Method and system for detection of faults in electric grid

Abstract

Disclosed is a method (700) and a system (600) for detecting a traveling wave in an electric grid (100). The method comprises determining that a traveling wave is propagating in a transmission line (102) of the electric grid based on a set of amplitudes and a set of arrival times received from a set of sensors (104a-104c, 104a'-104c') in the electric grid and determining that a cause of propagation of the traveling wave through the transmission line is an occurrence of a lightning strike on the transmission line or near the transmission line. Additionally, a location (106) of occurrence of the lightning strike is determined based on a set of locations (108a-108c, 108a'-108c') of the set of sensors and parameters of the traveling wave at the set of locations. One or more locations of occurrence of same fault in the electric grid is determined based on the determined location.

Description

[0001] METHOD AND SYSTEM FOR. DETECTION OF FAULTS IN ELECTRIC GRID

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to a method and a system for detection of faults in an electric grid. Moreover, the present disclosure relates to a method and a system for detection of faults in an electric grid due to lightning strikes.

[0004] BACKGROUND

[0005] An electric grid may include components such as power lines with phase wires, cables, power poles, transformers, switching circuits, protection circuits, substations, and so on, to enable electricity transmission over long distances from electricity producers to electricity consumers. The electric grid is prone to faults which may occur due to a natural cause such as lightning strikes and wind, or objects, such as trees, falling on transmission lines of the electric grid. The electric grid is also prone to faults caused by electric grid equipment failure. Occurrence of such fault(s) may create conditions such as overcurrent, undervoltage, unbalancing of three phases, or high voltage surges. Examples of the faults include, but are not limited to, transient faults, earthing faults, arcing faults, short circuit faults, open circuit faults, overload faults, broken conductors, lost phases, and partial discharges. Many of the faults occurring in the electric grid are transient in nature.

[0006] It has been observed that lightning strikes are the cause of occurrences of many of the abovementioned faults in the electric grids. Furthermore, lightning strikes contribute significantly to the count of faults that occur in overhead lines of the electric grids over a period. The lightning strikes on a transmission line may involve several stages. The lightning typically begins with a stepped leader, which is a series of electrical discharges from a cloud towards the ground. When this leader gets close to ground, a connection is established, and a return stroke follows. The return stroke is generally a bright visible flash that is associated with lightning. When a lightning bolt directly strikes a power line, conductive materials in the power line provide a low-resistance path for electrical energy to flow via the low-resistance path. The lightning is accompanied by generation of an intense heat that causes air around a strike point (i.e., a point on the power line) to rapidly expand, thereby creating a shock wave. The shock wave is an audible wave that is referred to as thunder. The lightning strike can introduce a surge of extremely high voltage and current into the electrical distribution system. This surge can travel along the power line, thereby affecting everything that is connected to a network including the power line.

[0007] The electric grid (i.e., transmission / distribution systems) is designed to include protective devices which include, but may not be limited to, surge arrestors, fuses, and static / shield wires. The protective devices are strategically placed to divert or absorb any excess electrical energy that is introduced in the electric grid due to a lightning strike. The protective devices protect equipment and appliances that are connected to or a part of the electric grid. However, strategically placing the protective devices, selecting appropriate protective devices, and determining an optimum number of protective devices required to be installed is challenging. Grounding also plays a critical role in dissipating the excess electrical energy, as for example high grounding (earthing) impedance of surge arresters may allow the lightning induced surge pulse to flash over an insulator, directing the energy to a grid phase conductor, which is unwanted. Equipment of the electric grid such as transformers, electric poles, and other components of a distribution system are grounded to provide a path for lightning-induced currents to flow into the ground. Thus, proper grounding may be essential for minimizing an impact of lightning strikes on reliability and safety of the electric grid. However, implementing a sufficient number of grounding electrodes is challenging in certain scenarios such as non-conductive ground (for example, certain types of soils with low conductivity, rocks, and sand).

[0008] After a lightning strike, it is often necessary to assess damage caused to different portions of the electric grid (such as the distribution system). The damage assessment involves inspecting power lines, transformers, and other infrastructure, for identification of components that may be requiring repair or replacement. The actual effects of a lightning strike on an electrical distribution network or transmission network can vary based on factors such as the intensity of the strike, the design of the distribution or transmission network, and the effectiveness of protective measures in place. Thus, carrying out damage assessment may be challenging in terms of ensuring efficiency in locating faults in the electric grid within a specific period, especially if the electric grid is large (i.e., covers a larger area). Furthermore, effective damage assessment requires expending significant financial resources for manual assessment or automatic assessment using specialized equipment.

[0009] During a thunderstorm, and after, it is important, or at minimum very useful, to know if the protection devices of the electrical grid operated due to a lightning strike, or if the fault was caused by a local event, such as fallen tree, a car hitting a pole, a broken grid component, etc. Knowing this allows the grid operator to plan the inspection and service work needed for this particular grid fault event.

[0010] A grid utility may obtain lightning strike information from various paid or free weather or lightning data sources. However, these data sources, even the best ones, typically have several drawbacks. First, the lightning location accuracy claimed by the lighting data providers is very often not up to the promise, the lightning strike locations often missing the real strike location at the electrical grid. Secondly, the lightning data providers very often miss a lightning strike, or thirdly, the lightning data providers may claim the lightning strike being a cloud-to-cloud strike instead of cloud-to-ground strike, while at the electrical grid level it is clear, in hindsight, that the lightning strike really hit the ground or the components of the electrical line. A fourth drawback is that connecting (correlating) the grid event information to a lightning strike information from the lightning data provider, is often difficult, requiring manual inspection, which especially during a storm is tedious and time consuming, deeming the process of combining the two in practice impossible to execute properly.

[0011] Therefore, in light of the foregoing discussion, there exists a need to overcome the aforementioned drawbacks by determining whether faults in an electric grid are caused by lightning strikes, in addition to locating the lightning strikes.

[0012] SUMMARY

[0013] The present disclosure relates to the detection of traveling waves in electrical grids and determining their source, with a particular focus on differentiating lightning-induced traveling waves from other events. The method uses data collected from sensors placed strategically across the grid, to locations, such as electric poles, towers, and / or substations, to analyze signal characteristics and signal propagation patterns. The detailed description of embodiments provides clarity on how the claims are implemented and highlights practical considerations in deploying the method in real-world grid systems.

[0014] As the modern electrical grids have typically effective protection against lightnings, when a lightning impacts the electrical grid, it is important to the grid operators to understand first if the event detected was caused by a lightning strike or not, as most often the lightning events can cause temporary outage as the protection elements on the grid operate, but there is no permanent damage or fault, and typically the grid automation restores the power, often within one second, and with mesh-connected transmission lines often within a few power frequency cycles.

[0015] The aim of the present disclosure is to provide a method and a system for detecting traveling waves in an electric grid to determine the origin, nature, and propagation of the waves. The aim of the disclosure is achieved by a method and a system for detecting traveling waves in an electric grid based on signal data comprising amplitudes and arrival times received from sensors located in the electric grid, as defined in the appended independent claims to which reference is made.

[0016] The embodiments of the present disclosure substantially enable to improve the detection, categorization, and localization of traveling waves in electric grids by distinguishing between locally induced disturbances and atmospheric events such as lightning strikes. The disclosed embodiments provide robust signal correlation techniques that analyze amplitudes, arrival times, and waveform characteristics to enhance the accuracy of fault detection and location determination.

[0017] Additional aspects, advantages, features, and objects of the present disclosure would be made apparent from the drawings and the detailed description of the illustrative embodiments constructed in conjunction with the appended claims that follow.

[0018] Throughout the description and claims of this specification, the words "comprise" , "include", "have", and "contain" and variations of these words, for example "comprising" and "comprises" , mean "including but not limited to" , and do not exclude other components, items, integers, or steps not explicitly disclosed also to be present. Moreover, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] FIG. 1A illustrates two exemplary portions of an electric grid which includes an arrangement for determining occurrences of an event or a fault in the electric grid, in accordance with various embodiments of the present disclosure;

[0020] FIGS. 1B-1D illustrate by way of examples propagation paths through which a fault current signal can propagate in the electric grid, with reference to the same electric grid as illustrated in FIG. 1A;

[0021] FIG. 2 illustrates an exemplary determination of location of occurrence of the lightning strike in the electric grid, in accordance with various embodiments of the present disclosure;

[0022] FIG. 3 illustrates a first set of graphs that indicate a variation of amplitude of a traveling wave detected by a set of sensors in an electric grid, in accordance with an embodiment of the present disclosure;

[0023] FIG. 4 illustrates a second set of graphs that indicative of a variation of amplitude of a traveling wave detected by a set of sensors in an electric grid, in accordance with an embodiment of the present disclosure;

[0024] FIG. 5 depicts an exemplary user interface of a map application that indicates a location of occurrence of a lightning strike, in accordance with an embodiment of the present disclosure;

[0025] FIG. 6 illustrates a block diagram of a system for detecting traveling waves in an electric grid, in accordance with an embodiment of the present disclosure;

[0026] FIGS. 7A and 7B illustrate a flowchart of a method for detecting a traveling wave in an electric grid, in accordance with an embodiment of the present disclosure; FIG. 8 illustrates an example of a lightning induced fault event triggered recording at two frequency bands;

[0027] FIG. 9 illustrates an exemplary electric grid on a geographical map, having an arrangement for determining occurrences of an event or a fault in the electric grid, in accordance with various embodiments of the present disclosure.

[0028] DETAILED DESCRIPTION OF EMBODIMENTS

[0029] The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practicing the present disclosure are also possible.

[0030] In a first aspect, an embodiment of the present disclosure provides a method, implemented in a processor, for detecting a traveling wave in an electric grid, the method comprising: receiving a set of signal data, comprising a set of amplitudes and a set of arrival times, from a set of sensors, associated with a transmission line or a distribution line of the electric grid, the set of sensors comprising a first set of sensors, positioned at locations in a first section of the electric grid, and a second set of sensors, positioned at locations in a second section of the electric grid; determining the presence of a traveling wave propagating along the transmission line or the distribution line of the electric grid based on the received signal data; categorizing the traveling wave as either a local event or an atmospheric event, such as a lightning strike, wherein the categorizing comprising: recording waveforms of the detected traveling waves; analyzing whether the detected traveling wave exhibits similar amplitude sequences and arrival times across the set of sensors; distinguishing between a locally detectable traveling wave, where the traveling wave is detected either by any or several of the sensors in the first set of sensors or any or several of the sensors in the second set of sensors, and / or whether a traveling wave is detectable by any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors; and identifying the source of the traveling wave as a local grid event if the traveling wave is detected only with any or several of the sensors in the first set of sensors or only with any or several of the sensors in the second set of sensors; and identifying the source as an atmospheric event, if the traveling wave is detected by both any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors.

[0031] The set of sensors is positioned in different sections of the electric grid, such as two distinct areas separated by step-up or step-down transformers. Preferably the mentioned two different sections of the electrical grid are also isolated spatially so that there is minimal possibility for cross-induction of the traveling wave signals from one section of the grid to the other section of the grid. These sensors detect transient signals resulting from disturbances in the electric grid. Each sensor is capable of recording the arrival time of the traveling wave and its amplitude. For instance, when a lightning strike occurs, the sensors across multiple sections detect the high-frequency traveling wave almost simultaneously, whereas a local event, like a grid fault, is detected only by sensors in proximity, typically only by the sensors in a single separate section of a grid, as explained above.

[0032] The method involves analyzing the amplitude sequences and arrival times to determine whether the disturbance is local or atmospheric. In accordance with an embodiment analyzing whether the detected traveling wave exhibits similar amplitude sequences and arrival times across the set of sensors is executed within a predefined temporal threshold. A predefined temporal threshold accounts for time measurement errors, which depend on time measurement technology used, but when using GPS (Global Positioning System) / GNSS (Global Navigation Satellite System) time the timing error is typically negligible, as it is relatively easy to achieve, e.g., 1 microsecond level or better timing accuracy, corresponding to 300 meters distance at speed of light, and close to the speed of light velocity in overhead lines, typically. The signal propagation velocity from the event point (a local grid event, or an atmospheric event, such as lightning) to the point of detection is at minimum the delay of speed of light, and maximally follows coaxial cable path along the grid lines, the signal velocity along coaxial cables being typically 0.5 up to 0.8 times the speed of light, and in overhead lines typically 0.95 to 0.999 times the speed of light. By mathematically correlating waveforms, including filtered versions to reduce noise, wave fronts detected, and / or signal amplitude envelopes detected, the processor distinguishes whether the traveling wave originated locally or propagated across a larger area. Notably, correlating high-frequency transient signals to determine if they originate from the same source such as a lightning strike, does not necessarily require very accurate time synchronization between the recorded signals by sensors, as the event signals, especially lightning strike signals, have distinct and unique waveform amplitude or envelope "fingerprint", with which an event can be identified. For correlating, using multiple possible correlation methods, the signals for the purpose of determining a local or an atmospheric event, it is possible to select frequency band for signal recordings from a wide area of frequencies, ranging from 3 or 10 kHz, up to 1 GHz and even above, depending on the sensor design. As an example, frequency bands from 50 kHz to 50 MHz, or a narrower frequency band 100 kHz to 1MHz perform rather well for this purpose. Also, one could estimate correlations per frequency band, or using Fourier or wavelet transforms, or any combination of the mentioned. The sampling rate for digitalizing the signals before correlation can be also selected rather freely, and down sampling before correlation operation may be preferred for noise reduction and reduced processor performance requirements. For example, receiving signal frequencies between 100 kHz and 5 MHz, sampling at 1 MHz, calculating signal amplitude envelope every 1 milliseconds, or demodulating the signal, and correlating the obtained signals for a duration of 300 milliseconds, or from 30 milliseconds to 1 second, may provide good performance in detecting if the signals received by two different sensors originate from the same signal source or not, even if the time synchronization would be as poor as 1 second level or worse. Therefore, to an expert in art, it is obvious that selecting the received signal bandwidth, sampling frequency, filtering and down sampling method, and clock synchronization method for the purposes of is merely an implementation detail which the expert in art may select and optimize rather freely for the purposes of determining if the event was of local or atmospheric nature. Also, based on these design parameters selected, the required correlation amount between event recordings between sensors to determine that the signals had the same origin event, is an implementation design parameter, which value depends on these implementation details chosen and other system performance requirements and parameters, and the true and false positive, and true and false negative indication requirements set by the expert in art implementing the method into an operable system.

[0033] The above mentioned direct or manipulated signals are referred to as signal amplitude sequences in this invention.

[0034] In addition, knowing the signal reception to signal recording transfer function, even at a rough accuracy, gives the additional benefit of being able to estimate roughly the area of origin of an atmospheric signal, as the signal amplitudes will decrease along the path of the signal, and also the highest frequency components of the signal attenuate more per distance compared to lower frequency signals.

[0035] The method provides an accurate mechanism to distinguish between local grid faults and lightning-induced traveling waves. By correlating sensor data and analyzing waveform characteristics, the method can identify the nature of disturbances and locate their sources efficiently, and the system incorporating the method can inform the grid operators about the nature and location of the event, so that adequate inspection and service actions can be planned and initiated. This enhances grid reliability, minimizes downtime, and ensures appropriate responses to both local and atmospheric events.

[0036] In a second aspect, the present disclosure provides a system for detecting a traveling wave in an electric grid, the system comprising: a set of sensors, the set of sensors comprising a first set of sensors configured to be positioned at locations in a first section of the electric grid; and a second set of sensors configured to be positioned at locations in a second section of the electric grid; wherein each sensor of the set of sensors is configured to monitor one or more phase conductors and / or neutral conductors and / or ground conductors of a transmission line or a distribution line of the electric grid and detect a set of signal data, comprising a set of amplitudes and a set of arrival times; a processor configured to be associated with the set of sensors, wherein the processor is configured to: receive the set of signal data from the set of sensors; determine the presence of a traveling wave propagating along a transmission or distribution line of the electric grid based on the received signal data; categorize the traveling wave as either a local event, or an atmospheric event, such as a lightning strike; record waveforms of the detected traveling waves; analyze whether the detected traveling wave exhibits similar amplitude sequences and arrival times across the first set of sensors and the second set of sensors; distinguish between a locally detectable traveling wave, where the traveling wave is detected either by any or several of the sensors in the first set of sensors or any or several of the sensors in the second set of sensors, and / or whether a traveling wave is detectable by any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors, and identify the source of the traveling wave as a local grid event if the traveling wave is detected only by any or several of the sensors in the first set of sensors or only by any or several of the sensors in the second set of sensors, and identify an atmospheric event, such as a lightning strike, if the traveling wave is detected by both any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors.

[0037] The system includes a distributed arrangement of sensors that monitor specific portions of the electric grid. The sensors are located in grid sections that may be separated by transformers or other galvanic isolation points. Each sensor is capable of capturing critical parameters of traveling waves, such as amplitudes and arrival times, and transmits this data to the processor. The processor analyzes the sensor data to identify and characterize the traveling wave.

[0038] The processor performs multiple analyses, including waveform recording, amplitude and arrival time correlation, and mathematical filtering, to determine whether the traveling wave originated from a local event or an atmospheric disturbance. For instance, if sensors in both isolated grid sections simultaneously detect the wave, the event is categorized as atmospheric (e.g., lightning strike). Conversely, if the wave is detected only by sensors in one section, it is classified as a local grid fault.

[0039] The system provides an integrated and automated solution for detecting, categorizing, and localizing traveling waves in electric grids. By leveraging sensor networks and advanced signal processing capabilities, the system enables accurate fault detection, distinction between lightning-induced and local disturbances, and real-time decision-making for grid operators. This ensures enhanced grid stability, reliability, and operational efficiency.

[0040] In a third aspect, the present disclosure provides a computer program product, downloadable and installable on a computer, for detecting a traveling wave in an electric grid according to the method provided by the first aspect.

[0041] The present disclosure provides the aforementioned first aspect and the aforementioned second aspect to facilitate detection of a traveling wave in a transmission line, or in a distribution line, of an electric grid, and determining that a cause of the detection of the traveling wave (i.e., propagation of traveling wave) on the transmission line, or on the distribution line, is a lightning strike directly on said line or in the vicinity of said line.

[0042] The present disclosure further provides that once it is determined that the lightning strike is the cause of detection of the traveling wave, a strike location, where the lightning had actually struck, may be determined. Additionally, the determination that the lightning strike is cause of detection of the traveling wave in the transmission line, or in the distribution line, facilitates distinguishing faults in the electric grid that have been caused by lightning strikes and faults that are caused by other factors.

[0043] The method begins by detecting the propagation of a traveling wave in an electric grid. Sensors distributed across the electric grid capture wave signals, comprising a set of amplitudes and a set of arrival times. A processor receives the data from the sensors and analyzes the data to determine if a traveling wave is present. The method further categorizes the traveling wave based on its origin. By analyzing whether the wave signals are detectable both locally and at a long distance, the system identifies whether the event was localized (e.g., a fault or switching event within the electric grid) or atmospheric (e.g., a lightning strike). This determination leverages the physical principle that lightning-induced traveling waves exhibit distinct propagation patterns, being detectable across larger grid areas.

[0044] To determine the locality of a traveling wave involves analyzing the spatial distribution of sensors detecting the event. The signals are considered local if all the detecting sensors are confined to a defined local area (segments or sections) within the electric grid. This area may be predefined based on operational grid parameters and / or historical data, i.e., an information that has been detected and / or collected over time.

[0045] A traveling wave signal is identified as local if no sensors, among the set of sensors, is outside a predefined maximum distance to detect the traveling wave. The distance threshold is set to reflect the physical limitations of signal propagation within the electric grid. For example, local grid disturbances have propagation characteristics that prevent them from traveling beyond a certain range, unlike lightning strikes that produce waves detectable over much larger areas.

[0046] The determination considers both the galvanic connectivity of the electric grid and the distinct (isolated) electric grids. If all sensors detecting the signal are on sections of the electric grid that are electrically connected without being separated by step-up or step-down transformers or open switches (apart from neutral or ground connections), the signal is likely local. This consideration acknowledges that certain devices, such as step- up or step-down transformers, isolate sections of the grid from higher frequency signals, well above the mains frequency of the electric grid, traveling, limiting signal propagation. Thus, if all sensors detecting the signal are on sections of the electric grid that sections are electrically connected without being separated by transformers or open switches, for example, (apart from neutral or ground connections), the signal is likely caused by a local grid event, i.e., being an indicative of a local fault.

[0047] After identifying the source of the traveling wave, the strike location and, consequently one or more locations of occurrences of the same event (which is caused by the lightning strike) may be determined. The strike location may be determined based on the set of arrival times and the set of amplitudes if the strike location is on the transmission line. On the other hand, if the strike location is in the vicinity of the transmission line, the strike location may be determined based on locations of the set of traveling wave sensors. A user interface, operatively connected to the processor, may display the strike location and a location of the fault caused by the lightning strike. The location of the fault in the electric grid is determined based on the determined strike location.

[0048] The term "electric grid" as used herein refers to a type of interconnected electrical network that is operable to deliver electricity to (or from) the connected elements therein via a plurality of associated transmission lines or distribution lines. The electric grid may consist of power plants, the high-voltage distribution grid, and the local distribution grid to which the electricity users are connected. The electric grid may be composed, for example, of 400 kV (kilovolt) or higher transmission lines, 220 kV transmission lines, and / or 110 kV transmission lines, and / or substations, and for example any of lkV to 100 kV medium voltage distribution lines and below 1 kV low voltage distribution lines. The electric grid comprises a set of traveling wave sensors coupled to at least one event recorder that monitors (or records) transient events (such as occurrence of lightning strikes) in the electric grid. The electric grid is a part of an electrical grid network which distributes electrical power from power generation stations and sites to distribution substation(s) to premises (such as homes, offices, factories, and the like) of consumers.

[0049] The terms "distribution grid" and "transmission grid", or "distribution lines" and "transmission lines", referring to systems transmitting and distributing electrical power and energy, are used herein interchangeably.

[0050] The term "traveling wave sensor" refers to a sensor with the purpose of measuring voltage and / or current information of an electrical grid or line, which are capable of detecting information of higher frequency content above the mains frequency and also can time stamp the recorded information with an accuracy of at least 5 seconds, preferably better than 100 milliseconds, for correlating grid events measured by multiple sensors, and optimally the time stamping accuracy is better than 10 microseconds for locating the events in accuracy of kilometers, and 1 us for location accuracy range of 100 meters and even better.

[0051] A lightning strike refers to the overall event of lightning hitting a location, object, or ground. A lightning stroke refers to a single discharge of electrical energy as part of a lightning flash. It is a component of a complete lightning event. The lightning strokes of the same lightning strike typically hit the same place or follow nearly the same path. The lightning strikes and lightning strokes are used herein mostly interchangeably. However, it is possible, according to the invention, to detect, identify and locate each lightning stroke separately, as well as to provide a more generic, often average indication and location of the lightning strokes as a lightning strike. The following description further illustrates embodiments of the present disclosure and ways in which they can be implemented.

[0052] Initially, the method determines that a traveling wave is propagating in a transmission line of the electric grid. The determination of propagation of the traveling wave in the transmission line is based on one or more inputs, i.e., a set of signal data, comprising a set of amplitudes and a set of arrival times, received from a set of sensors in the electric grid. The set of sensors are operable to detect traveling waves and monitor, in the transmission line, one or more of a current level, a voltage level, an electromagnetic field, an electric field, a magnetic field, and so on. The monitoring by the set of sensors involves periodically measuring current level, voltage level, electric field strength, magnetic field strength, power level, and so on, on the transmission line. Such monitoring may facilitate detection of faults in the electric grid. Additionally, the set of sensors are operable to monitor equipment associated with the transmission network or the distribution network of the electric grid.

[0053] In accordance with an embodiment, each sensor of the set of sensors may detect a transient event during the monitoring. The transient event may be detected due to an occurrence of a grid event, a fault event, a partial discharge event, or a lightning strike in or near the electric grid. The detection of the transient event involves detecting a transient signal such as a high-frequency traveling wave. Each sensor of the set of sensors that detect the transient signal, convert the detected transient signal into an electrical signal, determine the set of inputs, the set of signal data, from the electrical signal, and transmit the set of inputs to facilitate determining that a traveling wave is propagating in the transmission line. The set of sensors may be synchronized using Global Positioning System (GPS) or Global Navigation Satellite System (GNSS), or even using less accurate time synchronization methods. The synchronization allows all sensors of the set of sensors, at different locations of the electric grid, to simultaneously detect the high-frequency traveling wave (i.e., the transient signal).

[0054] In accordance with an embodiment, determining that a traveling wave is propagating in the transmission line comprises analyzing variations in the electromagnetic field surrounding the transmission line, or distribution line, detected by any or several of the sensors in the set of sensors. The line may also have cabled, typically underground portions, and as the signals propagate from above ground sections to cable sections, the signals may be detected using voltage transformers, capacitive or resistive dividers, current transformers, Rogowski coils, or other inductive or capacitive or magnetic field sensors. The sensors are capable of detecting changes in the electromagnetic field, which are indicative of traveling waves caused by disturbances, such as faults or lightning strikes, propagating along the transmission or distribution line. For example, when a transient event occurs, the associated high-frequency electromagnetic impulses generate variations in the surrounding field. These variations are detected by the sensors. The sensors may be inductive sensors, Hall-effect sensors, or Rogowski coils, which convert the electromagnetic signals into measurable data. The variations in the magnetic, electric, or electromagnetic field or voltage or current, help validate the presence of a traveling wave, complementing the analysis of amplitude and arrival times. This approach enhances the reliability of detecting disturbances, as electromagnetic field analysis provides additional data to verify the traveling wave's origin and characteristics. The inclusion of electromagnetic field analysis improves the accuracy and robustness of the method. By leveraging multiple detection parameters (signal amplitudes, arrival times, and electromagnetic field variations), the system ensures a comprehensive and reliable identification of traveling waves, reducing the risk of false positives or undetected events. In accordance with an embodiment, analyzing comprises determining whether the detected traveling wave exhibits consistent amplitude sequences and arrival times across the first set of sensors and / or the second set of sensors, within a predefined amplitude tolerance range established based on signal variability. In this embodiment, the processor analyzes the amplitude values and arrival times of the traveling wave signals recorded, or detected, by the sensors in the first and second sets. Specifically, the analysis involves verifying whether the amplitude sequences and arrival times are within a predefined tolerance range. The amplitude tolerance range is determined based on factors such as grid configuration, sensor synchronization accuracy, and known variability of traveling wave signals. For example, if a lightning strike occurs, the amplitude sequences and arrival times will be similar across sensors in both the first and second sections of the grid, within the tolerance range. In contrast, a local grid event will exhibit significant amplitude discrepancies between sensors located in different sections, as the wave does not propagate as widely. The tolerance range ensures that minor variations caused by signal noise or environmental factors do not affect the analysis. This embodiment improves the precision of wave categorization by accounting for allowable variations in signal amplitudes and arrival times. By introducing a predefined amplitude tolerance range, the method effectively distinguishes between local disturbances and atmospheric events, ensuring accurate analysis even in the presence of minor signal inconsistencies.

[0055] In accordance with an embodiment, determining if the traveling wave was detectable only locally involves verifying that all sensors among the set of sensors detecting the signal and transmitting the corresponding signal data are located within a defined local area of the electrical grid. In this embodiment, the processor determines whether the traveling wave was of a local origin, and not an atmospheric event by analyzing the spatial distribution of sensors detecting the signal. Specifically, the sensors are located within a predefined local area, which can be defined as a section of the grid characterized by physical and operational constraints, such as a substation or an interconnected line segment (see, e.g., FIG. 1A). For example, this predefined area may span 5-20 kilometers, depending on the grid topology and signal transmission characteristics. If all sensors detecting the wave are confined within the predefined local area, the processor concludes that the disturbance is a local grid event. Conversely, if sensors outside the predefined area also detect the wave, the event is categorized as a more widespread disturbance, such as an atmospheric event, most probably a lightning strike. The processor uses precise sensor location data and arrival times to determine the geographic extent of the traveling wave's detection. This embodiment provides a clear and reliable method for determining the locality of disturbances. By restricting analysis to sensors within a predefined local area, the system effectively distinguishes localized events and faults or grid disturbances from broader atmospheric events. This improves operational decision-making and response times, enabling targeted maintenance and repairs.

[0056] In accordance with an embodiment, determining if the traveling wave signal was detectable only locally is based on the sensors detecting the signal being positioned in a section of the electrical grid that is galvanically connected, excluding sections separated by step-up or stepdown transformers or open switches. In this embodiment, the determination of whether a traveling wave is localized relies on analyzing the galvanic connectivity of the electric grid. A section of the grid is considered "galvanically connected" when there is continuous electrical connectivity between sensors without significant impedance changes caused, e.g., by step-up or step-down transformers or open switches. For example, if a disturbance occurs within a section of the grid that is galvanically connected to one or more other sections of the electric grid, the traveling wave is capable of propagating seamlessly across the sensors positioned within those sections. However, if transformers or switches isolate sections of the grid, the wave cannot propagate beyond these isolations. Thus, if the sensors detecting the wave are all positioned within a galvanically connected segment / section of the electric grid, the disturbance is identified as local. The processor determines this by correlating sensor locations with grid topology data, including transformer and switch positions. This embodiment enhances the accuracy of detecting localized disturbances by leveraging the physical and electrical characteristics of the grid. By excluding signals that cross galvanically isolated sections, the method minimizes misclassification of disturbances, ensuring precise differentiation between local and broader events.

[0057] The benefits of the above-mentioned galvanic isolation may be ruined due to mutual induction or capacitive coupling, if the electrical lines of the two galvanically isolated sections of a grid cross each other, or travel in parallel at close proximity, for example in the same trenches underground, or on the same utility poles, or at close proximity, for example along the same street. Improved performance of the local event I atmospheric event detection can be achieved if the two sensor groups are also in separate geographical areas, in addition to being galvanically isolated.

[0058] In accordance with an embodiment, determining if the traveling wave signal was detectable only locally is based on the proximity of all detecting sensors among the set of sensors, ensuring no sensor is located beyond a predefined maximum distance, determined based on physical and operational constraints of the electric grid. In this embodiment, the determination of locality is performed by analyzing the spatial proximity of the sensors that detect the traveling wave. Specifically, the method ensures that all detecting sensors are positioned within a predefined maximum distance from the disturbance point. This maximum distance is established based on the physical properties of the electric grid, such as conductor length, signal traveling speed of traveling waves, and operational constraints. For example, if the maximum distance is defined as 50 kilometers, the processor checks the positions of all detecting sensors. If no sensor lies beyond this distance threshold, the traveling wave is classified as a local event. In contrast, if sensors outside this range also detect the wave, the event is considered atmospheric or widespread. The processor uses location data from the sensors and time- stamped signal arrival data to perform this analysis. This embodiment provides an additional layer of accuracy for identifying local disturbances by introducing a strict physical boundary. By ensuring that detecting sensors remain within a predefined distance, the method effectively filters out non-local events, such as lightning strikes, which produce detectable traveling waves over larger areas. This improves the system's precision in categorizing grid events and locating faults.

[0059] In accordance with an embodiment, the predefined temporal threshold is selectable for each sensor pair, or a group of sensors, among the set of sensors, accounting for time measurement errors and signal propagation delays. In this embodiment, the temporal threshold for determining the arrival times of traveling wave signals at sensor pairs is configurable. This allows the system to accommodate time measurement inaccuracies caused by sensor synchronization errors, signal propagation delays, and grid-specific characteristics such as conductor length and signal velocity. For example, in a grid segment where sensors are 50 kilometers apart, signal propagation delays due to conductor properties might differ from segments with shorter distances. By selecting a temporal threshold specific to each sensor pair, the processor adjusts for such variations, ensuring accurate arrival time analysis. The threshold can be configured to within, e.g., 1 millisecond or any desired range based on sensor precision and operational requirements. Furthermore, the method accounts for measurement errors, which may arise from synchronization tools like GPS / GNSS receivers, cellular network, network server, or internal sensor clocks. This flexibility enables higher accuracy in correlating signal arrival times across multiple sensors. This embodiment improves the robustness and adaptability of the system by allowing configurable temporal thresholds for each sensor pair. By accounting for time delays and measurement inaccuracies, the method ensures reliable detection and classification of traveling waves, even in complex or large- scale grid topologies. This feature further enhances fault localization and disturbance analysis.

[0060] In accordance with an embodiment, similarity of amplitude sequences and arrival times of detected traveling waves is determined by calculating the propagation time of traveling waves along the atmospheric event flight path "as the crow flies". Typically, at frequencies between 5 kHz and 10MHz, a local grid event signals travel along the grid path increasing the signal propagation time, as opposed to atmospheric events, which signals at those frequencies travel a direct route via air, as the crow flies, at speed of light, and are picked up by the sensors directly or coupled to the electrical lines and then detected by the sensor. In this embodiment, the method determines the similarity of amplitude sequences and arrival times by calculating the maximum propagation time for the traveling waves along the grid topology. The calculation incorporates factors such as the physical distance between sensor locations and the signal velocity in conductors, which typically approaches the speed of light but may vary based on grid characteristics. For example, if sensors are installed 30 kilometers apart along a transmission line, the expected minimum propagation time for a traveling wave is approximately 0.1 milliseconds, given a signal velocity of 300,000 kilometers per second, or more accurately the speed of light 299,792.458 km / s. The processor compares this expected time with the measured arrival times of the traveling wave signals at the sensors to determine consistency. Similarly, amplitude sequences are analyzed for uniformity based on physical distances and attenuation rates, ensuring that discrepancies are identified and accounted for. By calculating the propagation time, the system accurately distinguishes between traveling waves originating from atmospheric events, which exhibit broader uniformity across sensors, and local events, where variations occur due to signal attenuation, grid topology or reflection. This embodiment enhances the accuracy of wave analysis by leveraging the grid topology and signal propagation characteristics. By determining minimum and maximum propagation times and analyzing amplitude consistency, the system reliably distinguishes local events from atmospheric disturbances. This approach improves fault detection precision and reduces false classifications, contributing to more efficient grid management.

[0061] In accordance with an embodiment, analyzing comprises recording waveforms of the detected traveling waves. In accordance with another embodiment the analyzing comprises comparing the recorded waveforms or filtered waveforms to historical databases of validated lightning event recordings using correlation methods or machine learning models. In these embodiments, the method involves the recording of waveforms of detected traveling waves and their subsequent comparison with historical databases. The waveforms include raw or filtered or modified versions that retain key characteristics of the traveling wave signals, such as amplitude, time, frequency content, and shape. These waveforms are compared using advanced correlation techniques or machine learning models. For instance, a recorded waveform can be compared to historical lightning event waveforms using mathematical correlation techniques such as convolution or ross-correlation. In addition, machine learning models, such as neural networks or transformer-based models, are trained on extensive datasets of validated traveling wave recordings. These datasets include examples of lightning-induced waves and non- lightning-induced waves, enabling the system to classify the source of a detected wave with high accuracy. The use of machine learning improves the system's ability to identify patterns in traveling wave data that may not be evident through conventional analysis. As a result, the system can distinguish between disturbances caused by atmospheric events, such as lightning strikes, and local grid events, such as faults or switching operations. This embodiment significantly enhances the accuracy and reliability of wave analysis by leveraging machine learning models and historical databases. By training on historical data, the system develops the capability to identify subtle differences between atmospheric and non-atmospheric disturbances. This reduces false positives, improves fault localization, and ensures quicker identification of lightning-induced events, contributing to the robustness of grid monitoring.

[0062] In accordance with an embodiment, analyzing comprises comparison of one or more signals, wherein the comparison of two signals, or a signal to a database of validated recordings, is performed using a mathematical or machine learning-based correlation technique, such as crosscorrelation, Fourier transform, dynamic time warping, or a machine learning model trained on historical data. In this embodiment, the system performs advanced signal analysis to compare detected traveling wave signals against other signals or historical recordings. The comparison is executed using mathematical correlation techniques or machine learning models. The techniques comprise, for example, cross-correlation, Fourier transform, dynamic time warping, or a machine learning model trained on historical data. The cross-correlation measures the similarity between two signals as a function of time-lag, enabling identification of overlapping patterns. Fourier transform analyzes the frequency spectrum of signals, particularly useful for identifying high-frequency components associated with lightning strikes. Dynamic time warping aligns signals with time-dependent variations, ensuring accurate comparisons even when signals are not perfectly synchronized. Machine learning models trained on historical traveling wave datasets, including atmospheric and non-atmospheric events, enable them to classify the source of the wave. For example, when a traveling wave is detected, its waveform is cross correlated with validated historical waveforms in a database. If the crosscorrelation coefficient exceeds a predefined threshold, the signal is identified as matching a known event type, such as a lightning strike. Similarly, the Fourier transform identifies frequency components in the signal that correspond to patterns observed in atmospheric disturbances. Dynamic time warping accounts for signal arrival time discrepancies caused by grid topology, ensuring that signals from sensors in different sections of the grid are compared accurately. Machine learning techniques further enhance the system's accuracy by identifying complex patterns and relationships in signal data that may not be evident through mathematical methods alone. This embodiment enhances the accuracy and efficiency of traveling wave analysis through robust signal comparison techniques. By combining mathematical and machine learning-based approaches, the method ensures reliable classification of wave origins, even in noisy or variable conditions. This capability significantly improves fault localization and identification of lightning- induced events, strengthening grid stability and response management.

[0063] In accordance with an embodiment, the signal records and / or metadata consisting of time-stamps, traveling wave analysis results and mains frequency signal analysis results, including event classifications and locations, are stored in a database, and this collected history data is later utilized, using statistical or heuristic analysis, for example by counting events or event types correlating with lightning strikes per grid section, to identify sections in electrical grid which are most prone to impact of lightning strikes.

[0064] Specifically, the method involves recording time-stamped signals, traveling wave analysis results, and mains frequency signal analysis, including classifications and event locations, into a database. This collected historical data is then utilized through statistical or heuristic analysis, for example, by counting events or event types correlated with lightning strikes per grid section. By analyzing historical data, grid sections most prone to lightning strikes or faults can be identified, allowing for proactive maintenance and targeted reinforcement of vulnerable grid areas, which reduces outage durations and improves grid reliability. The statistical insights derived from stored signal data enable grid operators to make data-driven decisions, optimizing resource allocation and improving fault management strategies. Additionally, the method leverages historical event patterns to identify recurring issues, enabling predictive maintenance, and reducing the frequency of unexpected grid failures. The collected data can be seamlessly integrated into outage management systems or other grid monitoring tools, enhancing situational awareness and response time during fault conditions. By introducing this into the method, the method not only detects and categorizes faults but also provides a means for continuous grid analysis and long-term operational improvement, leading to a more resilient and efficient electric grid.

[0065] In accordance with an embodiment, waveforms of the detected traveling waves are compared to historical databases of validated lightning strike recordings using one or more of the following correlation methods:

[0066] Cross-Correlation, Normalized Cross-Correlation, Sliding Dot Product, Fourier Transform (FFT), Phase Correlation, Pearson Correlation Coefficient, Spearman's Rank Correlation, Kendall Tau Correlation, Lagged Correlation, Windowed Cross-Correlation, Coherence, CrossAttention Mechanisms, Dynamic Time Warping (DTW), Wavelet CrossCorrelation, Hilbert Transform, Cross-Spectral Density (CSD), Empirical Mode Decomposition (EMD), Speech recognition algorithms or their derivatives, Mutual Information, Canonical Correlation Analysis (CCA), Euclidean Distance Correlation, Dynamic Cross-Correlation (DCC), Principal Component Analysis (PCA), Cross-Covariance, Phase-Shift Analysis, Zero-Crossing Analysis, Envelope Correlation, Autoencoders for Latent Space Correlation, Convolutional Neural Network (CNN)-Based Correlation, Recurrent Neural Networks (RNNs), Kernel Methods (e.g., Gaussian Kernel), Cross-Bispectrum, Transfer Entropy, Synchrosqueezed Transform, Nonlinear Regression Correlation, Self-Similarity and Fractal- Based Correlation, Machine Learning Methods, Granger Causality, Partial Correlation, Bayesian Inference Correlation, Predictive Correlation, Linear Predictive Coding (LPC), Cross-Recurrence Quantification Analysis (CRQA), Spectrogram Cross-Correlation, Cosine Similarity, Distance Correlation (dCor), Joint Entropy Correlation, Canonical Time Warping (CTW), Sparse Cross-Correlation, Cross-Amplitude Modulation (X-AM), Graph-Based Correlation, Energy Envelope Similarity, Nonlinear Dimensionality Reduction (e.g., t-SNE, UMAP), Information Theoretic Metrics, Higher-Order Spectral Analysis (HOSA), Maximum CrossCorrelation Coefficient (MCCC).

[0067] In accordance with an embodiment, comparison of two signals, or a signal to a history database of validated lightning strike recordings, is based on method wherein signals are transformed into high-dimensional embeddings via a large language model (LLM) or related neural network architecture, and similarity is determined using metrics such as cosine similarity, dot products, or other distance measures, or temporal correlation analysis, wherein sequential signal patterns are represented using LLM-derived embeddings or attention mechanisms, and correlations are identified by analyzing temporal dependencies within or between the sequences, or cross-attention mechanisms, wherein an LLM or transformer-based architecture learns interdependencies between two signals by aligning and weighing their representations through an attention mechanism, or signal-to-text translation, wherein signals are transformed into textual or symbolic representations, and the similarity is inferred by applying natural language processing techniques or direct input to an LLM, or multi-input models, wherein an LLM processes two or more input signals, either in raw or preprocessed form, to generate outputs that identify and quantify correlations, or signal pre-processing and embedding fusion, wherein signals are first converted into a domainspecific representation (e.g., frequency domain via Fourier Transform or spectrograms), and these representations are embedded and correlated using LLM-derived features, or nonlinear semantic correlation, wherein symbolic, event-driven, or feature-annotated signals are processed by an LLM to determine their similarity based on inferred semantic or contextual relationships, or unsupervised discovery of latent relationships, wherein an LLM is employed to identify correlations without explicit prior labels by learning latent spaces that encode the structure of both signals, or hybrid systems with signal processing models, wherein traditional signal processing techniques are used to preprocess the signals into interpretable formats, and LLMs are then utilized to infer or enhance correlations between them, or cross-domain correlations in multimodal datasets, wherein an LLM analyzes signals represented in different modalities (e.g., textual, visual, or numerical) to extract and quantify cross-domain relationships, or prompt engineering for correlation inference, wherein signals are described using natural language prompts, and the LLM outputs a similarity score or description based on its interpretation of the provided features, or predictive correlation, wherein an LLM generates predictions for one signal using the other as input, and similarity is quantified by comparing the prediction accuracy or residuals, or dimensionality reduction for embeddings, wherein an LLM is used to extract latent features from signals, and these features are mapped to a lower-dimensional space for comparison, or multi-scale attention mechanisms, wherein an LLM analyzes signals at multiple temporal or feature resolutions to determine correlations across different scales, or energy envelope similarity, wherein signal energy envelopes are extracted and processed via an LLM to determine similarity based on time-varying energy patterns. In accordance with an embodiment, determining whether a traveling wave signal detected was of an atmospheric origin or due to a local event in an electrical grid, can be done by fitting the signal amplitude sequences and their respective times of arrivals to a straight-line path over the air i.e., "as the crow flies", and fitting the arrival times of the signals to the electrical grid topology, assuming the signals would have travelled along the electrical grid lines and cables. If the arrival times of the signals better fit the straight-line paths over the air, there is high probability that the event was of an atmospheric origin. If the arrival times of the signals better fit to signals traveling along an electrical grid topology at the speed of the signal paths in question, then there is high probability that the event originated from a local event, such as a fault or a predictive event of an electrical grid component, near or in the electrical grid.

[0068] In accordance with an embodiment, determining if the traveling wave signal was originated by a local grid event, or if the traveling wave signal was of atmospheric event such as a lightning strike, includes fitting the arrival times of the signals as if the signals would follow the paths and signal speed of an electric grid topology and fitting the signal amplitude sequences and their respective times of arrivals to a straight-line paths over the air i.e. "as the crow flies", and if the signal arrival times fit better to the paths of a grid topology rather than straight lines at speed of light from a single event location, the event is of a local grid origin, and if the signal arrival times better fit to a model of straight lines at speed of light from an event location, the event is of an atmospheric origin, such as a lightning strike.

[0069] Sensors detecting electrical current and voltage signals in underground cables also detect lightning strikes, and these detections can be used to determine the signal type (local grid event or an atmospheric event such as lightning). However, the lightning strike traveling wave signal pickup point is the interconnection of an overhead line and the underground cable leading to the sensor. Therefore, the length of the cable and the signal speed in that cable will be added to the time delay of the detected lightning strike originated signal. In many cases an underground sensor is connected via one or more sets of cable to one or more segments of overhead lines. In these cases, for fitting the time of arrival of the signal, all these overhead-to-cable connection points and their respective time delays must be considered. Fitting the arrival times can be done in various ways, for example using Multilateration (TOA / TDOA, i.e., Time of Arrival I Time Difference of Arrival), Least Squares Estimation, Weighted Least Squares (WLS), Kalman Filtering, Angle of Arrival (AoA), Nonlinear Least Squares Localization, Optimization-Based Localization, Electromagnetic time reversal (EMTR), Time Reversal-Based Methods, and / or Fingerprinting. These methods, including generic two-ended traveling wave event location methods, can be referred to as traveling wave fault location methods.

[0070] In accordance with an embodiment, the method comprises calculating a straight-line distance between the location of the lightning strike and each sensor, wherein the straight-line distance corresponds to the propagation of the lightning-induced wave through air at the speed of light; calculating a grid line distance along the transmission line, wherein the grid line distance corresponds to the propagation of the traveling wave along the transmission line at a signal velocity determined by the line characteristics; comparing the straight-line distance to the grid line distance for the set of signal data; determining the location of the lightning strike based on a better fit of the arrival times to the straight- line distance compared to the grid line distance; wherein the better fit indicates that the lightning strike occurred at a distance away from the transmission line rather than directly on the transmission line.

[0071] By comparing straight-line distances (lightning-induced waves traveling through air at the speed of light) with grid line distances (signal propagation along the transmission line and / or cable), the method reliably identifies whether the lightning strike occurred directly on or away from the transmission line. This approach allows grid operators to distinguish between grid faults caused by nearby lightning strikes and those caused by direct hits, improving fault diagnosis. In addition, by utilizing the physical propagation characteristics of the traveling wave (air vs grid), the system pinpoints the strike location more accurately, reducing inspection time and improving response efficiency.

[0072] In accordance with an embodiment, the comparison of signals incorporates advanced machine learning techniques, such as embedding signals into high-dimensional spaces using neural networks or leveraging transformer-based models to analyze temporal and semantic dependencies between signals. In this embodiment, the system enhances signal analysis by employing advanced machine learning techniques. Specifically, the system embeds traveling wave signals into highdimensional feature spaces using neural networks or transformer-based models. These models are capable of capturing complex temporal relationships and dependencies within the signal data. For example, when a traveling wave is detected, its waveform can be represented as a series of amplitude and time points. A neural network maps this waveform into a high-dimensional vector space, where features such as amplitude patterns, frequency variations, and arrival times are encoded. Transformer-based models, widely used in temporal data analysis, further enable the system to identify semantic dependencies and temporal correlations between multiple waveforms recorded across sensors. Embedding the signals into high-dimensional spaces allows the system to identify subtle differences between traveling waves caused by local grid events and atmospheric disturbances, such as lightning strikes. The machine learning model analyzes these high-dimensional features to classify waveforms accurately and efficiently. This embodiment improves the precision of traveling wave analysis by leveraging cutting-edge machine learning techniques. By embedding signals into highdimensional spaces and analyzing temporal and semantic relationships, the system can detect patterns that are not apparent through traditional methods. This results in more accurate classification of wave sources and enhanced fault localization capabilities, contributing to improved grid stability and reliability.

[0073] In accordance with an embodiment, the present disclosure provides a method wherein the non-lightning induced traveling waves are categorized by amplitude attenuation and frequency analysis compared to atmospheric events. In this embodiment, the system differentiates between traveling waves caused by atmospheric events, such as lightning strikes, and those originating from non-lightning grid disturbances, such as faults or switching operations, such a tree falling on a transmission line. The differentiation is achieved by analyzing amplitude attenuation and frequency characteristics of the traveling waves.

[0074] For instance, lightning-induced traveling waves typically exhibit high initial amplitudes and steep rise times, followed by rapid attenuation as the wave propagates. These waves contain high-frequency components ranging from as low as 3 kHz up to 300 MHz in radio frequencies, depending on the nature and severity of the lightning strike. A lightning strike, being of very broadband in nature, also emits energy at other frequencies, such as Extremely Low Frequency (ELF): 3 Hz to 30 Hz; Very Low Frequency (VLF): 3 kHz to 30 kHz; Low to High Frequency (LF to HF) 30 kHz to 30 MHz; Very High Frequency (VHF): 30 MHz to 300 MHz; Microwave / Infrared : 300 MHz to ~10 THz; Visible Light: 380 nm to 750 nm; Ultraviolet (UV): 10 nm to 400 nm; X-Rays: ~10 keV to ~100 keV; Gamma Rays: ~100 keV to 20 MeV. Of these, the VLF, LF, HF and VHF frequencies are typically detectable using traveling wave sensors used on electrical grids. The electrical impulse of a lightning strike may cause a surge in the current level or voltage level at the transmission line. The surge may be characterized as an impulsive transient pattern, which may be referred to as a "burst". The surge (i.e., burst) is of a short duration (in the range of nanoseconds to microseconds). Non-lightning induced waves, such as those generated by switching operations or grid faults, tend to have similar amplitudes, and rise times. The processor compares the detected traveling wave's amplitude, temporal, and frequency profile to known patterns for lightning and non-lightning events. Amplitude attenuation is analyzed by measuring the reduction in wave magnitude over the distance between sensors, while frequency analysis identifies the dominant frequency components in the wave. By correlating these factors, the system accurately categorizes the wave's origin. This embodiment enhances the system's ability to distinguish between lightning-induced and non-lightning traveling waves by leveraging amplitude attenuation and frequency analysis. The approach ensures that atmospheric events are correctly identified, while grid disturbances such as faults or switching operations are accurately classified. This improves fault diagnosis, enables efficient resource allocation, and enhances overall grid reliability.

[0075] In accordance with an embodiment, the present disclosure provides a method wherein the predefined temporal threshold is set based on the propagation delay of the associated transmission line. In this embodiment, the predefined temporal threshold used to analyze the arrival times of traveling waves at sensors is configured based on the propagation delay characteristics of the associated transmission line. The propagation delay is determined using the physical length of the transmission line and the signal velocity in the conductor, which typically approaches 300,000 kilometers per second but can vary depending on conductor material and grid conditions. For example, if the transmission line is 100 kilometers long, the propagation delay for a traveling wave to traverse this distance is approximately 0.33 milliseconds. To account for such delays accurately, the predefined temporal threshold is set with a precision of 1 millisecond or better. The processor uses this threshold to compare signal arrival times recorded at sensors located in different sections of the grid, ensuring accurate identification of the wave's origin and propagation characteristics. By setting the temporal threshold based on the propagation delay, the system compensates for variations in wave arrival times due to grid topology and conductor characteristics. This allows for precise categorization of traveling waves as either local disturbances or atmospheric events. This embodiment ensures the system achieves high accuracy in signal arrival time analysis by setting a predefined temporal threshold aligned with the propagation delay of the transmission line.

[0076] In accordance with an embodiment, the set of sensors corresponds to a set of traveling wave sensors.

[0077] In accordance with an embodiment, each sensor of the set of sensors is one of: a coil wound around a ferrite core, an air-core coil, a hall-effect sensor, a metal core sensor, a current transformer, a voltage transformer, a voltage divider, electrical field sensor, capacitive sensor, or a Rogowski coil.

[0078] In accordance with an embodiment, a sensor corresponds to a sensor arrangement that comprises three sensors, arranged in such a way that a first sensor overlaps with a second sensor to form a cross-positioned two-sensor configuration, and a third sensor is positioned at a distance from this cross-positioned two-sensor configuration. The three sensors are configured to measure (electro)magnetics field in different directions relative to the transmission line. This may involve sensing magnetic fields in the vicinity of the transmission line. In accordance with an embodiment, the determination the presence of the traveling wave may involve amplification of the transient signal (i.e., the high-frequency traveling wave) detected by any or several of the sensors among the set of sensors and filtering of the amplified transient signal for removal of one or more of undesired frequency and undesired harmonics from the transient signal. Thereafter, a set of edges can be detected from a digitized version of the transient signal detected by any or several of the sensors. The detection of the set of edges in the digitized version of the transient signal comprises applying an edge detector function on the digitized version of the transient signal. The edge detector function may be an envelope peak- detector function. An edge of the set of edges, detected by any or several of the sensors among the set of sensors, may be an "edge of importance". The "edge of importance" is a rising edge that is detected at a particular instance by a sensor. This is the instance at which a first surge in the transient signal is detected. The instance corresponds to the arrival time. Thus, instances of detection of the first surge at the locations where the sensors are located correspond to the set of arrival times. Furthermore, an amplitude of the first surge, detected by any or several of the sensors, at the set of arrival times represents the set of amplitudes.

[0079] The current level is of the current signal that follows the impulsive transient pattern. Since the current signal corresponds to the high- frequency traveling wave, the "edge of importance" may be detected at an instance when the current level rises sharply (i.e., the rise follows an impulsive transient pattern). Thus, the instance at which the current level follows the impulsive transient pattern is the arrival time of the traveling wave. Each sensor detecting the traveling wave determines an instance at which the current level follows the impulsive transient pattern. Thus, based on detection of the high-frequency traveling wave by each sensor, the set of arrival times and the set of amplitudes may be determined. In accordance with an embodiment, a set of event detectors, associated with the set of sensors, may determine the set of arrival times and the set of amplitudes. Thereafter, the set of arrival times and the set of amplitudes are transmitted to the processor from the set of sensors.

[0080] It is to be noted that each sensor may detect multiple surges and, for example, a second surge, which arrives after the first surge, may be of a higher amplitude compared to an amplitude of the first surge.

[0081] Once the set of arrival times and the set of amplitudes are received by the processor, the cause of propagation of the traveling wave (i.e., the transient signal) through the transmission line is determined. The surge in the current levels in the transmission line, which is caused by the propagation of the traveling wave, may lead to an occurrence of a fault in one or more locations of the electric grid. The fault may be a transient fault, an earthing-related fault, an arcing fault, a short circuit fault, an open circuit fault, an overload fault, a partial discharge fault, or other faults related to broken conductors, broken components, and lost phases. The fault may be detected at the one or more locations by use of a variety of means.

[0082] Optionally, the determination of the cause of propagation of the traveling wave through the transmission line is based on the current signal or the current level (corresponding to the traveling wave that is detected by the set of sensors in the transmission line) following the impulsive transient pattern. This is because the lightning strike on the transmission line or in the vicinity of the transmission line may cause the transient event, which may lead to appearance of the impulsive transient pattern in the current signal. This impulsive transient pattern is detected by all sensors of the set of sensors at the set of locations of the electric grid at the set of arrival times (i.e., at instances of arrival of the impulsive transient pattern at the set of locations where the set of sensors are located). Furthermore, an envelope pattern (which represents an amplitude) of the (high-frequency) traveling wave (i.e., the transient signal), detected by each sensor of the set of sensors, is compared with envelope patterns of the traveling wave detected by other sensors of the set of sensors. The envelope of the traveling wave may be detected (for example by using the envelope peak- detector function with or without a rectifier) at each sensor prior to or during the detection of the set of edges from the digitized version of the transient signal. For example, the set of sensors may include three sensors. An envelope pattern of the traveling wave detected at a first sensor may be compared with envelope patterns of the traveling wave detected by a second sensor and a third sensor. Furthermore, an envelope pattern of the traveling wave detected by the second sensor may be compared with envelope patterns of the traveling wave detected by the first sensor and the third sensor, and an envelope pattern of the traveling wave detected by third sensor may be compared with envelope patterns of the traveling wave detected by the first sensor and the second sensor.

[0083] Optionally, based on a similarity between the envelopes of the traveling wave detected by each sensor of the set of sensors, it may be determined, by the processor, that only a lightning strike can cause this traveling wave to propagate in the transmission line. The one or more inputs received from the set of sensors may further include the envelopes of the traveling wave detected by each sensor of the sensors. Once the envelopes are received as the one or more inputs, they can be compared with each other and the similarity between the envelopes can be determined.

[0084] Additionally, the determination of the cause of propagation of the traveling wave through the transmission line (as due to the occurrence of the lightning strike or in vicinity of the transmission line) is also based on an area encompassed by the set of sensors and / or distances between different pairs of sensors of the set of sensors (i.e., different pairs of locations of the set of locations). Optionally, the determination of the cause of propagation of the traveling wave through the transmission line is further based on a distance between each pair of sensors of the set of sensors. This is because the surge in the current or voltage level at the transmission line, caused by the traveling wave propagating through the transmission line, leads to an occurrence of a fault in one or more locations of the electric grid. The fault may be detected at the one or more locations by various means.

[0085] The location of occurrence of the lightning strike is determined based on the set of locations of the set of sensors and / or one or more parameters of the traveling wave at the set of locations of the set of sensors. Optionally, the one or more parameters of the traveling wave at the set of locations (where the set of sensors are located) include the set of arrival times of the traveling wave at the set of locations of the set of sensors and the set of amplitudes of the traveling wave at the set of locations.

[0086] In a first scenario, if lightning has struck the transmission line directly, then the location of occurrence of the lightning strike is determined based on the set of arrival times. The location may be a point on the transmission line. In an example, if two locations of the set of locations, where two sensors of the set of sensors may be located, align themselves into a straight line and a transmission line that is connecting the two sensors gets struck by a lightning at a particular point on the transmission line, then a traveling wave propagating on the transmission line will arrive at the two locations (i.e., the two sensors) at different arrival times, (i.e., two arrival times of the set of arrival times). In this scenario, if the distance between the two locations or the two sensors is known, e.g., based on GPS / GNSS data, the point on the transmission line (where the lightning has struck) may be determined based on a difference between the two arrival times. The distance between the two locations is determined based on the two locations of the set of locations. This method works especially with the lower end of the frequency spectrum and when the lightning strikes very near the sensors and the sensors have a relatively straight electrical line path between them. In other cases, a second scenario should be used to locate the lightning strike.

[0087] In a second scenario, where the lightning strikes in the vicinity of (i.e., near), higher frequency spectrum of the lightning strike are used for locating the lighting event, when there are no straight paths for the lightning strike induced traveling wave to travel between the sensors, or when the lightning strike occurs further away of the sensors, the location of occurrence of the lightning strike is determined based on the set of locations of the set of sensors. In such scenarios a triangulation method is used for determining the location of occurrence of the lightning strike. The triangulation utilizes the two locations and another location of the set of locations where another sensor of the set of sensors is located. The two locations and the other location, i.e., three locations, when connected to each other form a triangle. The three locations of the three sensors are the three vertices of the triangle. In such scenarios, if the distance between each pair of sensors of the three sensors is known, if the three locations of the three sensors are known, and an angle between each pair of segments (of the triangle) connecting two vertices (i.e., each pair of locations where each pair of sensors are located) is known, then the location in the vicinity of the transmission line (where the lightning has struck) may be determined by use of sine formulas. Preferably, more than three sensors detect the traveling wave originating from a lightning strike, and then Multilateration (TOA / TDOA), Least Squares Estimation, Weighted Least Squares (WLS), Kalman Filtering, Angle of Arrival (AoA), Nonlinear Least Squares Localization, Optimization-Based Localization, or Fingerprinting method can be used. Optionally, a first angle, formed based on an intersection of a first segment and a second segment, may be obtained. The first segment may connect a first location of the set of locations and a second location of the set of locations. The second segment connects the first location and a third location of the set of locations. The first location is a location of a first sensor of the set of sensors. The second location is a location of a second sensor of the set of sensors. The third location is a location of a third sensor of the set of sensors. Thereafter, a second angle, formed based on an intersection of the first segment and a third segment, may be obtained. The third segment connects the second location and the third location. Thereafter, a third angle, formed based on an intersection of the second segment and the third segment, may be obtained. The first location, the second location, and the third location may be connected for triangulation and represent vertices of a triangle. The first segment, the second segment, and the third segment represent sides of the triangle.

[0088] Thereafter, a first distance between the first location and the second location may be obtained. Thereafter, a second distance between the first location and the third location may be obtained. Thereafter, a third distance between the second location and the third location may be obtained. The determination of the location of occurrence of the lightning strike is based on one or more of the first angle, the second angle, the third angle, the first distance, the second distance, and the third distance by use of sine formulas. The location of occurrence of the lightning strike is in the vicinity of a portion of the transmission line that is connecting two sensors out of the first sensor, the second sensor, and the third sensor.

[0089] Once the location of occurrence of the lightning strike is determined, one or more locations in the electric grid, where the same fault has occurred, may be determined. The determined one or more locations of occurrence of the same fault may indicate that the cause of propagation of the traveling wave (which had caused the same fault to occur at the one or more locations) in the transmission line is due to the occurrence of a lightning strike on the transmission line or in vicinity of the transmission line. Additionally, based on the location of occurrence of the lightning strike, one or more locations in the electric grid, where faults are likely to occur in the future, may be predicted.

[0090] In accordance with an embodiment, the method comprises determining a location of a lightning strike in an electric grid, wherein determination of the location of the lightning strike comprises utilizing a triangulation method based on arrival times of the lightning-induced wave at at least three sensors positioned at distinct locations along the transmission line; calculating angles formed between the straight-line distances and segments connecting the positions of the sensors; and determining the location of the lightning strike based on the calculated angles and distances using trigonometric relations.

[0091] In this embodiment, triangulation enhances the precision of determining the lightning strike location by leveraging arrival time differences at multiple sensor locations and applying trigonometric calculations. Triangulation considers the airborne wave propagation, enabling accurate localization even if lightning strikes occur at a distance from the line. By incorporating angles and distances, the method reduces errors caused by signal delays, noise, or other disturbances, leading to highly reliable results for grid operators.

[0092] In accordance with an embodiment, the sensors used for triangulation are selected based on their geographical separation (see e.g., FIGs. 1A- 1D, FIG. 2 and FIG. 9), ensuring the sensors are positioned such that the straight-line paths between the sensors and the lightning strike form a triangle with sufficient angular separation, and the signal quality, wherein the sensors providing the most consistent amplitude and arrival time data within a predefined threshold are prioritized for triangulation. Prioritizing sensors based on geographical separation ensures the triangulation method forms an appropriate triangle with sufficient angular separation, which is critical for achieving accuracy. Selecting sensors based on signal quality (amplitude consistency and arrival time precision) reduces the impact of noise or unreliable data, leading to more dependable results. This approach dynamically selects the most relevant sensors, making the system adaptable to varying grid topologies, sensor placements, and lightning strike scenarios. By focusing on high-quality sensor data, the system minimizes computational load while maximizing accuracy, improving operational efficiency.

[0093] Optionally, a user interface element, which is indicative of the location of occurrence of the lightning strike, may be rendered on a user interface of a map application. The user interface of the map application is operable to render a map of a region of the real-world where the electric grid may be situated. Furthermore, additional user interface elements indicative of the set of locations of the set of sensors may be rendered on the user interface. These user interface elements may also be indicative of the set of electric poles in which the set of sensors are installed.

[0094] The present disclosure also relates to the aforementioned second aspect. The various embodiments and variants disclosed above, with respect to the aforementioned first aspect, apply mutatis mutandis to the second aspect without any limitations.

[0095] The system comprises a set of sensors and a processor / a user device. The user device may include a processor. The user device may include programmable and / or non-programmable components that are operable to store information, process instructions, or receive / transmit instructions to detect occurrence of lightning strikes and resulting faults in the electric grid. The user device further includes a display, the processor, memory, input / output devices, and so on. The user device may further include physical and / or virtual computational entities that are capable of processing information for performance of various computational tasks. The processor of the user device is implemented as a microcontroller, a complex instruction set computing (CISC) microcontroller, a reduced instruction set (RISC) microcontroller, a very long instruction word (VLIW) microcontroller, or a Field Programmable Gate Array (FPGA).

[0096] Each sensor refers to a structure and / or module that may include programmable and / or non-programmable components that are operable to store, process, and share a set of arrival times and a set of amplitudes to the processor. The processor may receive the set of arrival times of the traveling wave and the set of amplitudes from the set of sensors and based on the reception, determine that a traveling wave is propagating in the transmission line. Optionally, the reception may be based on transmission of the set of arrival times and the set of amplitudes by the set of sensors to the processor. The processor determines, based on the received arrival times and amplitudes, that a traveling wave is propagating in the transmission line.

[0097] The set of sensors and the processor (i.e., the user device) communicate with each other using a communication interface. The communication interface includes a medium (e.g., a communication channel) through which the communication takes place. Examples of the communication interface include, but are not limited to, a communication channel in a computer cluster, a Local Area Network (LAN), a Metropolitan Area Network (MAN), a Wireless Sensor Network (WSN), network, a cloud, and / or the Internet.

[0098] Optionally, the processor determines, based on a set of arrival times and a set of amplitudes, that the current signal (i.e., the current level at the transmission line), which corresponds to the traveling wave, follows an impulsive transient pattern. The set of arrival times represents the instants at which a rising edge of the transient signal, corresponding to the surge, is detected by the any or several of the sensors among the set of sensors (i.e., traveling wave sensors). The set of amplitudes represents the amplitudes of the surge detected by the any or several of the sensors among the set of sensors. Thus, based on the appearance of an impulsive transient pattern in the current level (i.e., the current signal / level following the impulsive transient pattern), the processor determines that the cause of propagation of the traveling wave through the transmission line is the occurrence of a lightning strike on the transmission line or near the transmission line.

[0099] Optionally, the determination of the cause of propagation of the traveling wave through the transmission line may be further based on a distance between each pair of sensors of the set of sensors. This is because the surge in the current level in the transmission line leads to an occurrence of the same event in one or more locations of the electric grid which are sufficiently far from each other. The distance (for example, a value in a range 20-60 km) between each pair of sensors of the set of sensors, or a group of sensors of the set of sensors, is such that the same fault, if caused by a transient event other than a lightning strike, is not likely to occur at the one or more locations.

[0100] The processor determines a location of the occurrence of the lightning strike based on the set of locations of the set of sensors and one or more parameters of the traveling wave at the set of locations. Based on the determined location of occurrence of the lightning strike, one or more locations of occurrence of the same fault in the electric grid is determined. Optionally, the one or more parameters of the traveling wave include the set of arrival times and the set of amplitudes. If the location of occurrence of the lightning strike is a point on the transmission line, the location of occurrence of the lightning strike is determined based on the set of arrival times and the set of amplitudes. On the other hand, if the location of occurrence of the lightning strike is a point in the vicinity of the transmission line, then the location of occurrence of the lightning strike is determined based on the set of locations of the set of sensors. The present disclosure also relates to the aforementioned third aspect. The various embodiments and variants disclosed above, with respect to the aforementioned first aspect and the second aspect, apply mutatis mutandis to the third aspect without any limitations.

[0101] DETAILED DESCRIPTION OF THE DRAWINGS

[0102] Referring to FIG. 1A, illustrated is an exemplary portion of an electric grid 100 which includes an arrangement for determining the occurrences of an event or a fault in the electric grid 100, in accordance with various embodiments of the present disclosure. The exemplary portion of the electric grid 100 comprises two transmission line sections, a transmission line 102, in a first section 100A of the electric grid, and a transmission line 102', in a second section 100B of the electric grid. The transmission line 102 contains three phase conductors 102a, 102b, and 102c, and the transmission line 102' contains three phase conductors 102a', 102b', and 102c'. The first section 100A includes a first set of sensors 104a, 104b, and 104c, installed at locations 108a, 108b, and 108c, respectively. The second section 100B includes a second set of sensors 104a', 104b', and 104c' installed at locations 108a', 108b', and 108c'. The sensors 104a-104c are installed in a set of electric poles 110a- 110c. The sensors 104a'-104c' are installed in a set of electric poles 110a'-110c'. The electric poles HOa-llOc, and 110a'-110c' are located at locations 108a-108c, and 108a'-108c', respectively.

[0103] The first section 100A is in the area marked A, where the circle marked with a solid black line, around the transmission line 102, describes the outer boundary of area A. The second section 100B is in the area marked B, where the circle marked with a solid black line, around the transmission line 102', describes the outer boundary of area B. The areas A and B are at a distance from each other. The electric grid section 1OOA and the electric grid section 1OOB may be, for example, in some circumstances, electrically isolated from each other, e.g., by step-down and step-up transformers, although both sections 1OOA and 1OOB are part of the same electric grid 100. The distance of the electric grid section 100A from the electric grid section 100B may be, for example, 20-1000 kilometers.

[0104] The first set of sensors 104a, 104b, 104c are configured to monitor the phase conductors 102a, 102b, and 102c and to detect signal data comprising a set of amplitudes and a set of arrival times of transient signals. The sensors 104a, 104b, and 104c are operable to monitor magnetic fields, voltage levels, electrical fields, or current signals associated with transmission line 102. The signals are transmitted to a processor, which analyzes the data to determine whether the traveling wave originates locally or from an atmospheric event, such as a lightning strike. Respectively, the second set of sensors 104a', 104b', and 104c' are configured to monitor the phase conductors 102a', 102b', 102c', and to detect signal data comprising a set of amplitudes and a set of arrival times of transient signals. The sensors 104a', 104b', and 104c' are operable to monitor magnetic fields, voltage levels, electrical fields, or current signals associated with transmission line 102'.

[0105] A lightning strike may affect the electric grid 100. Lightning may strike in area A near the transmission line 102 at a location marked 106A. A traveling wave may propagate directly over the air, and along the grid lines and cables, due to the lightning strike, at location 106A. The location 106A of occurrence of the lightning strike is determined based on the set of locations 108a-108c, 108a'-108c' of the set of sensors 104a-104c, 104a'-104c', and one or more parameters of the traveling wave at the set of locations 108a-108c, 108a'-108c'. The one or more parameters comprise the set of arrival times of the traveling wave at the set of locations 108a-108c, 108a'-108c' and the set of amplitudes of the traveling wave at the set of locations 108a-108c, 108a'-108c'.

[0106] FIGS. IB, 1C and ID illustrate by way of examples propagation paths through which a fault current signal can propagate in the electric grid, with reference to the same electric grid as illustrated in FIG. 1A. Referring to FIG. IB, illustrated is an example where an event, for example a fault, such as a line fault, occurs approximately at a location 108b. The event location has a lightning bolt denoted as reference numeral 106B (also referred herein as an event location). The fault in a transmission line generates traveling waves (fault current) that propagate from the event location 106B towards the ends (terminals) of the transmission line at approximately the speed of light. The fault current is divided into two parts and it travels predominantly along the grid lines 102 from the location 106B in two directions. The propagation directions of the traveling waves along the grid lines are indicated by arrow 106B* and arrow 106B". When a traveling wave propagating in direction of arrow 106B' arrives at the end of the transmission line at location 108a, sensor 104a, at location 108a, detects the traveling wave signal and records the set of amplitudes and set of arrival times of the traveling wave. When a traveling wave propagating in direction of arrow 106B" arrives at the end of the transmission line at location 108c, sensor 104c, at location 108c, detects the traveling wave signal and records the amplitude and time of arrival of the arriving traveling wave. The method determines the set of arrival times and a set of amplitudes, by means of the set of sensors, and by processing the time and assigning timestamps to them. The difference in timestamps enables determining the fault location.

[0107] Referring to FIG. 1C, illustrated is another example where an event that occurs at a location 108b of electric pole 110b is a lightning strike. An event location is denoted with 106B. In this example, the lightning strike current is divided into two and it travels in waves in both directions of the grid lines 102 as at the speed of light. The waves travel along the electric grid lines. The waves further propagate through the air and travel the shortest path from the event location 106B to the sensors 104a-104c, 104a'-104c'. The directions of propagation are illustrated, by arrow lines 106B', 106B", 106B'", 106B"" and 106B'"". The arrow-dashed lines 106B'", 106B"" and 106B'"" are defined herein as straight-line paths, distances, or as atmospheric event flight paths, i.e., "as the crow flies" between the event location 106B and the location of at least one sensor, in the set of sensors 104a-104c, 104a'-104c'. The waves propagate between different sections 1OOA and 1OOB of the electric grid and are detected by a first set of sensors 104a-104c, and by a second set of sensors 104a'-104c'. All sensors 104a-104c, 104a'-104c' are capable of detecting the waves even though the section 1OOA and section 1OOB would be separate from each other, i.e., electrically isolated.

[0108] Referring to FIG. ID, illustrated is yet another example where an event occurring at a location 106 is a lightning strike. In this example, there is an additional (traveling wave) sensor 104d installed on an underground cable 105 at a distance from the electric pole 110c. The sensor 104d is, for example, located approximately at the length of the underground cable 105 from the electric pole 110c. The sensor 104d is configured to monitor the signal flowing along the underground cable 105 from the associated transmission line 102, via an electric pole 110c. The fault current generated by the lighting strike at the location 106 propagate to the sections of the electric grid and are detected herein by a set of sensors 104a-104c at locations 108a-108c and by a sensor 104d at a location 108d. The signals travel on the shortest path and propagate through air to the sensors 104a-104c. The length is illustrated herein as straight- line distances denoted as 106', 106", 106'". The arrow-dashed lines 106', 106" and 106'" are defined as straight-line distances, or as atmospheric event flight paths, between the event location 106 and the locations of the set of sensors 104a-104c. The signal received by the sensor 104d at a distance from other sensors 104a-104c sees an additional delay corresponding to the length of the underground cable 105 and the signal speed in the underground cable, with respect to the sensors 104a-104c, 104a'-104c' installed at the electric poles 110a- 110c, 110a'-110c'.

[0109] FIGS. 1A, IB, 1C and ID are merely examples, which should not unduly limit the scope of the claims herein. A person skilled in the art will be able to recognize many variations, alternatives, and modifications of embodiments of the present disclosure.

[0110] Referring to FIG. 2, illustrated is an exemplary determination of location of occurrence of the lightning strike in the electric grid 100, in accordance with various embodiments of the present disclosure. Lightning strikes at location 106 which is in a vicinity of a portion of the transmission line. Due to occurrence of a lightning strike an electromagnetic wave is generated at the location 106 that electromagnetic wave propagates from the location of the lightning strike to the locations of the sensors 104a, 104b, 104c and 104d installed at the electric poles and reach the sensors, e.g., in some microseconds. Herein, in FIG. 2, each of the four sensors, detect an envelope pattern corresponding to the electromagnetic wave that envelope pattern detected further corresponds to amplitude variation of the electromagnetic wave over a first period. The envelope patterns the sensors detect function as one or more inputs, being indicative of arrival times and amplitudes which inputs each of the four sensors transmits to a processor for further analysis. Accordingly, the processor receives four envelope patterns from the four sensors. The system then compares each envelope pattern received with the other three envelope patterns. Based on the comparison, a similarity between the four envelopes of the traveling wave, detected by each of the four sensors, may be determined. Similarity between the envelope patterns of the traveling wave may be based on at least one of: polarity, amplitude, frequency, of the travelling wave.

[0111] The location of occurrence of the lightning strike 106 is determined based on a triangulation method by using three locations of the set of locations 108a-108c where three sensors of the set of sensors 104a-104c are located. The three locations of the three sensors represent three vertices of a triangle. The triangle includes three segments that connect the three locations (vertices) and represent the three sides of the triangle.

[0112] A first angle 112 is formed based on an intersection of a first segment 118 and a second segment 120. A second angle 114 is formed based on an intersection of the first segment 118 and a third segment 122. A third angle 116 is formed based on an intersection of the second segment 120 and the third segment 122. The segments of the triangle, viz., the first segment 118, the second segment 120, and the third segment 122, connect the vertices of the triangle.

[0113] The first segment 118 connects a first location 108a of the set of locations 108a-108c with a second location 108b of the set of locations 108a-108c. A length of the first segment 118 represents a first distance between the first location 108a and the second location 108b. The second segment 120 connects the first location 108a and a third location 108c of the set of locations 108a-108c. A length of the second segment 120 represents a second distance between the first location 108a and the third location 108c. The third segment 122 connects the second location 108b and the third location 108c. A length of the third segment 122 represents a third distance between the second location 108b and the third location 108c. The first location 108a is a location of a first sensor 104a of the set of sensors 104a-104c, the second location 108b is a location of a second sensor 104b of the set of sensors 104a-104c, and the third location 108c is a location of a third sensor 104c of the set of sensors 104a-104c. The location of occurrence of the lightning strike is determined based on one or more of the first angle 112, the second angle 114, the third angle 116, the first distance 118, the second distance 120, and the third distance 122 by use of sine formulas. Specifically, a segment 200 may be considered and "x" and "y" may be determined using the sine formulas. Once "x" and "y" are determined, the location of occurrence of the lightning strike may be determined.

[0114] FIG. 2 is merely an example, which should not unduly limit the scope of the claims herein. A person skilled in the art will recognize many variations, alternatives, and modifications of embodiments of the present disclosure.

[0115] FIG. 3 illustrates an example of a first set of graphs 300A, 300B, 300C, 300D that indicate a variation of amplitude of a traveling wave detected by a set of sensors in an electric grid, in accordance with an embodiment of the present disclosure. In the example illustrated in FIG. 3, the set of sensors in the electric grid include four sensors. Each of the four sensors is a traveling wave sensor that is capable of detecting, i.e., determining, a traveling wave that may be propagating in a transmission line of the electric grid due to an occurrence of a lightning strike on the transmission line or in vicinity of the transmission line. The four sensors may be installed in electric poles in the electric grid at geographical locations. Each of the four sensors, at different geographical locations, may detect an envelope pattern of the traveling wave that corresponds to amplitude variation of the traveling wave over a first period. The envelope pattern may be an input which each of the four sensors may transmit to a processor. The processor may receive four envelope patterns from the four sensors and compare each envelope pattern with the other three envelope patterns. Based on the comparison, a similarity between the four envelopes of the traveling wave, detected by each of the four sensors, may be determined. The similarity between the envelope patterns indicates that a lightning strike has caused the traveling wave to propagate in the transmission line of the electric grid.

[0116] FIG. 3 is merely an example, which should not unduly limit the scope of the claims herein. A person skilled in the art will recognize many variations, alternatives, and modifications of embodiments of the present disclosure. For example, the first set of graphs 300 may include any number of graphs.

[0117] FIG. 4 illustrates a second set of graphs 400 that are indicative of a variation of amplitude of a traveling wave detected by a set of sensors in an electric grid, in accordance with an embodiment of the present disclosure. For example, the set of sensors in the electric grid may include four sensors and each of the four sensors are traveling wave sensors. Each of the four sensors, at different geographical locations, may detect an envelope pattern of the traveling wave that corresponds to amplitude variation of the traveling wave over a second period. The envelope pattern may be an input which each of the four sensors may transmit to a processor. The processor may receive four envelope patterns from the four sensors and compare each envelope pattern with the other three envelope patterns. Based on the comparison, a similarity between the four envelopes of the traveling wave, detected by each of the four sensors, may be determined. The similarity between the envelope patterns indicates that a lightning strike has caused the traveling wave to propagate in the transmission line of the electric grid.

[0118] FIG. 4 is merely an example, which should not unduly limit the scope of the claims herein. A person skilled in the art will recognize many variations, alternatives, and modifications of embodiments of the present disclosure. For example, the second set of graphs 400 may include any number of graphs. FIG. 5 depicts an exemplary user interface 500 of a map application that indicates a location of occurrence of a lightning strike, in accordance with an embodiment of the present disclosure. The user interface 500 of the map application is operable to render the map of a region of the real- world where an electric grid may be situated. The user interface 500 of the map application renders a user interface element 502 on the map of the region. The user interface element 502 may be indicative of the location in the electric grid, where a lightning strike might have occurred. Furthermore, additional user interface elements 504a-504f, indicative of a set of locations of a set of sensors in the electric grid, may be rendered on the user interface 500. The user interface elements 504a-504f may also be indicative of the set of electric poles in the electric grid where the set of sensors may be installed. The set of sensors may detect a traveling wave and based on the set of locations of the set of sensors (indicated by the user interface elements 504a-504f), the location of occurrence of the lightning strike may be determined. Once the location is determined, the user interface element 502 may be rendered on the user interface 500. The measuring line at the bottom of FIG. 5 indicates example distances between the set of locations of the set of sensors, by means of user interface element 504a-504f: the measuring line corresponds to a distance of 10 kilometers on the map, on the user interface 500 of a map application.

[0119] FIG. 5 is merely an example, which should not unduly limit the scope of the claims herein. A person skilled in the art will recognize many variations, alternatives, and modifications of embodiments of the present disclosure.

[0120] Referring to FIG. 6, illustrated is a block diagram of a system 600 for detecting faults in an electric grid, in accordance with an embodiment of the present disclosure. The system 600 comprises a set of sensors 104a- 104c, 104a'-104c' and the processor 602. The set of sensors 104a- 104c, 104a'-104c' are at a set of locations within the electric grid 100. The set of sensors 104a-104c,104a'-104c' correspond to a set of traveling wave sensors that are configured to be installed in the set of electric poles HOa-llOc, 110a'-110c'. The electric grid 100 includes a transmission line 102, in a first section of the electric grid, and a transmission line 102', in a second section of the electric grid, that passes through the set of electric poles HOa-llOc, 110a'-110c'. The processor 602 is operable to receive a set of amplitudes and a set of arrival times from the set of sensors 104a-104c, 104a'-104c' (i.e., the set of traveling wave sensors). The reception is based on transmission of the set of amplitudes and the set of arrival times by the set of sensors 104a-104c, 104a'-104c' to the processor 602. The processor 602 is further operable to determine that a traveling wave is propagating in the transmission line 102, 102' in the electric grid 100 based on the received set of amplitudes and the received set of arrival times. The processor 602 is further operable to determine that the traveling wave is propagating through the transmission line 102, 102' due to an occurrence of a lightning strike on the transmission line 102, 102' or near the transmission line 102, 102'. The processor 602 is further operable to determine a location of occurrence of the lightning strike based on the set of locations and / or one or more parameters of the travelling wave at the set of locations of the set of sensors 104a-104c, 104a'-104c'. Based on the determined location of occurrence of the lightning strike, one or more locations of occurrence of fault (caused by the traveling wave) in the electric grid 100 is determined. Furthermore, based on the determined location of occurrence of the lightning strike, the processor 602 may predict one or more locations where a fault is likely to occur in the future.

[0121] It may be understood by a person skilled in the art that FIG. 6 includes a simplified architecture of the system 600, for sake of clarity, which should not unduly limit the scope of the claims herein. It is to be understood that the specific implementation of the system 600 is provided as an example and is not to be construed as limiting. The person skilled in the art will recognize many variations, alternatives, and modifications of the system 600 of the present disclosure.

[0122] Referring to FIGS. 7A and 7B, illustrated is a flowchart 700 that includes steps of a method for detecting a traveling wave in an electric grid, in accordance with an embodiment of the present disclosure.

[0123] At step 701, the method comprises receiving a set of signal data, comprising a set of amplitudes and a set of arrival times, from a set of sensors associated with a transmission line of the electric grid, the set of sensors comprising a first set of sensors positioned at locations in a first section of the electric grid; and a second set of sensors positioned at locations in a second section of the electric grid.

[0124] At step 702, the method comprises determining the presence of a traveling wave propagating along a transmission of the electric grid based on the received signal data.

[0125] At step 703, the method comprises categorizing the traveling wave as either a local event or, an atmospheric event, such as a lightning strike, wherein the categorizing comprising.

[0126] At step 704, the method comprises recording waveforms of the detected traveling waves.

[0127] At step 705, the method comprises analyzing whether the detected traveling wave exhibits similar amplitude sequences and arrival times across the set of sensors.

[0128] At step 706, the method comprises distinguishing between a locally detectable traveling wave, where the traveling wave is detected either by any or several of the sensors in the first set of sensors or any or several of the sensors in the second set of sensors, and if a traveling wave is detectable at both any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors.

[0129] At step 707, the method comprises identifying the source of the traveling wave as a local grid event if the traveling wave is detected only with any or several of the sensors in the first set of sensors or only with any or several of the sensors in the second set of sensors; and an atmospheric event, such as a lightning strike, is identified if the traveling wave is detected by both any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors.

[0130] It may be appreciated that the steps 701, 702, 703, 704, 705, 706, and 707 are only illustrative, and other alternatives can also be provided where one or more steps are added, one or more steps are removed, or one or more steps are provided in a different sequence without departing from the scope of the present disclosure.

[0131] Referring to FIG. 8, illustrated is an example of a lightning induced fault event triggered recording at two frequency bands. Reference number 801 represents absolute values of signal rising edges, or envelope, from 50 kHz to 3 MHz in 1 millisecond resolution as detected by a high- frequency sensor, whereas reference number 802 shows time-aligned 3- phase line currents in Amperes of the electrical line in question. An earlier part of the signal 801, labeled as 803, was deemed as an atmospheric event, most probably a lightning event, as similar correlating recordings were measured at sensors in multiple separate electric grids. The later labeled part 804 of the high-frequency signal 801 was only recorded in the sensors connected to the same step-down transformer, and therefore this event was likely not of atmospheric origin, but local to the part of the grid where this sensor in question is located. It is also visible that simultaneously to the high-frequency signals of the detected local event 804 in the high-frequency signal graph 801, the line currents 802 exhibit symptoms of 2-phase short circuit event 805, followed by a line protection device tripping 806 in order to try to clear the fault by deenergizing the line. In this manner, detecting :

[0132] 1. which part of the recorded signal was of atmospheric origin and which was local;

[0133] 2. what the phase current or phase voltage signals and higher frequency signals measured by sensors along the grid indicate of this event's type and classifying the event moment accordingly this example shoes a short circuit between two phases);

[0134] 3. the atmospheric event 803 can be identified, and the lightning event can be located using traveling wave fault location principles of this invention;

[0135] 4. the fault situation 806 can be identified and as the high-frequency signals 804 correlating in time with the fault situation 806 are identified to be of local origin and not atmospheric, the grid fault event location can be found using traveling wave fault location methods, and

[0136] 5. the grid operators and the service crew can be notified of the nature and location of these events and possible fault types using push notifications, text messages, email, integrations to an outage management system, in form of an event or alert log, and showing the events on a map view 500, preferably together with the topology of the part of the grid in question.

[0137] Referring to FIG. 8, there are also visible multiple high energy (sensor ADC input voltage), and lower energy (sensor ADC input voltage) spikes and / or wavefronts / edges in the graph 801. In practice, most of these events can be located using TOA / TDOA principles, and strategies explained in this invention, depending on whether that particular traveling wave spike / wavefront / edge was found to be of local grid event origin, or from an atmospheric event. Therefore, from a single set of recordings, mixed of atmospheric and local grid event information. The above-mentioned method improves the detection, classification, and localization of grid fault events by analyzing recorded signals from multiple sensors. The method first identifies which parts of the signal originate from atmospheric events, such as lightning strikes, and which correspond to local grid faults. It then uses traveling wave fault location techniques to locate the exact fault event. By analyzing current and / or voltage values of the mains frequency and its harmonics, the method accurately identifies fault situations and distinguishes high-frequency signals of local origin from atmospheric noise. The fault location is determined using traveling wave methods, ensuring precision. Additionally, the method notifies grid operators and service crews of the fault's nature and location, providing the information through event logs, grid topology map views, or integrations with outage management systems. By separating local grid fault signals from atmospheric noise, the method improves fault classification and accuracy. Further, traveling wave fault location methods ensure the exact location of the fault is pinpointed efficiently. The inclusion of notifications (e.g., event logs or map views) enables grid operators and service crews to respond quickly, minimizing downtime. The method leverages measurable parameters like mains frequency harmonics and high-frequency signals, ensuring reliable performance in real-world grid conditions. In summary, this approach enhances grid reliability and operational efficiency by providing clear, actionable insights into the nature and location of grid faults.

[0138] Referring to FIG. 9, there is illustrated an exemplary electric grid on a geographical map, the electric grid having an arrangement for determining occurrences of an event or a fault in the electric grid, in accordance with various embodiments of the present disclosure. The map illustrates, for example, a part of a country. In FIG. 9, reference numeral 900 is an electric grid, for example a national electrical grid, extending federally over the geographical area, or portion thereof, shown in FIG. 9 (the detailed configuration and lines of the electrical grid are not shown on the map). Electric grid 900 comprises a plurality of regional electric grids 900A, 900B, 900C, 900D, 900E, 900F, 900G, 900H, and 9001. The electric grids 900A-900I are part of a national electric grid 900. All of these electric grids 900A-900I are isolated from each other, for example by step-down and / or step-up transformers, although they are part of the same national electric grid. The electric grids may be similar to electric grids as illustrated above with reference to FIG. 1A. Thus, all the electric grids include an arrangement for determining occurrences of an event or a fault in the electric grid, in accordance with various embodiments of the present disclosure. Thereby, for example, if sensors in more than one grid simultaneously detect similar and simultaneously traveling wave signals, it is likely that an atmospheric event, such as lightning, is involved.

[0139] Modifications to embodiments of the present disclosure described in the foregoing are possible without departing from the scope of the present disclosure as defined by the accompanying claims. Expressions such as "including", "comprising", "incorporating", "have", "is" used to describe, and claim the present disclosure are intended to be construed in a nonexclusive manner, namely allowing for items, components or elements not explicitly described also to be present. Reference to the singular is also to be construed to relate to the plural.

Claims

CLAIMS1. A method, implemented in a processor, for detecting a traveling wave in an electric grid (100), the method comprising: receiving a set of signal data, comprising a set of amplitudes and a set of arrival times, from a set of sensors associated with a transmission line (102, 102') of the electric grid (100), the set of sensors comprising a first set of sensors (104a-104c) positioned at locations (108a-108c) in a first section (100A) of the electric grid; and a second set of sensors (104a'-104c') positioned at locations (108a'-108c') in a second section (100B) of the electric grid; determining the presence of a traveling wave propagating along a transmission lines (102, 102') of the electric grid based on the received signal data; categorizing the traveling wave as either a local event or, an atmospheric event, such as a lightning strike, wherein the categorizing comprising; recording waveforms of the detected traveling waves; analyzing whether the detected traveling wave exhibits similar amplitude sequences and arrival times across the set of sensors (104a- 104c, 104a'-104c'); distinguishing between a locally detectable traveling wave, where the traveling wave is detected either by any or several of the sensors in the first set of sensors (104a-104c) or any or several of the sensors in the second set of sensors (104a'-104c'), and / or whether a traveling wave is detectable any or several of the sensors in the first set of sensors (104a-104c) and any or several of the sensors in the second set of sensors (104a'-104c'); identifying the source of the traveling wave as a local grid event if the traveling wave is detected only with any or several of the sensors inthe first set of sensors or only with any or several of the sensors in the second set of sensors; and an atmospheric event, such as a lightning strike, is identified if the traveling wave is detected by both any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors.

2. A method according to claim 1, wherein determining comprises analyzing variations in the magnetic, electric, or electromagnetic field surrounding the transmission line, or current or voltage detected by any or several of the sensors in the set of sensors (104a-104c, 104a'-104c').

3. A method according to claim 1, wherein the detected traveling wave comprises multiple traveling wavefronts or edges, each representing a local grid event or an atmospheric event, and each traveling wavefront or edge is individually identified and classified as a local grid event or an atmospheric event4. A method according to claim 1, wherein the traveling wave signal, filtered waveforms of the detected traveling waves, and / or waveform envelopes are processed to separate local grid event signals and atmospheric event signals, by removing the atmospheric event signals from the original signal data to yield recordings containing only local grid event signals, and removing the local grid event signals from the original signal data to yield recordings containing only atmospheric event signals.

5. A method according to claim 1, 2 or 3, wherein analyzing whether the detected traveling wave exhibits similar amplitude sequences and arrival times across the set of sensors (104a-104c, 104a'-104c') is executed within a predefined temporal threshold.

6. A method according to any of the preceding claims, wherein analyzing further comprises determining whether the detected traveling wave exhibits consistent amplitude sequences and arrival times across the firstset of sensors (104a-104c) and / or the second set of sensors (104a'- 104c'), within a predefined amplitude tolerance range established based on signal variability.

7. A method according to any of the preceding claims, wherein analyzing further comprises mathematically correlating the waveforms of the detected traveling waves, filtered waveforms of the detected traveling waves, and / or waveform envelopes of the detected traveling waves in order to test their similarity between each other in order to gain confirmation that these signals have the same origin event.

8. A method according to any of the preceding claims, wherein determining if the traveling wave was detectable only locally involves verifying that all sensors among the set of sensors (104a-104c, 104a'- 104c') detecting the signal and transmitting the corresponding signal data are located within a predefined local area of the electrical grid.

9. A method according to any of the preceding claims, wherein determining if the traveling wave signal was detectable only locally is based on the sensors among the set of sensors (104a-104c, 104a'-104c') detecting the signal being positioned in a section of the electric grid that is galvanically connected, excluding sections of the electric grid separated by isolating transformers, open contacts, or missing connection, and where the electrical lines are geographically separated.

10. A method according to any of the preceding claims, wherein determining if the traveling wave signal was detectable only locally is based on the proximity of all detecting sensors among the set of sensors (104a-104c, 104a'-104c'), ensuring no sensor is located beyond a predefined maximum distance, determined based on physical and operational constraints of the electric grid.

11. A method according to claim 5, wherein the predefined temporal threshold is selectable for each sensor pair, or sensor group, among the set of sensors (104a-104c, 104a'-104c'), accounting for time measurement errors and signal propagation delays.

12. A method according to any of the previous claims, wherein determining if the traveling wave signal was originated by a local grid event, or if the traveling wave signal was of atmospheric event such as a lightning strike, includes fitting the arrival times of the signals as if the signals would follow the paths and signal speed of an electric grid topology and fitting the signal amplitude sequences and their respective times of arrivals to a straight-line paths over the air i.e. "as the crow flies", and if the signal arrival times fit better to the paths of a grid topology rather than straight lines at speed of light from a single event location, the event is of a local grid origin, and if the signal arrival times better fit to a model of straight lines at speed of light from an event location, the event is of an atmospheric origin, such as a lightning strike.

13. A method according to any of the preceding claims, wherein analyzing comprises recording waveforms of the detected traveling waves and comparing the recorded waveforms or filtered waveforms to historical databases of validated lightning event recordings using correlation methods or machine learning models, wherein the machine learning models are trained on a dataset comprising historical traveling wave recordings from both atmospheric and non-atmospheric events.

14. A method according to any of the preceding claims, wherein analyzing comprises comparison of one or more signals, wherein the comparison of two signals, or a signal to a database of validated recordings, is performed using a mathematical or machine learning-based correlation technique, such as cross-correlation, Fourier Transform, Dynamic Time Warping, or a machine learning model trained on historical data.

15. A method according to any of the preceding claims, wherein the comparison of signals incorporates advanced machine learning techniques, such as embedding signals into high-dimensional spaces using neural networks or leveraging transformer-based models to analyze temporal and semantic dependencies between signals.

16. A method according to any of the preceding claims, wherein the nonlightning induced traveling waves are categorized by amplitude attenuation and frequency analysis compared to atmospheric events.

17. A method according to any of the preceding claims, wherein the method further comprises rendering a user interface element (502), indicative of a location (106) of occurrence of the lightning strike, on a user interface (500) of a map application that is operable to render a map of a region that includes the electric grid (100), wherein the user interface indicates the set of locations (504a-504f).

18. A method according to any of the preceding claims, wherein analyzing the records from the multiple sensors comprises: determining which temporal parts of the recorded signal correspond to atmospheric origin and which correspond to a local grid event origin; locating the temporal parts of the signal using traveling wave fault location methods; identifying and classifying a fault / event situation (806) based on current and / or voltage values of the mains frequency and its harmonics; determining that high-frequency signals (804), correlating in time with the fault / event situation, are of local origin and not atmospheric, and therefore originating from the fault point; locating the grid fault / event using traveling wave fault location methods using the identified traveling waves of the local origin correlating with the fault / event situation; andnotifying grid operators and / or service crews of the nature and location of the grid fault event, the notification being provided through an event log, a map view (500) showing the grid topology, or via integrations with an outage management system.

19. A method according to claim 18, wherein the signal records and / or metadata consisting of time-stamps, traveling wave analysis results and mains frequency signal analysis results, including event classifications and locations, are stored in a database, and this collected history data is later utilized, using statistical or heuristic analysis, for example by counting events or event types correlating with lightning strikes per grid section, to identify sections in electrical grid which are most prone to impact of lightning strikes.

20. A system (600) for detecting a traveling wave in an electric grid (100), the system comprising: a set of sensors (104a-104c, 104a'-104c'), comprising: a first set of sensors (104a-104c) configured to be positioned at locations (108a-108c) in a first section (100A) of the electric grid, fed by a transformer at a first primary substation, and a second set of sensors (104a'-104c'), configured to be positioned at locations (108a'-108c') in a second section (100B) of the electric grid, wherein each sensor of the set of sensors being configured to monitor one or more phase conductors and / or neutral conductors, and / or ground conductors of a transmission line or a distribution line, of the electric grid, and detect a set of signal data, comprising a set of amplitudes and a set of arrival times; a processor (602), configured to be associated with the set of sensors (104a-104c, 104a'-104c'), the processor (602) being configured to receive the set of signal data from the set of sensors, and configured to:determine the presence of a traveling wave propagating along a transmission line (102, 102') of the electric grid based on the received signal data; categorize the traveling wave as either a local event, an atmospheric event, such as a lightning strike, wherein the processor is configured to; record waveforms of the detected traveling waves; analyze whether the detected traveling waves exhibit similar amplitude sequences and arrival times across the first set of sensors and / or the second set of sensors (104a-104c, 104a'-104c'); distinguish between a locally detectable traveling wave, where the traveling wave is detected either by any or several of the sensors in the first set of sensors (104a-104c) or any or several of the sensors in the second set of sensors (104a'-104c'), and / or whether a traveling wave detectable by any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors, and identify the source of the traveling wave as a local grid event if the traveling wave is detected only with any or several of the sensors in the first set of sensors or only with any or several of the sensors in the second set of sensors and identify an atmospheric event, such as a lightning strike, if the traveling wave is detected by any or several of the sensors in the first set of sensors and any or several of the sensors in the second set of sensors.

21. A system (600) according to claim 20, wherein the first section of the electric grid and the second section of the electric grid are galvanically isolated.

22. A system (600) according to claim 20 or 21, wherein the first section of the electric grid is fed by a transformer at a first primary substation and the second section of the electric grid is fed by a transformer at asecond primary substation, which second primary substation is distinct from the first primary substation.

23. A system (600) according to any of the preceding claims 20-22, wherein the processor is configured to mathematically correlate the waveforms of the detected traveling waves, filtered waveforms of the detected traveling waves, and / or waveform envelopes of the detected traveling waves.

24. A system (600) according to claim any of the preceding 20-23, wherein the processor is configured to categorize the non-lightning induced traveling waves by amplitude attenuation and frequency analysis compared to atmospheric events.

25. A system (600) according to any of the preceding claims 20-24, wherein the processor is configured to analyze variations in the magnetic, electric, or electromagnetic field surrounding the transmission line or distribution line, or current or voltage detected by any or several of the sensors in the set of sensors (104a-104c, 104a'-104c').

26. A system (600) according to any of the preceding claims 20-25, wherein the analysis whether the detected traveling wave exhibits similar amplitude sequences and arrival times across the set of sensors (104a- 104c, 104a'-104c') is executed within a predefined temporal threshold.

27. A system (600) according to any of the preceding 20-26, wherein the processor is configured to determine whether the detected traveling wave exhibits consistent amplitude sequences and arrival times across the first set of sensors (104a-104c) and / or the second set of sensors (104a'- 104c'), within a predefined amplitude tolerance range established based on signal variability.

28. A system (600) according to any of the preceding claims 20-27, wherein the processor is configured to determine if the traveling wavethat was detectable only locally involves verifying that all sensors among the set of sensors (104a-104c, 104a'-104c') detecting the signal and transmitting the corresponding signal data are located within a predefined local area of the electric grid.

29. A system (600) according to any of the preceding claims 20-28, wherein the processor is configured to determine if the traveling wave signal that was detectable only locally is based on the sensors among the set of sensors (104a-104c, 104a'-104c') detecting the signal being positioned in a section of the electric grid that is galvanically connected.

30. A system (600) according to any of the preceding claims 20-29, wherein the processor is configured to determine if the traveling wave signal that was detectable only locally is based on the proximity of all detecting sensors among the set of sensors (104a-104c, 104a'-104c'), ensuring no sensor is located beyond a predefined maximum distance, determined based on physical and operational constraints of the electric grid.

31. A system (600) according to any of the preceding claims 20-30, wherein the predefined temporal threshold is selectable for each sensor pair, or sensor group, among the set of sensors (104a-104c, 104a'- 104c'), accounting for time measurement errors and signal propagation delays.

32. A system (600) according to any of the preceding claims 20-31, wherein similarity of amplitude sequences and arrival times of detected traveling waves is determined by calculating the maximum propagation time of traveling waves along the electric grid topology, considering signal velocity in conductors and sensor locations.

33. A system (600) according to any of the preceding claims 20-32, wherein the processor analyzing comprises recording waveforms of thedetected traveling waves and comparing the recorded waveforms or filtered waveforms to historical databases of validated lightning event recordings using correlation methods or machine learning models, wherein the machine learning models are trained on a dataset comprising historical traveling wave recordings from both atmospheric and non- atmospheric events.

34. A system (600) according to any of the preceding claims 20-33, wherein the processor is further configured to estimate a straight-line distance between at least one sensor (104a-104c, 104a'-104c') and the event location and compare the estimated straight-line distance to a line distance, calculated along the grid lines between the at least one sensor (104a-104c, 104a'-104c') and the event location; and, based on comparison, determine whether the event is an atmospheric event or a local grid event.

35. A system (600) according to any of the preceding claims 20-34, wherein determining, by the processor, if the traveling wave signal was originated by a local grid event, or if the traveling wave signal was of atmospheric event such as a lightning strike, includes fitting the arrival times of the signals as if the signals would follow the paths and signal speed of an electric grid topology and fitting the signal amplitude sequences and their respective times of arrivals to a straight-line paths over the air i.e. "as the crow flies", and if the signal arrival times fit better to the paths of a grid topology rather than straight lines at speed of light from a single event location, the event is of a local grid origin, and if the signal arrival times better fit to a model of straight lines at speed of light from an event location, the event is of an atmospheric origin, such as a lightning strike.

36. A system (600) according to any of the preceding claims 19-35, wherein the processor is configured to render a user interface element (502), indicative of a location (106) of occurrence of the lightning strike,on a user interface (500) of a map application that is operable to render a map of a region that includes the electric grid (100), wherein the user interface indicates the set of locations (504a-504f).

Images