% according to the present subject matter. Thus, the present subject matter implements a hybrid approach including the time domain approach and the traveling wave approach, thus ensuring ultra-fast and reliable protection of the transmission line up to 66.67% of the total line length from the terminal.

[0063] In one example, a transmission line such as transmission line 512 may be divided into three segments: segment-1, segment-2, and segment-3. In one example, segment-1 and segment-3 may be referred to as end segments since they may be adjacent to terminal terminations, and segment-2 may be referred to as middle segment since it may include the midpoint of the transmission line. Segment-1 may cover the beginning of the line from bus M501 to P1, which is approximately 33.33%, i.e., one-third of the transmission line from the terminal. Segment-2 covers P1, which is one-third (33.33%) of the transmission line from terminal 501 to P2, which is two-thirds (66.67%) of the transmission line from terminal 501 to terminal 502 of the transmission line. Segment-3 covers P2, which is two-thirds of the transmission line from terminal 501 to terminal 502 of the transmission line, which is 66.67%, i.e., one-third of the transmission line from terminal 502. In one example, a time domain based zone identification technique may be implemented to protect segment-1 of the transmission line, and a traveling wave based technique may be implemented to protect segment-2 of the transmission line. Since segment-3 corresponds to segment-1 as seen from terminal 502, segment-3 may be protected using the time domain based zone identification technique when current or voltage measurements are taken from terminal 502. Thus, the present subject matter efficiently combines the advantages of both the time domain based technique and the traveling wave based technique to achieve fast and reliable zone identification.

[0064] 7 shows a lattice diagram of a traveling wave-based approach for line length calculation, according to one embodiment of the present subject matter. During the occurrence of a fault in a transmission line, a high frequency transient signal may be generated that can travel at a speed approximately equal to the speed of light along the transmission line to the nearest impedance discontinuity in the interconnected power system. From the impedance discontinuity, the wave can be reflected and refracted, based on which the line length can be calculated.

[0065] As mentioned above, the portion of the transmission line between one-third and two-thirds of the line (P1-P2) can be protected by traveling wave based techniques. In one example, consider a fault in segment-2 of the transmission line, Figure 7 represents its lattice diagram.

[0066]

number

[0067] The line length can be calculated based on equation (1-3) for a fault in segment-2 of the transmission line as shown in equation (4).

[0068]

number

[0069] In one example, the fault zone identification module 544 may be configured to compare the calculated line length to the actual known line length. It will be appreciated that in other implementations, other techniques for calculating the line length may be used. In one example, the actual known line length may be stored in memory 546. Further, if the difference between the calculated line length and the actual line length is less than the threshold line length, the fault may be identified as being in segment-2. However, if the difference between the calculated line length and the actual line length is greater than the threshold line length, a time domain approach may be used.

[0070] Thus, the present subject matter provides a fast and reliable fault classification system based on the traveling wave technique and a hybrid solution based on incremental time domain and traveling wave principles for identifying the segment (zone) in which the fault exists.

[0071] 8(a), 8(b), 9, and 10 illustrate methods for fault classification, fault-present zone identification, fault phase selection, and fault zone identification according to various embodiments of the present subject matter. The order in which methods 800, 801, 900, 1000 are described is not intended to be construed as a limitation, and some of the described method blocks may be performed in different orders to implement methods 800, 801, 900, 1000 or alternative methods. Furthermore, methods 800, 801, 900, 1000 may be implemented in any suitable hardware, computer readable instructions, firmware, or combinations thereof. For purposes of illustration, methods 800, 801, 900, 1000 are described with respect to the implementations shown in FIGS. 5(a) and 5(e).

[0072] 8(a) illustrates a method 800 for classifying a fault, according to one embodiment of the present subject matter. In the method 800, at block 802, voltage or current measurements are obtained at the terminals of the transmission line in each of the three phases measured during the fault. In one example, the terminals may be connected to a synchronous generation resource or a grid-connected transmission line or an inverter-based generation resource. In one example, the voltage measurements of each phase at the terminals of the transmission system may be obtained by a voltage or potential transformer, and the current measurements of each phase at the terminals of the transmission system may be obtained by a current transformer. In one example, the current transformer and the voltage transformer may be operably connected to the device at the terminals M501.

[0073] A mode transformation is performed on the voltage or current measurements for each of the three phases to obtain traveling wave signals at block 804. In one example, Clarke transformation, a method well known in the art, may be performed on the voltage or current measurements for phases A, B, and C, after which alpha, beta, and zero mode traveling wave signals may be obtained for each of the three phases as described above with respect to Figures 5(a)-5(d).

[0074] At block 806, the fault is classified based on the magnitude of the traveling wave signal. In one example, the fault may be classified as a single-phase fault, a two-phase fault, or a three-phase fault, whether or not it is a fault to ground. Fault classification and phase selection are described in more detail with reference to FIG. 9. Based on the fault classification, a single-phase trip or a three-phase trip signal may be generated.

[0075] In one example, the magnitude of the zero mode traveling wave signal can be compared to a zero mode threshold to distinguish ground faults from non-ground faults.

[0076] In one example, if the fault is a ground fault, the fault may be classified as involving a first of the three phases and ground when a first condition is met. The first condition may be met if the magnitude of the beta traveling wave signal for the first phase is the minimum of the magnitudes of the beta traveling wave signals for each of the three phases and is less than a beta mode threshold.

[0077] In one example, if the fault is a ground fault, the fault may be classified as involving two of the three phases and ground when the second condition is met. The second condition may be met when the magnitude of the alpha traveling wave signal for the third of the three phases is the minimum of the magnitudes of the alpha traveling wave signals for each of the three phases. Furthermore, the fault may be classified as involving three phases and ground when the first and second conditions are not met and the fault is a ground fault.

[0078] In another example, if the fault is a non-ground fault, the fault may be classified as involving two of the three phases when a third condition is met. The third condition may be met if the magnitude of the alpha traveling wave signal for the third of the three phases is the minimum of the magnitudes of the alpha traveling wave signals for each of the three phases and is less than the alpha mode threshold. Furthermore, the fault may be classified as involving three phases when the third condition is not met and the fault is a non-ground fault.

[0079] In one example, based on the fault classification as described above, a trip signal may be generated to cause a single phase trip or a three phase trip based on the classification of the fault. In one example, the fault classification module 522 of the device 508 may be configured to classify the fault.

[0080] 8(b) illustrates a method 801 for identifying a zone where a fault exists, according to one embodiment of the present subject matter. In the method 801, at block 810, a zone of the transmission line where a fault exists may be identified based on the traveling wave signal if the difference between the calculated line length and the actual line length is less than a threshold line length. Further, at block 812, a zone of the transmission line where a fault exists may be identified based on a rate of change of the incremental terminal current with respect to time if the difference between the calculated line length and the actual line length is equal to or greater than the threshold line length. In one example, determining a zone of the transmission line where a fault exists based on a rate of change of the incremental terminal current with respect to time includes calculating an actual rate of change of the incremental terminal current based on current measurements.

[0081] In one example, the zone identified based on the traveling wave signal is within one-third to two-thirds of the actual line length from the terminal, and the zone identified based on the rate of change of the incremental terminal current is within one-third of the actual line length from the terminal.

[0082] FIG. 9 illustrates a method 900 for fault phase selection for fault classification, according to one embodiment of the present subject matter. The phase selection can be performed based on the traveling wave principle as described above. At block 902, alpha, beta, and zero mode signals can be generated after performing a mode conversion for phase A on the voltage or current measurements taken at the terminals. Similarly, at block 904, alpha, beta, and zero mode signals can be generated after performing a mode conversion for phase B on the voltage or current measurements taken at the terminals, and at block 906, alpha, beta, and zero mode signals can be generated after performing a mode conversion for phase C on the voltage or current measurements taken at the terminals. In one example, the mode conversion performed for phase A, phase B, and phase C can be Clarke transformation.

[0083] At block 908, the zero mode signal for each phase may be compared to a zero mode threshold. A comparison of the magnitude of the zero mode signal to the zero mode threshold may be used to determine whether the fault is a fault to ground. In one example, the zero mode threshold may be 0.01. Although block 908 shows that the zero mode reference signal for phase A is used for the comparison for ease of representation, it will be understood that the same comparison may be made for the zero mode signals of phases B and C.

[0084] If the magnitude of the zero mode signal is greater than the zero mode threshold, the fault may be classified as a phase-to-ground fault, a phase-to-ground fault, or a three-phase-to-ground fault based on one or more conditions that are met.

[0085] At block 910, if the magnitude of the beta traveling wave signal for phase A is the minimum of the magnitudes of the beta traveling wave signals for the three phases A, B, and C, and the magnitude of the beta traveling wave signal for phase A is less than the beta mode threshold, the fault can be classified as an A-phase ground fault, and therefore a single-phase trip signal can be generated to trip phase A.

[0086] At block 912, if the magnitude of the beta traveling wave signal for phase B is the minimum of the magnitudes of the beta traveling wave signals for the three phases A, B, and C, and the magnitude of the beta traveling wave signal for phase B is less than the beta mode threshold, the fault can be classified as a phase B to ground fault, and therefore a single-phase trip signal can be generated to trip phase B.

[0087] Similarly, in block 914, if the magnitude of the beta traveling wave signal for phase C is the minimum of the magnitudes of the beta traveling wave signals for phases A, B, and C, and the magnitude of the beta traveling wave signal for phase C is less than the beta mode threshold, the fault can be classified as a phase C-to-ground fault, and therefore a single-phase trip signal can be generated to trip phase C.

[0088] At block 916, if the magnitude of the alpha traveling wave signal for phase C is the minimum of the magnitudes of the alpha traveling wave signals for the three phases A, B, and C, the fault can be classified as a phase-to-ground fault or a three-phase-to-ground (ABC-g) fault between phases A, B, and ground, and therefore a three-phase trip signal can be generated.

[0089] At block 918, if the magnitude of the alpha traveling wave signal for phase A is the minimum of the magnitudes of the alpha traveling wave signals for the three phases A, B, and C, the fault can be classified as a phase-to-ground fault or a three-phase-to-ground (ABC-g) fault between phases B, C, and ground, and therefore a three-phase trip signal can be generated.

[0090] At block 920, if the magnitude of the alpha traveling wave signal for phase B is the minimum of the magnitudes of the alpha traveling wave signals for the three phases A, B, and C, the fault can be classified as a phase-to-ground fault or a three-phase-to-ground (ABC-g) fault between phases C, A, and ground, and therefore a three-phase trip signal can be generated.

[0091] In one example, the fault classifications generated in blocks 916, 918, and 920 may be sent to an OR gate 950 to generate a three-phase trip signal. To ensure safety and reduce damage, a three-phase trip signal may be generated if a fault occurs between two or more phases.

[0092] Returning to block 908, if the fault is a non-ground fault, i.e., if the magnitude of any of the zero mode traveling signals is less than or equal to the zero mode threshold, the method 900 proceeds to block 922. At block 922, if the magnitude of the alpha traveling wave signal for phase C is the minimum of the magnitudes of the alpha traveling wave signals for the three phases A, B, and C, and the magnitude of the alpha traveling wave signal for phase C is less than the alpha mode threshold, the fault can be classified as a phase-to-phase fault between phase A and phase B.

[0093] In block 924, if the magnitude of the alpha traveling wave signal for phase A is the minimum of the magnitudes of the alpha traveling wave signals for the three phases A, B, and C, and the magnitude of the alpha traveling wave signal for phase A is less than the alpha mode threshold, the fault can be classified as an inter-phase fault between phase B and phase C.

[0094] Similarly, in block 926, if the magnitude of the alpha traveling wave signal for phase B is the minimum of the magnitudes of the alpha traveling wave signals for the three phases A, B, and C, and the magnitude of the alpha traveling wave signal for phase B is less than the alpha mode threshold, the fault can be classified as an inter-phase fault between phase C and phase A.

[0095] If the conditions in any one of blocks 922, 924, and 926 are not met, the fault can be classified as three phase (ABC fault). In either case, i.e., whether the fault is a two-phase fault or a three-phase fault, a three-phase trip signal can be generated. In one example, all the fault classifications generated from blocks 922, 924, and 926 can be sent to an OR gate 952, based on which a final three-phase trip signal can be generated. To ensure safety and reduce damage, a three-phase trip signal can be generated if a fault occurs between two or more phases.

[0096] 10 illustrates a method 1000 for identifying a zone in which a detected fault resides, according to one embodiment of the present subject matter. The segment in which the fault may have occurred may be determined as described above with respect to classifying the fault. In one example, the fault classification may be performed using the method described above with respect to FIG. 8(a) and FIG. 9. In other examples, any fault classification method known in the art may be performed.

[0097] In one example, the transmission line may be divided into three segments, segment-1, segment-2, and segment-3, as described with respect to FIG. 6. In classifying the fault, a hybrid time domain based traveling wave approach can be used to locate the faulty segment and generate a trip signal. For faults very close to the local terminal, the time domain approach can be implemented as it is fast and reliable, while for segments located at one-third to two-thirds of the transmission line, the traveling wave based approach can be used as it is very fast and accurate. Calculations based on both the time domain approach and the traveling wave approach are performed simultaneously, and based on the calculations, the faulty segment can be identified.

[0098] At block 1002, a time domain based approach may be implemented in which the reach (segment in which the fault exists) may be set to one third of the actual line length of the transmission line. To implement the time domain approach, the rate of change of the incremental terminal current with respect to time is monitored. A threshold rate of change of the incremental terminal current is calculated by assuming that the fault is at the segment-1 boundary. Further, the actual rate of change of the incremental terminal current is calculated based on current measurements taken from the terminals. If the actual rate of change of the incremental terminal current is greater than the threshold rate of change of the incremental terminal current, the fault may be identified as being within segment-1 and zone identification may be provided for trip signal generation. Alternatively, if the actual rate of change of the incremental terminal current is less than the threshold rate of change of the incremental terminal current, the fault may be identified as being beyond segment-1 and zone identification may not be provided from block 1002.

[0099] At block 1004, a travel-based approach may be implemented in which the arrival times of the first, second, and third travelling waves at terminal M are recorded. Based on parameters such as the arrival times and propagation speed, the line length may be calculated as shown in equation (4) above.

[0100] At block 1006, the calculated line length is compared to the actual known line length, and if the difference is less than the threshold line length, the fault is identified as being in segment-2 of the transmission line and fault zone identification is provided from block 1006. However, if the difference is greater than or equal to the threshold line length, no fault zone identification is provided from block 1006. In one example, the threshold may be set depending on the sampling frequency. If the sampling rate is 1 MHz, the threshold may be set to 300 m.

[0101] The fault zone identification from blocks 1002 and 1006 may be sent to an OR gate 1010 (a combinational logic gate) to generate a final trip signal. When OR gate 1010 receives a trip signal from either block 1002 or block 1006, it sends a signal to block 1008, and a trip signal may be generated accordingly.

[0102] Considering that the fault is identified as a single-phase to earth (Ag) fault, the above segment identification method can be illustrated by the following example. After classifying the fault as an A-phase to earth fault, zone identification using the time domain method as well as the traveling wave method is performed. If the time domain method issues a zone identification signal before the traveling wave method, the fault can be identified as being located in the end segment (first third of the line length), and if the traveling wave method issues a trip signal before the time domain method, the fault can be identified as being located in the central segment of the transmission line (middle third of the line segment) from the terminal where the current or voltage measurements were taken.

[0103] Thus, the present subject matter provides a fast and reliable fault classification technique based on traveling wave principles and identifies segments where faults may be located based on a hybrid time domain and traveling wave approach.

[0104] example To determine the performance of the fault classification and hybrid methods, different fault scenarios with different types of faults, different values ​​of fault resistance, source-to-line impedance ratio and fault location were simulated and tested using conventional sources and inverter-based resources.

[0105] In the first scenario, the system was tested on a transmission line system with a conventional generator for a 200 km long 400 kV, 50 Hz transmission line.

[0106] In the first example, a single phase-to-earth Ag fault on 10% of the line with a fault resistance of 5 ohms and a fault initiation time of 0.1 seconds was considered. The time domain approach and the hybrid approach were tested.

[0107] FIG. 11 illustrates a plot of threshold and actual rate of change of incremental current for fault zone identification by the time domain approach for a first example of a first scenario. Plot line 1102 represents the actual rate of change of current and plot line 1104 represents the threshold rate of change of current. It was observed that phase selection using the time domain based approach took approximately 3.17 ms to identify the fault type as an A-phase to ground fault. Furthermore, it was observed that the time domain based zone identification took 1.87 ms to identify that the fault is within zone-1.

[0108] Figure 12 12(a) and 12(b) show the first peaks of the beta and zero mode traveling wave signals recorded at bus M in a first example of a first scenario according to an embodiment of the present subject matter. FIG. 12(a) shows the traveling wave signal of an Ag fault in beta mode. Line 1202 shows the beta mode signal for phase C, and line 1204 shows the beta mode signal for phase A. Here, it can be observed that line 1204 for phase A is the smallest and has a negligible magnitude of 6.09e-9. FIG. 12(b) shows the traveling wave signal of the zero mode signal represented by line 1206. Since line 1206 shows that the zero mode signal has a significant magnitude, it is concluded that the fault is a ground fault and the fault is classified as an A-phase ground Ag fault. When using phase selection using a traveling wave based approach, it was observed that the time it takes for phase selection to identify the fault type as an A-phase ground fault is 34 μs.

[0109] FIG. 13 illustrates a traveling wave signal recorded at bus M showing the first three peaks and their arrival times in a first example of a first scenario, according to an embodiment of the present subject matter. For a fault initiation at 0.1 s, the first three peaks and the traveling wave signal recorded at terminal M were used to calculate the line length based on equation (4) as described above. Using equation (4), the calculated line length was determined to be 59.99 km, which is significantly different from the actual line length of 200 km. Therefore, no trip signal was generated based on the traveling wave-based approach. Rather, the time domain approach is used for zone identification (after phase selection by the traveling wave approach) and for generating a trip signal by the hybrid approach. Table 3 provides a comparative study of the operating times of the traditional time-based approach and the hybrid approach.

[0110] [Table 3]

[0111] From Table 3, it can be observed that in case of fault in segment-1 (0 to 1 / 3) of the transmission line, the time domain based approach alone takes 3.17 ms to generate the final trip signal. This time arises from the operation time of the phase selection module since phase selection and zone identification are performed in parallel. On the other hand, in the hybrid approach, the traveling wave based phase selection gives the phase selection signal within 34 μs and the time domain part of the zone identification approach generates the trip signal within 1.87 ms (zone identification is performed by the time domain approach since the traveling wave approach does not identify the zones accurately). Hence, the final trip time is 1.87 ms for the hybrid approach. Hence, the hybrid solution achieves faster operation as compared to the time domain approach.

[0112] In the second example, a phase-to-phase BC fault on 66% of the lines with a fault resistance of 50 Ohms and a fault initiation time of 0.095 seconds was considered. A conventional time domain based approach and a hybrid method combining the time domain based approach and the traveling wave based approach were tested.

[0113] FIG. 14 illustrates a plot of threshold and actual rate of change of incremental current for fault zone identification in a second example of the first scenario. Plot line 1402 represents the actual rate of change of current, and plot line 1404 represents the threshold rate of change of current. As can be seen from the plot, the actual rate of change of current line 1402 does not intersect the threshold rate of change of current line 1404, and therefore no trip signal is generated. It was observed that phase selection using the time domain based approach took approximately 4.58 ms to identify the fault type as a B-phase to C-phase fault. It was further observed that the time domain based zone identification did not generate a trip signal.

[0114] FIG. 15(a) and FIG. 15(b) show the first peaks of the alpha and zero mode traveling wave signals recorded at bus M in a second example of the first scenario according to an embodiment of the present subject matter. FIG. 15(a) shows the traveling wave signal of a BC fault in alpha mode. Line 1502 shows the alpha mode signal for phase C, and line 1504 shows the alpha mode signal for phase A. Here, it can be observed that line 1504 is the smallest and has a negligible magnitude of 11.9e-10. FIG. 15(b) shows the traveling wave signal of the zero mode signal represented by line 1506 having a negligible magnitude of around 1e-13. Since line 1506 shows that the zero mode signal has a negligible magnitude, it is concluded that the fault is a non-ground fault and the fault is classified as a phase B to phase C fault. In this case, when phase selection using a traveling wave based approach is used, it is observed that the time it takes for phase selection to identify the fault type as a phase B to phase C fault is 36 μs.

[0115] FIG. 16 illustrates a traveling wave signal recorded at bus M showing the first three peaks and their arrival times in a second example of the first scenario, according to an embodiment of the present subject matter. For fault initiation at 0.095 s, the first three peaks and the traveling wave signal recorded at terminal M were used to calculate the line length based on equation (4) as described above. Using equation (4), the calculated line length was determined to be 199.98 km, which is very close to the actual line length of 200 km. Thus, a trip signal was generated based on the traveling wave-based approach (of the hybrid approach) at about 0.14 ms. Table 4 provides a comparative study of the operating times of the traditional time-based approach and the hybrid approach.

[0116] [Table 4]

[0117] From Table 4, it can be observed that in case of fault in segment-2 (1 / 3 to 2 / 3) of the transmission line, the time domain algorithm alone is not able to generate the trip signal. However, in the hybrid approach, the traveling wave based phase selection gives the phase selection signal within 34 μs and the traveling wave based zone identification approach generates the trip signal within 0.14 ms. Hence, the final trip time is 0.14 ms for the hybrid approach. This example demonstrates that the solution generates the trip signal at a faster rate even when the time domain approach fails. Hence, the hybrid solution provides increased certainty due to higher fault resistance and fault location in one-third to two-thirds of the line as compared to the time domain approach.

[0118] In the second scenario, the system was tested for a transmission line system with a weak inverter-based power source on a 220 kV, 50 km single-circuit transmission line with a source-to-line impedance ratio of 30:5.

[0119] In the first example, a single phase-to-earth Ag fault on 32% of the line with a fault resistance of 5 ohms and a fault initiation time of 0.3 seconds was considered. The time domain and hybrid approaches were tested.

[0120] FIG. 17 illustrates a plot of threshold and actual rate of change of incremental current for fault zone identification in a first example of the second scenario. Plot line 1702 represents the actual rate of change of current, and plot line 1704 represents the threshold rate of change of current. As can be seen from the plot, the actual rate of change of current line 1702 does not intersect with the threshold rate of change of current line 1704, and therefore no trip signal is generated. It was observed that phase selection using the time domain based approach is unable to identify the exact fault loop based on the current signals. Furthermore, it was observed that the time domain based zone identification takes 17.5 ms to identify that the fault is within zone-1 and generate a trip signal.

[0121] FIG. 18(a) and FIG. 18(b) show the first peaks of the beta and zero mode traveling wave signals recorded at bus M in a first example of the second scenario according to an embodiment of the present subject matter. FIG. 18(a) shows the traveling wave signal of an Ag fault in beta mode. Line 1802 shows the beta mode signal for phase C, and line 1804 shows the beta mode signal for phase A. Here, it can be observed that line 1804 for phase A is the smallest and has a negligible magnitude of 8.12e-9. FIG. 18(b) shows the traveling wave signal of the zero mode signal represented by line 1806. Since line 1806 shows that the zero mode signal has a significant magnitude, it is concluded that the fault is a fault to earth and the fault is classified as an A-phase to earth Ag fault. In this case, it was observed that when using phase selection using a traveling wave based approach, the time it takes for phase selection to identify the fault type as an A-phase to earth fault is 34 μs.

[0122] Further, in the case of fault initiation at 0.1 s, the first three peaks and traveling wave signal recorded at terminal M were used to calculate the line length based on equation (4) as discussed above. Using equation (4), the calculated line length was determined to be 39.99 km, which is significantly different from the actual line length of 50 km. Hence, no trip signal was generated based on the traveling wave-based approach. Rather, the time domain approach is used for zone identification (after phase selection by the traveling wave approach) and generation of trip signal by the hybrid approach. Table 5 provides a comparative study of the operating times of the traditional time-based approach and the hybrid approach.

[0123] [Table 5]

[0124] From Table 5, it can be observed that in case of fault in segment-1 (0 to 1 / 3) of the transmission line, the time domain based approach fails to generate a trip signal since the fault loop is not identified accurately. But in the hybrid approach, the traveling wave based phase selection gives the phase selection within 34 μs and the time domain part of the zone identification approach generates a trip signal within 7.8 ms (zone identification is done by the time domain approach since the traveling wave approach does not identify the zones accurately). Hence, the final trip time is 7.8 ms for the hybrid approach. Hence, using only the time domain algorithm may reduce the certainty for systems with weak sources / renewables. The hybrid protection solution has been proven to increase the certainty especially for systems with inverter based sources.

[0125] In the second example, a phase-to-ground CA-g fault on 66% of the line with a fault resistance of 50 Ohms and a fault initiation time of 0.29 seconds was considered. A time domain based approach and a hybrid approach combining the time domain based approach and the traveling wave based approach were tested.

[0126] FIG. 19 illustrates a plot of threshold and actual rate of change of incremental current for fault zone identification in a second example of the second scenario. Plot line 1902 represents the actual rate of change of current and plot line 1904 represents the threshold rate of change of current. As can be seen from the plot, the actual rate of change of current line 1902 does not intersect the threshold rate of change of current line 1904 and therefore no trip signal is generated. It was observed that phase selection using the time domain based approach took approximately 13.7 ms to identify the fault type as a C-phase to A-phase ground fault. It was further observed that the time domain based zone identification does not generate a trip signal.

[0127] FIG. 20(a) and FIG. 20(b) show the first peaks of the alpha and zero mode traveling wave signals recorded at bus M in a second example of the second scenario, according to an embodiment of the present subject matter. FIG. 20(a) shows the traveling wave signal of a CA-g fault in alpha mode. Line 2002 shows the alpha mode signal for phase A, line 2004 shows the alpha mode signal for phase C, and line 2006 shows the alpha mode signal for phase B. Here, it can be observed that line 2006 is a minimum compared to lines 2002 and 2004. FIG. 20(b) shows the traveling wave signal of the zero mode signal represented by line 2008. Since line 2008 shows that the zero mode signal has a significant magnitude, it is concluded that the fault is a ground fault and the fault is classified as a phase C to phase A ground CA-g fault. In this case, it was observed that when phase selection using the traveling wave based approach was used, the time it took for phase selection to identify the fault type as a C phase to A phase ground fault was 36 μs.

[0128] FIG. 21 illustrates a traveling wave signal recorded at bus M showing the first three peaks and their arrival times in a second example of the second scenario, according to an embodiment of the present subject matter. For fault initiation at 0.29 s, the first three peaks and the traveling wave signal recorded at terminal M were used to calculate the line length based on equation (4) as described above. Using equation (4), the calculated line length was determined to be 59.99 km, which is very close to the actual line length of 50 km. Thus, a trip signal was generated based on the traveling wave-based approach (of the hybrid approach) at about 0.27 ms. Table 6 provides a comparative study of the operating times of the traditional time-based approach and the hybrid approach.

[0129] [Table 6]

[0130] From Table 6, it can be observed that in case of a fault in segment-2 (1 / 3 to 2 / 3) of the transmission line, the time domain algorithm alone is not able to generate a trip signal. However, in the hybrid approach, the traveling wave based phase selection gives the phase selection signal within 36 μs and the traveling wave based zone identification approach generates a trip signal within 0.27 ms. Hence, the final trip time is 0.27 ms for the hybrid approach. The hybrid approach generates a fast trip signal even if the time domain approach fails. Hence, the hybrid solution increases the certainty especially for faults very close to the zone boundaries. Also, the hybrid protection solution has been proven to increase the certainty especially for systems having inverter based power sources.

[0131] The system and method were tested on different systems including a normal single circuit transmission line, a double circuit line, a series compensated line and a line with a weak power supply. A summary of the results for each of these systems is provided below.

[0132] FIG. 22(a) and FIG. 22(b) show operation times of a system with weak inverter-based resources to determine the reliability of the system with the conventional approach and the hybrid approach in one example according to an embodiment of the present subject matter. A system with inverter-based resources for a 100 km transmission line with a source-to-impedance ratio of 15:2 was considered. FIG. 22(a) and FIG. 22(b) show plots of operation times obtained with the conventional time domain approach and the hybrid approach. In FIG. 22(a), the location of the fault is plotted against the maximum trip time measured in milliseconds. Line 2202 represents the operation time plot obtained for the conventional time domain based approach, and line 2204 represents the operation time plot obtained for the hybrid approach. In FIG. 22(b), the location of the fault is plotted against the average trip time measured in milliseconds. Line 2206 represents the operation time plot obtained for the conventional time domain based approach, and line 2208 represents the operation time plot obtained for the hybrid approach. From Figure 22(a) and Figure 22(b), it can be observed that as the position (%) increases along the x-axis, the time it takes for the system to generate a trip signal based on the conventional approach also increases, but the time it takes for the system to generate a trip signal based on the hybrid approach is comparatively very short, thereby increasing the reliability of the system.

[0133] Based on the operating times obtained from both conventional and hybrid methods, the robustness of the system using the two methods is shown in Table 7.

[0134] [Table 7]

[0135] It is observed that when the fault location increases beyond one-third of the transmission line length, the conventional approach does not give 100% accurate results, whereas the hybrid approach gives 100% reliable results.

[0136] FIG. 23(a) and FIG. 23(b) show the operation time of a system with weak inverter-based resources to determine the reliability of the system with the conventional approach and the hybrid approach in another example according to an embodiment of the present subject matter. A system with inverter-based resources for a 50 km transmission line with a source-to-impedance ratio of 30:5 was considered. FIG. 23(a) and FIG. 23(b) show plots of operation time obtained with the conventional time domain approach and the hybrid approach. In FIG. 23(a), the location of the fault is plotted against the average trip time measured in milliseconds. Line 2302 represents the operation time plot obtained for the conventional time domain based approach, and line 2304 represents the operation time plot obtained for the hybrid approach. In FIG. 23(b), the location of the fault is plotted against the maximum trip time measured in milliseconds. Line 2306 represents the operation time plot obtained for the conventional time domain based approach, and line 2308 represents the operation time plot obtained for the hybrid approach. From Figure 23(a) and Figure 23(b), it can be observed that as the position (%) increases along the x-axis, the time it takes for the system to generate a trip signal based on the conventional approach also increases, but the time it takes for the system to generate a trip signal based on the hybrid approach is comparatively very short, thereby increasing the reliability of the solution.

[0137] Based on the operating times obtained from both conventional and hybrid methods, the robustness of the system using the two methods is shown in Table 8.

[0138] [Table 8]

[0139] It is observed that when the fault location increases beyond one-third of the transmission line length, the conventional approach does not give 100% accurate results, whereas the hybrid approach gives 100% reliable results.

[0140] FIG. 24(a) and FIG. 24(b) show typical transmission line operation times for determining system reliability with the conventional and hybrid approaches in yet another example according to an embodiment of the present subject matter. A 200 km single circuit transmission line was considered. FIG. 24(a) and FIG. 24(b) show plots of operation times obtained with the conventional time domain approach and the hybrid approach. In FIG. 24(a), the location of the fault is plotted against the average trip time measured in milliseconds. Line 2402 represents the operation time plot obtained for the conventional time domain based approach, and line 2404 represents the operation time plot obtained for the hybrid approach. In FIG. 24(b), the location of the fault is plotted against the maximum trip time measured in milliseconds. Line 2406 represents the operation time plot obtained for the conventional time domain based approach, and line 2408 represents the operation time plot obtained for the hybrid approach. From Figure 24(a) and Figure 24(b), it can be observed that as the position (%) increases along the x-axis, the time it takes for the system to generate a trip signal based on the conventional approach also increases, but the time it takes for the system to generate a trip signal based on the hybrid approach is comparatively very short, thereby increasing the reliability of the system.

[0141] Based on the operating times obtained from both the conventional and hybrid methods, the robustness of the system using the two methods is shown in Table 9.

[0142] [Table 9]

[0143] It is observed that when the fault location increases beyond one-third of the transmission line length, the conventional approach does not give 100% accurate results, whereas the hybrid approach gives 100% reliable results.

[0144] FIG. 25(a) and FIG. 25(b) show the operation time of a dual circuit transmission line for determining the reliability of the system according to the conventional approach and the hybrid approach in yet another example according to an embodiment of the present subject matter. A 200 km dual circuit transmission line was considered. FIG. 25(a) and FIG. 25(b) show plots of the operation time obtained with the conventional time domain approach and the hybrid approach. In FIG. 25(a), the location of the fault is plotted against the maximum trip time measured in milliseconds. Line 2502 represents the operation time plot obtained for the conventional time domain based approach, and line 2504 represents the operation time plot obtained for the hybrid approach. In FIG. 25(b), the location of the fault is plotted against the average trip time measured in milliseconds. Line 2506 represents the operation time plot obtained for the conventional time domain based approach, and line 2508 represents the operation time plot obtained for the hybrid approach. From Figure 25(a) and Figure 25(b), it can be observed that as the position (%) increases along the x-axis, the time it takes for the system to generate a trip signal based on the conventional approach also increases, but the time it takes for the system to generate a trip signal based on the hybrid approach is comparatively very short, thereby increasing the reliability of the system.

[0145] Based on the operating times obtained from both conventional and hybrid methods, the robustness of the system using the two methods is shown in Table 10.

[0146] [Table 10]

[0147] It is observed that when the fault location increases beyond one-third of the transmission line length, the conventional approach does not give 100% accurate results, whereas the hybrid approach gives 100% reliable results.

[0148] FIG. 26(a) and FIG. 26(b) show the operation time of a series compensated line for determining the system reliability with the conventional approach and the hybrid approach in yet another example according to an embodiment of the present subject matter. A 200 km series compensated line was considered. FIG. 26(a) and FIG. 26(b) show plots of operation time obtained with the conventional time domain approach and the hybrid approach. In FIG. 26(a), the location of the fault is plotted against the average trip time measured in milliseconds. Line 2602 represents the operation time plot obtained for the conventional time domain based approach, and line 2604 represents the operation time plot obtained for the hybrid approach. In FIG. 26(b), the location of the fault is plotted against the maximum trip time measured in milliseconds. Line 2606 represents the operation time plot obtained for the conventional time domain based approach, and line 2608 represents the operation time plot obtained for the hybrid approach. From Figure 26(a) and Figure 26(b), it can be observed that as the position (%) increases along the x-axis, the time it takes for the system to generate a trip signal based on the conventional approach also increases, but the time it takes for the system to generate a trip signal based on the hybrid approach is comparatively very short, thereby increasing the reliability of the system.

[0149] Based on the operating times obtained from both the conventional and hybrid methods, the robustness of the system using the two methods is shown in Table 11.

[0150] [Table 11]

[0151] It is observed that when the fault location increases beyond one-third of the transmission line length, the conventional approach does not give 100% accurate results, whereas the hybrid approach gives 100% reliable results.

[0152] Thus, the present subject matter provides an accurate method for classifying faults and identifying the zone of a transmission line in which the fault is located. The method is robust to fault type, location and fault resistance. Thus, for inverter-based resources or conventional sources, the present subject matter provides high reliability while improving speed and safety.

[0153] Although the present subject matter has been described with reference to particular embodiments, this description is not meant to be construed in a limiting sense. Various modifications of the disclosed embodiments, as well as alternative embodiments of the subject matter, will become apparent to those skilled in the art upon reference to the description of the subject matter.

Claims

1. 1. A method for fault zone identification, comprising: identifying a zone of the transmission line in which the fault is present based on the traveling wave signal if the difference between the calculated line length and the actual line length is less than the threshold line length; identifying the zone of the transmission line in which the fault exists based on a rate of change of incremental terminal current with respect to time if the difference between the calculated line length and the actual line length is greater than the threshold line length; A method comprising:

2. receiving an identification of the transmission line having the fault; calculating the line length of the transmission line based on a traveling wave signal obtained from a mode conversion performed on voltage or current measurements determined at the terminals of the transmission line during the fault to obtain the calculated line length; The method of claim 1 , comprising:

3. 3. The method of claim 2, wherein the zone identified based on the traveling wave signal is within one-third to two-thirds of the actual line length from the terminal, and the zone identified based on the rate of change of incremental terminal current is within one-third of the actual line length from the terminal.

4. 1. A method for fault classification, the method comprising: obtaining voltage or current measurements at the terminals of the transmission line in each of the three phases measured during the fault; performing a mode conversion on the voltage or current measurements to obtain a traveling wave signal for each of the three phases; classifying the fault based on a magnitude of the traveling wave signal; Including, the traveling wave signals include a zero mode traveling wave signal for each of the three phases; The method includes comparing a magnitude of the zero mode traveling wave signal to a zero mode threshold to distinguish ground faults from non-ground faults.

5. The method of claim 4 , wherein classifying the fault is performed without using an incremental amount of the voltage or current measurements with respect to time.

6. The method according to claim 4 or 5, wherein the mode transformation is a Clarke transformation.

7. The method of claim 4 including generating a trip signal to cause a single phase trip or a three phase trip based on the classification of the fault.

8. The traveling wave signals further include a beta mode traveling wave signal for each of the three phases, and the method further comprises:

5. The method of claim 4, further comprising: if the fault is a ground fault, classifying the fault as involving a first of the three phases and ground when a first condition is met, the first condition being met when a magnitude of a beta traveling wave signal for the first phase is the minimum of the magnitudes of the beta traveling wave signals of each of the three phases and is less than a beta mode threshold.

9. the traveling wave signals further include an alpha mode traveling wave signal for each of the three phases; If the fault is a ground fault, the method includes: classifying the fault as involving two of the three phases and ground when a second condition is met, the second condition being met when an alpha traveling wave signal magnitude for a third of the three phases is the minimum of the magnitudes of the alpha traveling wave signals for each of the three phases; classifying the fault as involving the three phases and the ground when the second condition is not met; If the fault is a non-earth fault, classifying the fault as involving two of the three phases when a third condition is met, the third condition being met when the magnitude of an alpha traveling wave signal for a third of the three phases is the minimum of the magnitudes of the alpha traveling wave signals of each of the three phases and is less than an alpha mode threshold; classifying the fault as involving the three phases when the third condition is not met; and The method of claim 4 or claim 8, comprising:

10. The method according to any one of claims 4 to 9, wherein the terminals are connected to a synchronous generation resource or a grid-connected transmission line or an inverter-based generation resource.

11. A method for fault classification and zone identification comprising a method for zone identification according to any one of claims 1 to 3 and a method for fault classification according to any one of claims 4 to 10.

12. An intelligent electronic device comprising a processor configured to carry out the method according to any one of claims 1 to 3 and / or the method according to any one of claims 4 to 10.

13. A non-transitory computer readable medium comprising instructions which, when executed by a processor, cause the processor to perform a method for zone identification according to any one of claims 1 to 3 and / or a method for fault classification according to any one of claims 4 to 10.

Images