Electric power system monitoring

Abstract

A monitoring device (10) for an electric power system, comprising a vector network analyser (110), a network interface connection (120) and a controller (140), wherein: the network interface connection (120) is configured to electrically couple the vector network analyser (110) with the electric power system at a single location thereof and filter out a network power frequency of the electric power system from signals received by the vector network analyser (110); the vector network analyser (110) is configured to perform a spectral frequency analysis of the electric power system on at least one frequency range that is higher than the network power frequency, to obtain a frequency response of the electric power system; the controller (140) is configured to, while the network interface connection (120) electrically couples the vector network analyser (110) with the electric power system, cause the vector network analyser (110) to obtain a first frequency response, cause the vector network analyser (110) to obtain at least one subsequent frequency response, and detect a change in state of the electric power system based on a comparison of the at least one subsequent frequency response with the first frequency response; whereby the monitoring device (10) results to be a single device that monitors and analyses the electric power system, radially and in an asynchronous polling mode, from the single location where the vector network analyser (110) is electrically coupled with said electric power system, to detect faults and state variations of said electric power system.

Description

[0001] ELECTRIC POWER SYSTEM MONITORING

[0002] Cross-Reference to Related Applications

[0003] This Patent Application claims priority from Italian Patent Application No. 102023000011991 filed on June 12, 2023, the entire disclosure of which is incorporated herein by reference .

[0004] Field of the Invention

[0005] Example aspects herein relate to the monitoring of electric power systems, and more specifically to a monitoring device, a method for monitoring of an electric power system, and a computer program.

[0006] Background

[0007] Electric power systems are utilised to transmit / distribute alternating current (AC) or direct current (DC) electrical power from various sources (e.g. thermal, solar, wind power sources) to users, using electric lines (e.g. overhead (aerial) or underground lines that may be transmission or distribution lines) , transformers, and other electrical elements. To ensure an efficient transmission and distribution of the electrical energy (or power) , electric power systems can be monitored to detect faults to be repaired, or a modification in connections (e.g. connection with a source of electrical energy and / or with an electrical load i.e. an intended destination of the electrical energy) .

[0008] For example, current and voltage characteristics (e.g. homopolar or phase current / voltage ) may be measured (e.g. at electric stations or substations) to detect an issue in the electric power system or a change in connections to the electric power system.

[0009] However, not all faults or changes in connections would result in a noticeable change in the current or voltage characteristics .

[0010] There is therefore a need to improve the monitoring of electric power systems. In addition, the supply of the electrical energy may be lost ( e . g . due to a power outage , also referred to as a power loss or blackout ) , which prevents the measuring of current or voltage characteristics . It is also possible that more than one issue occurs in the electric power system concurrently, for example when a natural disaster causes damage to a wider geographical area . Even i f the power is restored, a previously undetected fault can cause delays in restoring the power supply, or a previously undetected modi fication in the connections cause a reduced amount of power being obtained and transmitted / distributed . It is however di f ficult to assess the condition of the electric power system over the wider geographical area, as this requires human technicians to control equipment at multiple locations . During emergency events , where maj or power cuts of the high voltage grid managed by a Transmission System Operator ( TSO) can last several hours , the only method currently available to determine the status of overhead medium voltage lines is to perform visual inspections .

[0011] In particular, while the high voltage grid is absent , medium power lines may present a fault ( e . g . due to fallen trees , ground conductors , conductor breaking due to snow burden, etc . ) . In that case , protection relays such as in case of fault on medium power lines are all unavailable , and therefore no protection mechanism is triggered .

[0012] Thus , a Distribution System Operator ( DSO) cannot take any actions besides performing visual inspections during a blackout of the high voltage grid managed by the TSO . Consequently, medium voltage lines condition remains unknown .

[0013] It would therefore be particularly advantageous to improve the monitoring of an electric power system even in the absence of electrical energy in the electric power system .

[0014] In this context , US 2012 / 271574 Al discloses various realtime and of f-line tools for monitoring and analysis of power system components , such as a method that includes obtaining raw power system data associated with a power system, crosscorrelating the raw power system data with a synchroni zed pseudo-random sequence signal inj ected into the power system to determine a correlated impulse response and determining a condition of the power system based at least in part upon the correlated impulse response . Additionally, US 2012 / 271574 Al further discloses a system that includes a plurality of signal inj ection systems and a data capture device coupled to a power system . A data analysis device cross-correlates raw power system data obtained by the data capture device with at least one synchroni zed pseudo-random sequence signal inj ected by a signal inj ection system and determines a condition of the power system based at least in part upon a frequency spectrum based upon a correlated impulse response .

[0015] Summary of Invention

[0016] According to a first example aspect herein, there is provided a monitoring device for an electric power system, the monitoring device comprising a vector network analyser, VNA, a network interface connection and a controller, wherein :

[0017] • the network interface connection is configured to

[0018] - electrically couple the VNA with the electric power system at a location thereof and

[0019] - filter out a network power frequency of the electric power system from signals received by the VNA;

[0020] • the VNA is configured to perform a spectral frequency analysis of the electric power system on at least one frequency range that is higher than the network power frequency, to obtain a frequency response of the electric power system;

[0021] • the controller is configured to , while the network interface connection electrically couples the VNA with the electric power system,

[0022] - cause the VNA to obtain a first frequency response ,

[0023] - cause the VNA to obtain at least one subsequent frequency response , and

[0024] - detect a change in state of the electric power system based on a comparison of the at least one subsequent frequency response with the first frequency response ; whereby the monitoring device monitors the electric power system radially from the location where the VNA is electrically coupled with said electric power system .

[0025] The electric power system, which may comprise , for example , one or more electric power network, carries the electrical energy ( or electrical power ) from one or more power sources to one or more destinations ( e . g . other electric power systems or users , such as factories , domestic users , etc . ) . The electric power system may be a monophase or multiphase ( e . g . having three ) AC electric power system carrying electrical power at a set frequency, such as 50 Hz or 60 Hz (based for example on the geographic location) . This frequency is typically defined as the power frequency of the electric power system (hereinafter network power frequency) . In case of an electric power system carrying DC electrical power, the network power frequency may be defined to be substantially 0 Hz . In any embodiment described herein, the electric power system may be a transmission and / or distribution power system .

[0026] The monitoring of the electric power system allows changes in a state of the electric power system to be detected . This change may indicate that the electric power system may not be carrying the electrical power in an ef ficient way towards the intended destination . For example , a fault may be preventing the electrical energy to be delivered to some destinations .

[0027] Similarly, a change in the electric power system conditions , such as the power transmission characteristics of the electric power system ( and speci fically, of elements such as the electric lines ) may change due to damage , wear, or other factors , leading to a loss in the ef ficient provision of the electrical energy .

[0028] As another example , a modi fication in connections ( e . g . to another electric power system, to a power source, or to a destination) may af fect the transmission / distribution of electrical energy . Any of these types of modi fications in connections to the electric power system may be preventing the electrical power to be carried by the electric power system in an efficient way towards the intended destination ( s ) .

[0029] In the present case, the change in state may be defined to be a change other than a substantial loss of electrical power at the network fundamental frequency in the electric power system (in other words, a change other than a power outage or blackout) .

[0030] The network interface connection, which may also be referred to as a network connector, or a network interface, electrically couples (i.e. connects) by allowing physical connection whilst avoiding dangerous voltage transfer between the electric power system and VNA at the network power frequency. The network interface connection may be coupling the VNA to at least one phase of the electric power system, or to each phase of the electric power system. The network interface connection may, in addition, be coupling the VNA to a reference point (e.g. electrical ground) , which may be connected to an element of the electric power system or outside the electric power system. Thus, the network interface connection may create phase-to-phase and / or phase-to-ground coupling ( s ) .

[0031] The specific location may be, for example, a location on transmission or distribution line (preferably an overhead transmission / distribution line, but may be an underground line, or any other type of electric line) , at or near connections to electric power sub-systems (e.g main dichotomies) , near connections to other electric power systems, etc. The monitoring device may therefore monitor the electric power system to detect any change in state observable from that specific location. The monitoring device monitors the electric power system radially from the specific location. That is to say, the monitoring is not limited to a specific direction, such as an upstream direction (i.e. towards a power source) or downstream direction (i.e. towards a destination for the electrical power) . The network interface connection may be configured to form a permanent coupling ( for example by using permanent af fixing means such as mechanical fastening on crossarms , or a semipermanent coupling ( for example by using means detachable af fixing means such as bolts , screws etc . ) . By using a permanent or semi-permanent coupling, the connection of the VNA to the electric power system should not change between successive spectral frequency analyses , keeping the speci fic location where the monitoring device is electrically coupled to the electric power system, the first frequency response and the subsequent frequency response ( s ) obtained by the VNA should not di f fer unless a physical change in state elsewhere in the electric power system occurs .

[0032] It would be understood by the person skilled in the art that characteristics of components of the network interface connection may be selected for compatibility with the electric power system to which the network interface connection is to be coupled . For example , these characteristics of the components may include an insulation voltage level that allows components to withstand the voltage level of the electrical energy carried by the electric power system .

[0033] Preferably, an insulation voltage level of at least one component of the network interface connection coupled to the electric power system may be at least about I kV ( e . g . for coupling to a low-voltage electric power system) . Preferably ( e . g . for coupling to a medium- voltage electric power system) , the insulation voltage level is at least about l OkV, preferably at least about 15kV, more preferably at least about 20kV .

[0034] In the electrical coupling, the network interface connection filters out the network power frequency, for example by using at least on filter circuit ( e . g . t- filter, low-pass filter, high-pass filter, bandpass filter or bandstop filter ) to decouple components of the monitoring device from the electric power system at or around the network power frequency . In other word, the network interface connection attenuates electric signals in the electric power system at or around the network power frequency from reaching other components of the monitoring device such as the VNA. For example , the network interface connection may comprise at least one high-pass filter with a cut-of f frequency above the network power frequency ( e . g . 100 Hz or 1000 Hz , i f the network power frequency is 60 Hz or below) , a band-stop or notch filter to attenuate electrical signals of a frequency around the network power frequency ( e . g . with cut-of f frequencies of 10 Hz and 1000Hz , i f the network power frequency is 50 or 60 Hz ) .

[0035] As the network interface connection filters out the network power frequency, it can be understood to mean that the network interface connection de-couples the monitoring device ( and in particular the VNA) from the electric power system at the network power frequency . Accordingly, trans fers of voltages at the network power frequency from the electric power system towards the VNA may be avoided, as these voltages may be at levels that are dangerous for the VNA, other elements of the monitoring device , or a human operator in proximity to the monitoring device .

[0036] Preferably, the network interface connection comprises a T- filter .

[0037] The VNA may be any analyser configured to perform a frequency analysis , such as a spectral frequency analysis ( or a frequency sweep analysis , a frequency range analysis , etc . ) and obtain a frequency response . For example , the VNA may be generating a frequency sweep signal ( i . e . a signal at a plurality of frequencies in a frequency range ) transmitted to the electric power system via the network interface connection, and makes measurement ( s ) on the signal returned from the electric power system at one or more frequencies . Based on the measurements , the VNA obtains a frequency response of the electric power system . The frequency response may comprise one or more parameters ( or electrical characteristic ) related to the electric power system such as s-parameters , y-parameters , z-parameters , h-parameters . The spectral frequency analysis is performed on at least one frequency range that is higher than the network power frequency, thus may be defined as a high- frequency analysis.

[0038] Preferably, a frequency response may indicate one or more specific frequency value. The frequency response may comprise, for each or a subset of the specific frequency values, one or more electrical characteristics (or parameters) , such as one or more of: an amplitude value of the response signal, an impedance value (resistance or complex impedance) , a return loss value, reflection coefficient, a voltage standing wave ratio (VSWR) , a quality factor, a series inductance value (series L) , a series capacitance value (series C) , a parallel resistance value (parallel R) , a parallel inductance value (parallel X) , an Sil phase value, a S21 gain value, and a S21 phase value. If the frequency response indicates multiple frequency values, the frequency response may comprise different parameters associated with different specific frequency values.

[0039] The first frequency response may be a reference frequency response, defining a response of the electric power system under normal conditions (e.g. without a fault and with a number of connections) , which can be defined as an open-loop response of the electric power system.

[0040] Each subsequent frequency response is obtained at a time subsequent to the first frequency response, and any reference to a subsequent frequency response may instead be referred to as "second frequency response" to differentiate them from the first frequency response.

[0041] The VNA may be configured to obtain a subsequent frequency response recurringly (e.g. periodically or at predetermined time intervals) , and / or when a triggered by the occurrence of an event such as the reception of a command, the determination that no spectral frequency analysis is being performed by another entity on the electric power system (e.g. based on a polling method) etc. Each subsequent frequency response may be based on a spectral frequency analysis on the same (one or more) frequency range (s) , or they may be based on spectral frequency analyses performed on different frequency range (s) from each other.

[0042] The controller may comprise hardware elements such as one or more processor (e.g. a single / multiple core CPU, one or more microprocessors etc.) , one or more working memories (e.g. random-access memory, RAM, flash memory etc.) and one or more non-volatile instructions stores (e.g. read-only memory (ROM) , programmable ROM (PROM) , erasable PROM (EPROM) , electrically erasable PROM (EEPROM) , flash memory, etc.) storing computer- readable instructions. The processor (s) execute the computer- readable instructions in the instruction store (s) to control various elements of the monitoring device (e.g. by transmitting one or more control signals) and in particular, to cause the VNA to obtain the first frequency response and the at least one subsequent frequency response, and to detect the change in state of the electric power system based on the comparison of the frequency responses. The controller may also control protection means in the monitoring device, such as protection relays, and / or other elements of the monitoring device.

[0043] The comparison of the at least one subsequent frequency response and the first frequency response may comprise a comparative spectral analysis of each frequency response, a pattern matching, or other comparative processing, which may be performed in the VNA (in which case the controller obtains a result of the comparison from the VNA) , in the controller (in which case the controller obtains information related to the frequency responses to be compared from the VNA) , or both (i.e. if the controller obtains from the VNA an intermediate result comparing the frequency responses, further processes the intermediate result to detect the change in state of the electric power system) .

[0044] The comparison may include performing one-to-one comparison ( s ) (e.g. separate comparisons between the first frequency response and each subsequent frequency responses) , and / or one-to-many comparison ( s ) comparing the first frequency response with a combination of a plurality of subsequent frequency responses (e.g. an average, weighted average, etc.) , such as by obtaining a number of subsequent frequency responses and comparing these in a First-In, First-Out (FIFO) order (i.e. where the subsequent frequency responses are compared with the first frequency response in the same order as they are obtained) or a Last-In, First-Out (LIFO) order where the subsequent frequency response that is obtained last is compared with the first frequency response first.

[0045] The frequency responses may indicate one or more value on a logarithmic scale (e.g. the frequency values, return loss, gain values, etc.) . Accordingly, the comparison of such value may also be performed on a logarithmic scale as well.

[0046] The controller may generate signals causing the VNA to perform a spectral frequency analysis, and obtains in return a frequency response from the VNA. The controller may cause the storage of the information on the frequency response (e.g. in a memory of the controller or in the monitoring device) ; specifically, the controller may be configured to cause the storage of the first frequency response in a memory of the monitoring device, such as a non-volatile memory.

[0047] Accordingly, the monitoring device can detect a change in state of the electric power system, which may be used to determine whether the electrical power is efficiently distributed / transmitted from electrical sources to intended destinations. With the monitoring device, the state of the electric power system is analysed in a frequency domain, rather than a voltage / current domain.

[0048] Furthermore, as the network interface connection filters out the network power frequency from signals received by the VNA, the frequency responses obtained by the VNA can be used independently of whether the electric power system carries electrical power (or, for example, the electric power system experiences a partial or total blackout) . In addition, by filtering out the network power frequency, the risk that the VNA is saturated by the electric power carried by the electric power system (which may occur if the amplitude of the signal at the network power frequency is greater than the signal generated for the spectral frequency analysis) can be reduced.

[0049] By electrically coupling the monitoring device at a specific location to obtain multiple frequency responses, the monitoring device can detect a change in state both in an upstream direction (i.e. towards a power source) and downstream direction (i.e. towards a destination for the electrical power) relative to the monitoring device. In other words, the monitoring device can detect a change in state caused by an event (e.g. a fault, short-circuit, a new connection to the electric power system or the removal of an existing connection) that occurs at a location in either upstream or downstream direction from the monitoring device. The terms downstream and upstream need may be defined based on a prevalent direction of transmission / distribution of the electrical power, which may change over time (for example if one or more electrical sources are connected or disconnected from the electric power system) .

[0050] The monitoring device may therefore monitor the electric power system to detect any change in state observable from that specific location. Accordingly, the monitoring device may be installed at a location that is remote or difficult to access, and monitor the electric power system in case of a blackout, to detect a change in state of the electric power system without requiring an operator or technician to be physically present at the location to be monitored.

[0051] Further, by coupling multiple monitoring devices to be coupled to the same electric power system at different locations, each monitoring device can be set to monitor different parts of the same electric power system.

[0052] Preferably, the change in state indicates a fault or a modification to connections to the electric power system.

[0053] For example, the fault may be a monophase or multi-phase failure due to a line coming in contact with the ground or electrically conductive elements external to the electric power system, or two phases coming in contact with each other (i.e. creating a short-circuit) . The modification to connections may be the disconnection, or the deterioration in connection of a power source or a power destination from the electric power system, or a new element being connected to the electric power system and drawing electrical power away from the intended destinations , thus leading to energy losses ( for example in cases of an erroneous or fraudulent connection to the electric power system) . Thus , by obtaining the first frequency response when the electrical is under normal conditions ( i . e . with no fault and with the expected connections to the electric power system) , the first frequency response can be used as a reference against which subsequent frequency responses are compared .

[0054] Accordingly, the detection of this change in state can be used to identi fy a loss of ef ficiency in distributing or transmitting the electrical power .

[0055] Preferably, the network interface connection comprises a band-pass or high-pass filter, preferably a T- filter, to filter out the network power frequency .

[0056] As these filters can operate on a wide range of voltage , the network interface with such filter allows the monitoring device to be coupled to a medium voltage electric power system ( e . g . an electric power system carrying electrical power in a range of voltage usually defined to be about I kV to 30kV) , or a lower voltage electric power system ( e . g . an electric power system carrying electrical power at voltage equal to or below about I kV) .

[0057] Preferably, the controller is configured to detect a change in state of the electric power system based on a deviation in amplitude and / or a frequency shi ft between the first frequency response and the at least one subsequent frequency response .

[0058] For example , the first frequency response may indicate a frequency value for which an electrical characteristic ( e . g . a return loss ) is a minimum over the range of frequency on which the spectral frequency analysis is performed . Then, a subsequent frequency response may indicate that the frequency value for which the electrical characteristic is a minimum has shi fted, such as a shi ft by at least a predetermined threshold, or by at least a predetermined percentage , by at least a ratio ( linear, logarithmic, etc . ) . In that case the controller may detect the change in state of the electric power system based on the frequency shi ft .

[0059] The first frequency response may, alternatively or additionally, indicate an amplitude value of an electrical characteristic ( e . g . a return loss in dB ) for a speci fic frequency ( e . g . a predetermined frequency value and / or a signi ficant frequency value such as the frequency value where the return loss is the minimum) . Then, i f in a subsequent frequency response , the amplitude value of the electrical characteristic at that speci fic frequency deviates from the amplitude value indicated in the first frequency response such as a deviation by at least a predetermined threshold, or by at least a predetermined percentage , by at least a ratio ( linear, logarithmic, etc . ) , the controller may detect the change in state of the electric power system based on the deviation in amplitude .

[0060] Preferably, the controller is configured to determine , from the first frequency response , a first value of at least one electrical characteristic . This at least one electrical characteristic may be related to the electric power system, and may be any of the electrical characteristics indicated above .

[0061] The controller may determine the first value by analysing the first frequency response , or by obtaining the first value from the VNA ( e . g . by receiving it from the VNA, i f the VNA outputs the first value for one or more of the electrical characteristic ) .

[0062] Preferably, the controller is configured to determine , from the at least one subsequent frequency response , a second value of the at least one electrical characteristic, and detect the change in state of the electric power system based on a comparison of each first value and the corresponding second value . Accordingly, the change in state of the electric power system may be detected based on a comparison of values ( the first and second value ( s ) ) , rather than entire frequency responses .

[0063] Preferably, the at least one electrical characteristic comprises a frequency at which the frequency response has a minimum return loss , the comparison of each first value and the corresponding second value comprises a comparison of a first frequency at which the first frequency response has a minimum return loss and a second frequency at which the at least one subsequent frequency response has a minimum return loss .

[0064] Accordingly, the change in state of the electric power system may be detected based on a change in the minimum return loss measured by the VNA.

[0065] Preferably, the controller is configured to cause the VNA to obtain the at least one subsequent frequency response on one or more second frequency ranges , the one or more second frequency ranges being determined based on the first value of at least one electrical characteristic .

[0066] For example , each first value may have one or more corresponding frequency value ( e . g . a frequency value where the first value of the electrical characteristic was obtained) . Each subsequent frequency response can be based on a spectral frequency analysis on second frequency range ( s ) that comprise the frequency values corresponding to each first value .

[0067] As another example , each of a plurality of second frequency ranges may compri se one or more of the corresponding frequency values , and each second frequency response may be based on a respective subset of the second frequency ranges . For example , the VNA may be arranged to alternate between obtaining a subsequent frequency response on a first subset of the second frequency ranges , and a second, di f ferent subset of the second frequency ranges (where the first and second subsets may have common elements or each second frequency range is in one and only one of the first and second subsets ) . It would be understood that the VNA may be arranged to alternate between three or more subsets of the second frequency ranges instead.

[0068] Accordingly, each subsequent frequency response is based on a subset of the second frequency ranges and may therefore be performed faster and / or with fewer resources (e.g. energy and / or processing resources) .

[0069] Preferably, the controller is configured to cause the VNA to obtain the first frequency response on a first frequency range including one or more first frequencies, and to cause the VNA to obtain the at least one subsequent frequency responses on one or more second frequency ranges, each second frequency ranges comprising a respective one of the one or more first frequencies, and each second frequency range being narrower than the first frequency range.

[0070] For example, each first frequency may be a frequency representative of the first frequency response (e.g. a frequency corresponding to a minima, a maxima, or a change in the sign of derivative, etc. for an electrical characteristic) , which may be indicative of a change in state of the electric power system.

[0071] In some cases, the first frequency range (s) may comprise the second frequency ranges, and the second frequency ranges may have no substantial overlap between each other.

[0072] Accordingly, the subsequent frequency response (s) may be based on a spectral frequency analysis performed on narrower frequency range (s) than those on which the first frequency response is based (e.g. on frequency range (s) each centred on a respective one of the first frequency) . Thus, by reducing the frequency range (s) used to obtain the subsequent frequency response (s) , each subsequent frequency response may be performed faster and / or with fewer resources (e.g. energy and / or processing resources) .

[0073] Preferably, the controller is configured to, upon detecting a predetermined trigger, cause the VNA to obtain the first frequency response again, and to detect the change in state of the electric power system using the new first frequency response.

[0074] For example, the predetermined trigger may be any one of: detecting that the device is connected to the electric power system (e.g. a calibration at start-up) , receiving an update instruction (e.g. an instruction received from a remote device) , or an input by a human operator located near the monitoring device (e.g. a physical switch or button being pressed) , a periodic trigger (e.g. based on a periodic activation of a re-calibration of the device, which may be based on a timer running in the controller or elsewhere in the monitoring device) .

[0075] Once the new first frequency response is obtained, it may be used as a reference for comparing any subsequently obtained frequency response (i.e. any frequency response obtained by the VNA after obtaining the new first frequency response) .

[0076] The new first frequency response may also be used in other ways. For example, the previously obtained first frequency response and the new first frequency response may be used to train a machine learning model, such as one comprising a neural network, to detect a change in state of the electric power system from the subsequent frequency response (s) . In particular, the VNA may recurringly obtain new first frequency responses to be used as a training set for the machine learning model .

[0077] As another example, the new first frequency response could be compared with the previously obtained first frequency response. If the controller detects no substantial change (e.g. by more than a threshold) between the previously obtained first frequency response and the new frequency response, one of the two first frequency responses may be discarded (e.g. the new first frequency response) , and the other one is kept (e.g. the previously obtained first frequency response) to compare with subsequently obtained frequency responses (or vice-versa) .

[0078] Preferably, the controller is configured to, upon detecting the change in state of the electric power system, cause the generation of an alert noti fication, or to directly command a protection relay .

[0079] The alert noti fication may be transmitted to a remote device , such as a device used in a centralised control of the electric power system, via one or more suitable communication link . Each communication link may be , for example , a wireless communication link ( for example a Wi-Fi , cellular telephone data link such as LTE / 5G, Bluetooth or Bluetooth Low Energy (BLE ) , LoRa ( Long Range loT ) , LEO ( Low Earth Orbit ) satellite link) or a wired communication link ( e . g . DSL, fibre-optic cable , Ethernet etc . ) . Each communication link may not be permanent .

[0080] Accordingly, the alert noti fication may help improve an investigation of the detected change in state of the electric power system and any required action .

[0081] In one embodiment , the power supply unit is configured to supply power from a power source which is not electrically coupled to the electric power system .

[0082] For example , the power supply unit may be powered by one or more power sources such as a battery, a renewable ( e . g . solar, wind or water based) energy source , a fuel cell , etc . or any combination thereof , to power elements of the monitoring device such as the controller and the VNA.

[0083] Accordingly, the monitoring device may operate regardless of whether the electric power system is carrying electrical energy . Additionally, any fault such as a power surge in the electric power system is less likely to af fect the powering of the monitoring device .

[0084] Preferably, the electric power system is a low voltage energy distribution electric power system or a medium voltage energy distribution electric power system, wherein the network interface connection is configured to electrically couple the VNA with either the low voltage energy distribution electric power system or the medium voltage energy distribution electric power system . Preferably, the monitoring device is coupled to a multiphase electric power system ( i . e . the electric power system has a plurality of phases ) , wherein the monitoring device comprises a plurality of network interface connection each for a respective one of the phases of the electric power system . More preferably, the monitoring device comprises a network interface connection for each of the phases of the electric power system .

[0085] Each of the plurality of network interface connection may electrically couple the VNA to respective phase of the electric power system at the speci fic location, and filter out the network power frequency from signals received by the VNA. In other words , each network interface connection may have a corresponding phase of the electric power system ( i . e . a corresponding monitored phase ) .

[0086] Accordingly, each phase of the electric power system coupled to the VNA via a network interface connection may be monitored for a change in a state of the electric power system .

[0087] By monitoring multiple phases ( and more preferably all phases ) of the electric power system, faults that af fect only one ( or a subset ) of phases of the multiphase electric power system may be detected by the monitoring device .

[0088] Preferably, the controller is configured to cause the VNA to obtain a first frequency response , and to obtain at least one subsequent frequency response, for each phase corresponding to a network interface connection, and more preferably for each phase of the electric power system .

[0089] For example , the controller may cause the VNA to obtain a respective first frequency response for each monitored phase of the electric power system, and to cause each respective first frequency response to be stored in memory . Then, the controller may cause the VNA to obtain at least one subsequent frequency response for each monitored phase of the electric power system . The controller may then detect a change in a state of the electric power system based on a comparison of the at least one subsequent frequency response with the first frequency response obtained for the same phase of the electric power system . The VNA may sequentially obtain frequency response for each monitored phase , or may obtain frequency responses for a plurality of the monitored phases ( e . g . i f the VNA includes multiple channels ) .

[0090] Preferably, the monitoring device comprises at least one switch connecting each of the plurality of network interface connection to the VNA, wherein the controller is configured to control a state of the switch to determine a phase of the electric power system to be coupled to the VNA.

[0091] Accordingly, the controller may select which of the phases of the electric power system is coupled to the VNA.

[0092] Preferably, the VNA comprises a plurality of channels each for performing a spectral frequency analysis of a respective one of the phases of the electric power system.

[0093] Accordingly, the VNA may obtain frequency responses from multiple phases of the electric power system contemporaneously .

[0094] According to a second example aspect herein, there is provided a method for monitoring an electric power system, the method comprising : obtaining a first frequency response of the electric power system at a speci fic location, wherein the first frequency response is based on a spectral frequency analysis on at least one frequency range , each frequency range being higher than a frequency of electrical energy carried by the electric power system, hereinafter network power frequency; obtaining at least one subsequent frequency response of the electric power system at the speci fic location, wherein each subsequent frequency response is based on a spectral frequency analysis on at least one frequency range , each frequency range being higher than the network power frequency; and detecting a change in state of the electric power system based on a comparison of the at least one subsequent frequency response with the first frequency response .

[0095] Preferably, the obtaining the first frequency response and the at least one subsequent frequency response comprises electrically coupling a vector network analyser, VNA, to the electric power system at the speci fic location, via a network interface connection configured to filter out the network power frequency from signals received by the VNA.

[0096] Preferably, the method may be performed by the monitoring device according to the first example aspect described above .

[0097] Preferably, the obtaining the at least one subsequent frequency response and / or the detecting the change in state of the electric power system is performed independently of whether the electric power system carries electrical energy . In other words , whether the electric power system is live or dead at the speci fic location ( e . g . whether the speci fic location corresponds to a live line or a dead line ) .

[0098] The electrical energy is , for example , the energy carried at the network power frequency to be distributed to intended destinations .

[0099] Preferably, the method comprises calibrating the VNA after coupling the VNA to the electric power system .

[0100] By calibrating the VNA, changes in a state of the electric power system ( e . g . fluctuations in amplitude of a frequency response ) may be more accurately detected .

[0101] Preferably, the first frequency response and the at least one subsequent frequency response are obtained for each of a plurality of phases of a multiphase electric power system . More preferably, the first frequency response and the at least one subsequent frequency response are obtained for each phase of the multiphase electric power system . The subsequent frequency response ( s ) for a phase is compared to the first frequency response obtained for the same phase, to detect the change in state of the electric power system .

[0102] According to a third example aspect herein, there is provided a computer program comprising instructions which, when executed by at least one processor, cause the at least one processor to perform the method according to the second example aspect described above , wherein the at least one processor is communicatively coupled to a vector network analyser, VNA, the VNA is electrically coupled to the electric power system at the speci fic location via a network interface connection, the network interface connection is configured to filter out the network power frequency, from signals received by the VNA.

[0103] According to a fourth example aspect herein, there is provided a method for monitoring an electric power system, wherein a vector network analyser, VNA, is electrically coupled to the electric power system at a speci fic location via a network interface connection, the network interface connection being configured to filter out a frequency of electrical energy carried by the electric power system, hereinafter network power frequency, from signals received by the VNA, the method comprising : causing the VNA to obtain a first frequency response of the electric power system at a speci fic location, wherein the first frequency response is based on a frequency analysis on at least one frequency range , each frequency range being higher than the network power frequency; caus ing the VNA to obtain at least one subsequent frequency response of the electric power system at the speci fic location, wherein each subsequent frequency response is based on a frequency analysis on at least one frequency range , each frequency range being higher than the network power frequency; and detecting a change in state of the electric power system based on a comparison of the at least one subsequent frequency response with the first frequency response .

[0104] According to a fi fth example aspect herein, there is provided a computer program comprising instructions which, when executed by at least one processor, cause the at least one processor to perform the method according to the fourth example aspect described above .

[0105] According to a sixth example aspect herein, there is provided a system comprising the electric power system and at least one monitoring device according to the first example aspect described above . Preferably, the system comprises two or more monitoring devices , each electrically coupled to the electric power system at respective locations that are separate from each other, each monitoring device being configured to perform a spectral frequency analysis on di f ferent parts of the electric power system .

[0106] According to a further example aspect herein, there is provided a computer-readable storage medium storing the computer program according to the third example aspect and / or the fi fth example aspect .

[0107] According to a further example aspect herein, there is provided a signal carrying the computer program according to the third example aspect .

[0108] List of Figures

[0109] Embodiments of the present invention, which are presented for better understanding the inventive concepts , but which are not to be seen as limiting the invention, will now be described with reference to the figures in which :

[0110] Figure 1A shows a schematic diagram showing an example of a monitoring device in example embodiments ;

[0111] Figure IB shows an example of installation of the monitoring device ;

[0112] Figure 2 shows a schematic diagram a schematic diagram showing an exemplary implementation of the controller of the monitoring device ;

[0113] Figure 3A shows a schematic diagram showing an exemplary implementation of the network interface connection of the monitoring device ;

[0114] Figure 3B shows a schematic diagram showing an example of a frequency response of the network interface connection, in example embodiments ;

[0115] Figures 4A to 4C show schematic diagrams showing examples of coupling of the monitoring device to the electric power system in example embodiments ; Figure 5 shows a process performed by the controller in example embodiments ;

[0116] Figures 6A and 6B show examples of frequency responses obtained on a first phase of the electric power system in an experimental example ;

[0117] Figures 7A and 7B show examples of frequency responses obtained on a second phase of the electric power system in an experimental example ;

[0118] Figures 8A to 8C show examples of frequency responses obtained on a third phase of the electric power system in an experimental example ;

[0119] Figures 8D and 8E show examples of frequency responses obtained on a third phase of the electric power system in an experimental example ;

[0120] Figures 9A to 9C show examples of frequency responses obtained on a three-phase electric power system in an experimental example ;

[0121] Figure 10 shows an example of a frequency response obtained on a phase of the electric power system in an experimental example ; and

[0122] Figures 11A and 11B show examples of frequency responses obtained on a phase of the electric power system in an experimental example .

[0123] Figure 12 shows a schematic diagram showing an example of a monitoring device for a multiphase electric power system in example embodiments .

[0124] Detailed Description

[0125] Although example embodiments will be described below, it will be evident that various modi fications may be made to these example embodiments without departing from the broader spirit and scope of the invention . Accordingly, the following description and the accompanying drawings are to be regarded as illustrative rather than restrictive .

[0126] Components described herein such as the monitoring device , the controller, the vector network analyser , the external communication device , and the remote device may use any suitable communication link to exchange data, such as wireless communication link ( for example a Wi-Fi , cellular telephone data link such as LTE / 5G, Bluetooth or Bluetooth Low Energy (BLE ) ) , a wired communication link ( e . g . DSL, fibre-optic cable , Ethernet etc . ) . Each communication link may not be permanent .

[0127] In the following description and in the accompanying figures , numerous details are set forth in order to provide an understanding of various example embodiments . However, it will be evident to those skilled in the art that embodiments may be practiced without these details .

[0128] For example , the figures and following description may omit certain connections between components , such as a connection to a reference point ( e . g . a ground) . The person skilled in the art would understand that such connection may be present , even i f not represented or described explicitly .

[0129] Figure 1A is a schematic diagram showing an example of a monitoring device in example embodiments .

[0130] The monitoring device 10 comprises a vector network analyser (VNA) 110 , a network interface connection 120 , a power supply unit 130 and a controller 140 .

[0131] By way of non-limiting example , the monitoring device 10 in Figure 1A is coupled to an external communication device , namely an antenna 150 for a 5G cellular telephone data link, although it would be understood that other communication device may be used instead, or the external communication device may be omitted .

[0132] It would be understood that although the elements of the monitoring device 10 are shown within a box, this is nonlimiting as the elements of the monitoring device 10 need not be in a single housing . Any of these elements , such as the network interface connection 120 , may be located outside a housing that comprises other elements of the monitoring device 10 . Some of the components of the monitoring device 10 , such as the vector network analyser 110 , the power supply unit 130 and / or a controller 140 , may be located at a distance ( e . g . a safety distance depending on the voltage of the electric line ) from the electric line and / or from the network interface connection 120 . For example , the network interface connection 120 may be located close to the electric line to be monitored and the VNA 110 may be connected to the network interface connection 120 by way of a cable for use in radio frequency ( e . g . a coaxial cable , an RG8 cable , an RG58 cable , etc . ) . By way of non-limiting example , for an overhead distribution line , the safety distance between the network interface connection 120 and the VNA 110 may be approximately 10 to 15 meters ( for example , by placing the network interface connection 120 near the overhead lines , and placing the VNA 110 near the ground) ; in another embodiment , the safety distance may be approximately 6 m .

[0133] In this connection, Figure IB shows an example of installation of the monitoring device 10 , where the latter is connected to an overhead distribution line 2 at a single point thereof from which it is able to perform, by the VNA 110 , a radial analysis , namely an analysis downstream and upstream .

[0134] More in detail , the VNA 110 is a device configured to perform a spectral frequency analysis on a given frequency range of an electric circuit it is coupled thereto . In the present disclosure , the VNA is used to make spectral frequency analyses of an electric power system ( described below) . The VNA 110 may be any known type of VNA with an internal signal generator or coupled to an external signal generator . For brevity, details of the VNA 110 that would be apparent to the person skilled in the art will not be described herein .

[0135] In the present example , the VNA 110 is configured to perform a spectral frequency analysis on a frequency range of 50 kHz to 6MHz . However, it would be understood that the frequency range covered by the VNA 110 is not limited to this example , and many other ranges of frequencies may be selected instead, which may depend on the electric power system to which the monitoring device will be coupled, the location where the monitoring device is to be coupled with the electric power system, etc .

[0136] The VNA 110 is to be coupled to the electric power system via the network interface connection 120 . To perform the spectral frequency analysis , the VNA 110 causes a s ignal to be generated and transmitted via the network interface connection 120 to the electric power system, and receives the response , measuring electrical characteristics for one or more frequency values . For example , the VNA 110 may obtain electrical characteristics ( such as the s-parameters ) for 100 frequency values between 50 kHz and 6MHz , at intervals on a logarithmic scale .

[0137] The network interface connection 120 comprises an electrical circuit connected to the VNA 110 and to the electric power system . The electrical circuit allows electrical signals at the frequency range covered by the VNA 110 ( e . g . 50 kHz to 6 MHz ) to pass through ( i . e . to flow between the electric power system and the VNA 110 with an amplitude substantially unaf fected by the electrical circuit ) , and filters out electrical signals at a network power frequency ( e . g . 50 Hz ) in the electric power system from reaching the VNA 110 ( i . e . the electrical circuit substantially attenuates the amplitude of these signals ) . For brevity, details of the network interface connection 120 that would be apparent to the person skilled in the art will not be described herein .

[0138] The power supply unit 130 is coupled to a power source (not shown) and is configured to supply electrical power from the power source to the VNA 110 and the controller 140 .

[0139] The power source may be a component of the monitoring device 10 or a power source that is external to, but coupled to the monitoring device 10 ( and in particular to the power supply unit 130 ) .

[0140] By way of non-limiting example , the power supply unit 130 may be coupled to a battery system and / or a photovoltaic panel system . The power supply unit may draw power from the battery system and / or the photovoltaic panel to supply power to the components of the monitoring device 10 . I f both the battery system and the photovoltaic panel system are provided, the power supply unit 130 may use a surplus of electrical power from the photovoltaic panel to control a recharge of the battery system . It would be evident that the power supply unit 130 may be coupled to any other type of power source ( or combination of power sources ) .

[0141] The controller 140 controls components of the monitoring device 10 . In particular, the controller 140 causes the VNA 110 to perform a spectral frequency analysis of the electric power system, to obtain frequency responses from the electric power system .

[0142] In the present example , the controller 140 generates a signal transmitted to the VNA 110 that triggers the VNA 110 to perform the spectral frequency analysis and retrieves a signal from the VNA 110 comprising information related to the frequency response .

[0143] The controller 140 then processes the information related to the frequency response to detect a change in state of the electric power system, as will be explained below .

[0144] Figure 2 shows a schematic diagram illustrating an exemplary implementation of the controller 140 .

[0145] In an embodiment , the processing device 40 includes at least processor 42 , a memory 43 and a I / O interface 41 . The processor 42 is configured to execute instructions comprised in a computer program stored on the memory 43 to execute any of the functions of the controller 140 , as described herein . The memory 43 is configured to store the computer program, and may also store additional information, such as information related to frequency response obtained by the VNA 110 , a machine learning model and data related to the machine learning model ( e . g . training data such as information related to the first frequency response ( s ) obtained by the VNA 110 ) . When executing the instructions of the computer program, the processor 42 receives information ( e . g . information related to the first frequency response , information related to each of the at least one subsequent response, etc.) from other elements of the monitoring device 10 (e.g. the VNA 110) and / or to the external communication device (e.g. antenna 150) by means the I / O interface 41, and output information to the other elements of the monitoring device 10 and / or the external communication device by means of the I / O interface 41.

[0146] Figure 3A is a schematic diagram showing an exemplary implementation of the network interface connection 120 of the monitoring device 10.

[0147] In the present example, the network interface connection 120 comprises a T-filter to electrically couple the VNA 110 to the electric power system and filter out the network power frequency from signals received by the VNA 110.

[0148] Specifically, the T-filter comprises four electrodes 121, 122, 123 and 124. The first electrode 121 and the third electrode 123 are coupled to the VNA 110, and the second electrode 122 and the fourth electrode 124 are to be coupled to the electric power system.

[0149] A first capacitor 125 (e.g. having a capacitance of 240pF) and a second capacitor 126 (e.g. having a capacitance of 75pF) coupled in series, between the first electrode 121 and the second electrode 122. An inductor 127 (e.g. having an inductance of 2.2mH) is coupled on one side to the common terminal of the first capacitor 125 and the second capacitor 126, and on the other side to the third electrode 123 and the fourth electrode 124. Figure 3A also shows the parasitic resistance 128 of the inductor 128.

[0150] By way of non-limiting example, the first capacitor 125 and the second capacitor may be, for example, formed by dipping plates in a resin-based material to form the dielectric medium. However, it would be apparent to the person skilled in the art that other materials (or combination of materials) may be used as the dielectric medium. Additionally, it would be apparent to the person skilled in the art that the values of the capacitance and inductance of the elements may vary, for example based on the range of frequency on which the VNA 110 is to perform the spectral frequency analysis and / or the network power frequency to be filtered out .

[0151] An insulation voltage level of at least components of the network interface connection 120 facing the electric power system, such as the second electrode 122 , the fourth electrode 124 , the second capacitor 126 and the inductor 127 , may be at least about I kV i f the monitoring device is to be coupled to a low-voltage electric power system, or e . g . , at least about 16kV or 24 kV respectively for Latin America and European Union i f the monitoring device 10 is to be coupled to a mediumvoltage electric power system .

[0152] In the example shown on Figure 3A, the common terminal of the third electrode 123 , the fourth electrode 124 and the inductor 127 is grounded . However, this is non-limiting as it would be understood that other configurations to couple the VNA to the phases of the electric power system are possible , such as by creating a phase-to-phase connection ( i . e . the second electrode 122 and the fourth electrode 124 are each coupled to di f ferent phases of the electric power system) .

[0153] Figure 3B shows the trans fer function of amplitude from the second electrode 122 to the first electrode 121 , for frequency values ranging between 20 Hz and 200kHz .

[0154] As illustrated on Figure 3B, the signal is signi ficantly filtered out ( i . e . the amplitude of the signal is attenuated) as it approaches 20 Hz . Figure 3B shows that , at around 700Hz , the signal attenuation is around 180 dB . Accordingly, any network power frequency below 700 Hz would be substantially filtered out from signals reaching the VNA 110 . By way of example , i f the electrical power carried by the electric power system has a voltage of 20kVAC ( in case of a medium voltage electric power system) and a frequency below 700 Hz , the signal received by the VNA 110 would be insigni ficant as it would have an amplitude less than 2 * 10A- 14 V .

[0155] However, it would be understood by the person skilled in the art that the network interface connection 120 need to achieve this level of filtering or attenuation, and that any filtering allowing the VNA 110 to perform the spectral frequency analysis without being substantially disturbed by the electrical power carried at the network power frequency would be acceptable . The required filtering may be determined for example based on one or more of the frequency range ( s ) on which the VNA 110 is to perform the spectral frequency analysis , the network power frequency, the amplitude of the voltage of the electrical power carried by the electric power system and the amplitude of the signals generated ( e . g . by the VNA 110 or by a signal generator ) for the spectral frequency analysis .

[0156] Figures 4A to 4C are schematic diagrams showing examples of coupling of the monitoring device to the electric power system in example embodiments .

[0157] In a first example shown on Figure 4A, the monitoring device 10 is coupled to the electric power system 20 , and speci fically to two phases of a three-phase overhead line .

[0158] In a second example shown on Figure 4B, the monitoring device 10 is coupled to a single phase of the three-phase overhead line .

[0159] In a third example shown on Figure 4C, the monitoring device 10 is coupled to each of the three-phase overhead line .

[0160] By way of non-limiting example , the electric power system 20 may be a medium voltage energy distribution electric power system, or a low voltage energy distribution electric power system . In examples where the monitoring device 10 is coupled to more than one phase of the electric power system 20 , the network interface connection 120 may comprise a switch to selectively connect the VNA 110 with one of the phases , for example by cycling through each connected phase . Thus , the VNA 110 may perform the spectral frequency analysis on the one of the phases to which it is connected to via the network interface connection 120 . The controller 140 may generate a signal to control the state of the switch in the network interface connection 120 . Alternatively, the monitoring device 10 may comprise a plurality of network interface connection 120 , each connected to a respective phase of the electric power system 20 to be monitored, and a switch selectively connecting one of the network interface connection 120 to the VNA 110 . In this case as well , the controller 140 may generate a signal to control the state of the switch to determine which network interface connection 120 ( and thus which phase of the electric power system 20 ) is connected to the VNA 110 .

[0161] Thus , a first frequency response may be stored in the memory of the monitoring device for each phase of the electric line to be monitored .

[0162] Alternatively, the VNA 110 may have a connection point for each phase of the electric power system 20 and perform the spectral frequency analysis in the phases in parallel , or in successive order . By way of non-limiting example , the VNA may be multiple-channel VNA. Each channel of the VNA being associated to one of the monitored phases of the electric power system 20 via a network interface connection, to perform a spectral frequency analysis on that monitored phase and obtain a frequency response for that monitored phase . In this case , switch for connecting the VNA to the network interface connection may not be required .

[0163] However, it would be apparent to the person skilled in the art that these examples of coupling to the electric power system are purely illustrative , and other coupling may be used instead . Similarly, the monitoring device 10 need not be coupled to a three-phase electric line , as the electric power system may have any other number of phases , ( e . g . a single phase for a DC voltage electric power system) , and the monitoring device 10 may be coupled to any other suitable element carrying electrical power of the electric power system 20 .

[0164] Figure 5 shows a process performed by the controller in example embodiments . In the following, all the following steps of the process are referred to a single phase , however it is understood that these steps may be performed for each phase of the electric power system to be monitored, e . g . by using a switch cycling through the di f ferent phases or controlling a multiple-channel VNA to obtain a frequency response from di f ferent phases of the electric power system .

[0165] At step S502 , the controller 140 causes the VNA 110 to obtain a first frequency response . The controller 140 then obtains from VNA 110 information related to the first frequency response .

[0166] For example , the controller 140 may transmit a signal instructing the VNA to perform a spectral frequency analysis on one or more first frequency ranges .

[0167] The VNA 110 may perform the spectral frequency analysis as instructed, and obtain a first frequency response . The controller 140 then obtains from VNA 110 information related to the first frequency response (hereinafter " first information" ) . The first information may comprise one or more frequency values and optionally one or more other electrical characteristics for each of these frequency values , such as s- parameters , a return loss , etc . The information related to the first frequency response may include solely a speci fic frequency value ( e . g . a resonant frequency at which the return loss is the lowest ) , or it may include a frequency response on full spectrum analysis .

[0168] At step S504 , which occurs at a time subsequent to step S502 , the controller 140 causes the VNA 110 to obtain a subsequent frequency response . The controller 140 then obtains from VNA 110 information related to the subsequent frequency response (hereinafter " second information" ) . As for step S504 , the controller 140 may transmit a signal instructing the VNA to perform a spectral frequency analysis , and obtain in return the information related to the subsequent frequency response comprising one or more frequency values ( e . g . a resonant frequency) and optionally one or more other electrical characteristics for each of these frequency values . At step S506, the controller 140 compares the first information received at step S502 and the second information received at step S504 to detect whether the comparison indicates a change in state of the electric power system 20.

[0169] If a change in state of the electric power system 20 is detected (YES at step S506) , processing proceeds to step S508.

[0170] At step S508, the controller 140 causes an alert notification to be generated.

[0171] For example, the controller 140 may cause a transmission of a signal indicating the detected change in state via the external communication device (e.g. antenna 150) , to a remote device. The signal may include at least one of information related to the detected change in state, information identifying the monitoring device (e.g. to individuate the monitoring device in case multiple monitoring devices may transmit information to the same remote device) , information related to the first frequency response, information indicating a time at which the first frequency response was obtained, information related to the subsequent frequency response, and information indicating a time at which the subsequent frequency response was obtained. Alternatively, the signal may include no payload, and the transmission of the signal itself may indicate that a change in state in the electric power system 20 was detected. In such cases, the monitoring device 10 may be identified by characteristics of the transmitted signal such as the specific frequency used to transmit the signal.

[0172] Then, after step S508, the processing ends.

[0173] If instead no change in state of the electric power system 20 is detected (NO at step S506) , processing proceeds to step S510.

[0174] At step S510, the controller 140 determines whether an event triggering the obtention of a first frequency response is detected.

[0175] For example, the event may be that the external communication device (e.g. antenna 150) receives a specific signal from a remote device, or it may be that a time since the first frequency response was last obtained has elapsed, that a specific time and / or day (e.g. time of the day, day of the week, of the month or of the year) has been reached, etc.

[0176] If an event triggering the obtention of a first frequency response is detected (YES at step S510) , the processing returns to step S502.

[0177] If no event is detected at step S510, the processing proceeds to step S512.

[0178] At step S512, the controller 140 determines whether an event triggering the obtention of a subsequent frequency response is detected.

[0179] For example, the controller 140 may be configured to obtain a subsequent frequency response recurringly, e.g. at a predetermined interval. Thus, the event detected at step S512 may be that a time since the subsequent frequency response was last obtained has elapsed.

[0180] If an event triggering the obtention of a subsequent frequency response is detected (YES at step S512) , the processing returns to step S504.

[0181] If no event is detected at step S512, the processing ends.

[0182] By way of non-limiting example, frequency responses obtained in various experimental setups will be described below, to illustrate aspects of the present invention. Although the following examples refer to the return loss being obtained as part of the frequency response, this is non-limiting. Any other suitable electric characteristics may be used instead or in addition to the return loss, such as any other s-parameter, or any y- , z-, and / or h-parameter ( s ) , a VSWR, an equivalent impedance, etc.

[0183] Figures 6A and 6B show examples of frequency responses obtained on a first phase of the electric power system in an experimental example, and specifically the return loss in dB between around 50kH and 10MHz.

[0184] In this experimental setup, a three-phase overhead line of 80 meters of length was used as a simulation of the electric power system. The pole was lowered for ease of access, with a first phase being at 0.8m from the ground, a second phase being at 1.66m from the ground and a third phase at 0.3m from the ground. The monitoring device was coupled to the electric power system via an RG58 co-axial cable. As no electrical power was carried by the simulated electric power system, the monitoring device did not include a network interface connection.

[0185] Figure 6A shows an example of a first frequency response obtained on the first phase when no fault is present. A notch in the return loss is apparent around the frequency of about 810kHz, and the minimum return loss is measured near 810 kHz to be around -20.501dB.

[0186] Accordingly, when the controller 140 obtains information related to the frequency response shown on Figure 6A, a threshold can be set for a frequency shift (e.g. 20kHz) around the frequency of 810kHz, and / or an amplitude shift (e.g. 3dB) around the minimum return loss of -20.501dB.

[0187] Figure 6B shows an example of a subsequent frequency response obtained on the first phase, with a simulated ground fault near the end of the overhead line opposite to the monitoring device. The notch in the return loss around the frequency of 810kHz is no longer visible. For the same frequency of 810kHz, the return loss is now around -0.572dB. Instead, the minimum return loss is measured around the frequency of about 4.5MHz .

[0188] Accordingly, a deviation of about 20dB (from -20.501dB to -0.572dB) in the amplitude of the return loss and / or a shift of 3.7MHz (from 810kHz to 4.5MHz) occurs, which may be detected by the controller 140 using the defined threshold (s) .

[0189] Figures 7A and 7B show examples of frequency responses obtained on the second phase of the electric power system in an experimental example, and specifically the return loss in dB between around 50kH and 6MHz, in the same experimental setup as Figures 6A and 6B .

[0190] Figure 7A shows an example of a first frequency response obtained on the second phase under the same conditions as those of Figure 6A, and Figure 7B shows an example of a subsequent frequency response obtained on the second phase under the same conditions as those of Figure 6B, that is with a ground fault on the second phase at the same distance from the monitoring device as in the case of figure 6B .

[0191] Figures 8A and 8B show examples of frequency responses obtained on the third phase of the electric power system in an experimental example , and speci fically the return loss in dB between around 50kH and 6MHz , in the same experimental setup as Figures 6A and 6B and Figures 7A and 7B .

[0192] Figure 8A shows an example of a first frequency response obtained on the third phase under the same conditions as those of Figure 6A, and Figure 8B shows an example of a subsequent frequency response obtained on the second phase under the same conditions as those of Figure 6B, that is with a ground fault on the third phase at the same distance from the monitoring device as in the case of figure 6B .

[0193] As with the other phases , a large deviation in the amplitude of the return loss at a frequency around 810kHz ( from around -23dB to - 0 . 8dB ) is apparent , and a shi ft in frequency of the minimum return loss ( from around 810kHz to 4 . 5MHz ) is apparent .

[0194] Accordingly, a comparison of the frequency responses shown on Figures 7A and 7B ( or a comparison of those on Figures 8A and 8B ) also shows a deviation in the amplitude of the return loss at a frequency around 810kHz from around -23dB to around - 0 . 5dB, and a shi ft in the frequency of the minimum return loss from 810kHz to 4 . 5MHz . Thus , the controller 140 would, when receiving information related to the frequency response shown on Figure 6A, 7A or 8A, and information related to the frequency response shown on Figure 6B, 7B, or 8B, be able to detect the change in state of the electric power system .

[0195] Thus , in the experimental setup of Figures 6A to 8B, a connection to a single phase of the electric power system is suf ficient to detect a change in state of the electric power system af fecting the phase connected to the monitoring device . Figure 8C shows an example of a subsequent frequency response obtained on the third phase , with a simulated ground fault near the middle of the overhead line .

[0196] In this case as well , the controller 140 receiving the information related to the frequency responses shown on Figures 8A and 8C would be able to detect a change in state of the electric power system, for example by detecting the increase of around 22dB ( from -23 . 4db to - 1 . 42dB ) in the return loss at 810kHz .

[0197] Figures 8D and 8E show frequency responses obtained with the same experimental setup as Figures 8A and 8B, but including the network interface connection shown on Figure 3A . As explained above , the network interface connection would attenuate signals at around 50Hz by more than 200dB, and would therefore be suitable to couple the monitoring device to a medium voltage electric power system .

[0198] For simplicity, the range of frequency shown on Figures 8D and 8E is narrower than the range on Figures 8A and 8B .

[0199] The frequency response shown on Figure 8D includes a minimum return loss of about -23 . 92 dB at the frequency of 806 . 217 kHZ . These results are simi lar to those shown on Figure 8A.

[0200] Then, when simulating the same fault as the one represented on Figure 8B, namely that a fault is simulated by connecting the ground to the overhead line near the end of the overhead line opposite to the monitoring device .

[0201] As shown on Figure 8E , the return loss at the frequency of 806 . 217 kHz increases to - 0 . 437dB . In addition, the Figure 8E shows dips in the return loss around the frequency of 1 . 287MHz and 2 . 6MHz , which are also apparent on Figure 8B .

[0202] These results therefore show that the monitoring device with the network interface connection can detect changes in state of the electric power system related to a monitored phase of the electric power system .

[0203] Figures 9A to 8C show examples of frequency responses obtained on a three-phase electric power system in an experimental example performed using real measurements in a field test , and speci fically the return loss in dB between around 50kH and 4MHz .

[0204] In this experimental setup, to represent the electric power system, a real , dead three-phase 20 kV overhead line connected to a distribution substation, where the monitoring device was connected to the overhead line .

[0205] Figure 9A shows a frequency response measured on the entire line , showing a resonant frequency around 203kHz where the measured return loss was about - 10 . 079dB .

[0206] Then, at a distance of about 100m from the monitoring device , the line was connected to the ground to simulate a fault , and the frequency response shown on Figure 9B was measured . As shown on Figure 9B, the return loss at 203 kHz was about - 1 . 297 dB . Accordingly, a change in state of the electric power system can be detected due to the nearly 10 dB deviation in the return loss amplitude .

[0207] Then, at a distance of about 500m from the monitoring device , the three phases of the overhead line were connected together to simulate a short-circuit and connected to the ground for safety, and the frequency response shown on Figure 9C was measured .

[0208] As shown on Figure 9C, the return loss at 203 kHz was about - 0 . 302 dB . Accordingly, a change in state of the electric power system can be detected even when the fault occurs at greater distance of the monitoring device , due to the nearly 10 dB deviation in the return loss amplitude . Furthermore , as the line was out-of-service , it is apparent that the monitoring device can detect the change in state even when the electric power system experiences a power outage , and that the monitoring device can detect the change in state due to a change in connections or a fault in either the upstream or downstream directions .

[0209] Figure 10 show an example of a frequency response obtained on a phase of the electric power system the same experimental setup as Figures 9A to 9C . Speci fically, the monitoring device was recalibrated when disconnected from the electric power system, and then reconnected to only one of the three phases .

[0210] Figure 10 shows the frequency response obtained on the entire line when not connected to ground . As shown on Figure 10 , the return loss is measured to have a minimum value of about - 9 . 897dB at the frequency of about 203 kHz , thus showing coherent results with those shown on Figure 9A. The presence of minor di f ferences between the frequency response on Figure 9A and Figure 10 are due to di f ferences in the connection between the monitoring device and the overhead line .

[0211] The experimental test shown on Figure 9A- 9C and 10 show that the resonant frequency can be used to detect a fault or other change in state of the electric power system, even i f distant from the monitoring device, and that the monitoring device can determine the frequency of minimum return loss and monitor that frequency to detect , in real-time , any noteworthy variation in the return loss .

[0212] Figures 11A and 11B show examples of frequency responses obtained on the third phase of the electric power system in the same experimental setup as that described for Figures 6A- 6B, 7A-7B and 8A- 8C above .

[0213] However, in the case of Figures 11A- 11B, the VNA was calibrated when already coupled ( or connected) to the third phase of the electric power system .

[0214] As shown on Figure 11A, this leads to a return loss that is below well below the -20dB for most of the frequency range of 50kHz to 3 . 7MHz .

[0215] Then, by simulating the same fault as the one described in reference to Figures 6B, 7B and 8B above , the frequency response shown on Figure 11B was obtained .

[0216] In this case , the return loss for substantially the entire frequency range was shi fted up to be around OdB .

[0217] A di f ference of up to 40dB in return loss was determined between the frequency response shown on Figure 11A and the frequency response shown on Figure 11B . From the experimental test of Figures 11A- 11B, it is apparent that the monitoring device need not focus on a speci fic frequency where the return loss is the minimum under normal conditions .

[0218] Additionally, calibrating the monitoring device ( and speci fically the VNA) when it is already connected to the electric power system ( e . g . as shown on Figures 11A- 11B ) improves the detection of a change in a state of the electric power system relative to when the VNA is calibrated before being coupled to the electric power system ( e . g . as was the case in the experimental example shown on Figures 6A- 6B ) . Namely, the change in state may cause a di f ference in amplitude of the return loss up to 40dB, from the 20dB di f ference in amplitude apparent from Figures 6A and 6B .

[0219] Figure 12 is a schematic diagram showing an example of a monitoring device in example embodiments . In particular, the monitoring device 10 ' is for a multiphase electric power system .

[0220] The monitoring device 10 ' comprises a VNA 110 , four network interface connections 120a - 120d, a power supply 130 , a controller 140 , and a switch 160 .

[0221] The VNA 110 , the power supply 130 , and the controller 140 are the same as those described with reference to Figure 1A. Their description will therefore be omitted for brevity .

[0222] Each network interface connection 120a, 120b, 120c and 120d is to couple a phase of the electric power system to the VNA 110 , and each is the as the network interface connection 120 as described above with reference to Figure 3A.

[0223] The switch 160 selectively connects one of the network interface connections 120a to 120d to the VNA 110 . The controller 140 generates a control signal that controls the state of the switch 160 , thus controlling which of the network interface connections is connected to the VNA 110 .Accordingly, by controlling the state of the switch, the controller 140 may cause the VNA 110 to be coupled to each of the phases of the electric power system and to obtain a frequency response from each of the phases. The controller 140 can then cause each of these frequency responses to be stored in memory (in particular if it is a first frequency response for that phase) , or used to detect a change in state of the electric power system affecting that phase (for example if it is a subsequent frequency response) .

[0224] REMARKS REGARDING INNOVATIVE ASPECTS OF THE INVENTION WITH RESPECT TO THE PRIOR ART

[0225] The monitoring device according to the present invention is a single equipment that can be positioned at any location on a medium voltage (MV) electric power system and / or a low voltage (LV) electric power system, in particular:

[0226] • in the first branch of the MV line at the exit of the primary substation;

[0227] • on the backbone, at one or more sectioning points;

[0228] • on the derivations .

[0229] From the signal injection point, coinciding with the positioning of the monitoring device, the portion of the underlying electrical network is analyzed in radial, i.e. non- directional, mode, differently from what happens with the use of current protection relays.

[0230] The type of analysis carried out concerns the comparison of different quantities (e.g., frequency response images) acquired by the VNA at time To and in the subsequent timed polling instants (To+AT) .

[0231] The first image at time To becomes the reference for the self-calibration of the monitoring device, regardless of the consistency and type of the underlying electrical network.

[0232] The variations detected by the analysis algorithm generate a series of status words to communicate critical events directly to the SCADA, such as typically:

[0233] • one or more conductors earthed ( single-phase / multi-phase earth fault) ;

[0234] • short-circuited conductors; • accidental temporary contacts (e.g. shrubs carried by the wind onto the conductors) .

[0235] Additionally, it is worth stressing the importance of the following three key aspects of the monitoring device according to the present invention:

[0236] 1) the monitoring device works even in the absence of electrical energy in the electric power system;

[0237] 2) the monitoring device is coupled with the electric power system at a single point, whereby it has a plug-and-play-like installation; thence, with multiple devices it is possible to monitor multiple critical points of one and the same power line (thereby avoiding the use of vehicles and people) with consequent minimization of risks and costs;

[0238] 3) the monitoring device is a potential "enabler"; in fact, frequency domain data analysis of distribution networks could be integrated both with SCADA and third part systems for remote control and network automation; acquiring frequency response data from the electric grid, historicizing and analyzing it, opens up new analysis scenarios (e.g., predictive maintenance with Al support) .

[0239] The following table lists main differences and innovative aspects of the monitoring device according to the present invention with respect to the solution according to US 2012 / 271574 Al.

[0240]

[0241]

[0242]

[0243] MODIFICATIONS AND VARIATIONS

[0244] Although the electric power systems are described to have alternative current (AC ) electrical energy, it would be understood that the electric power system ( e . g . medium voltage electric power system MV and / or the low voltage electric power system LV) may operate with direct current ( DC ) electrical energy instead . The electric substation 2 may comprise , in some cases , means for converting AC electrical energy to DC electrical energy, or vice-versa .

[0245] Although in the above example , the network interface connection 120 comprises a T- filter, this is not limiting as other electric circuits may be used to filter out the network power frequency . For example , a band-pass or high-pass filter allowing frequencies higher than the network power frequency may be used . Alternatively or additionally, a bandstop filter may be used to filter out the network power frequency .

[0246] Similarly, a combination of filters may be used in the network interface connection 120 .

[0247] Although the above example embodiments describe the monitoring device 10 being coupled to an overhead line , this is not limiting as the monitoring device may be coupled to any other element of the electric power system carrying electrical power .

[0248] In the above example embodiments , the first frequency response may also be referred to as a reference frequency response ( as it is used for comparison with the subsequent frequency response ( s ) to detect a change in state ) , and the subsequent frequency response ( s ) may be referred to as a measurement frequency response .

[0249] Although the above example embodiments describe the processing operations in a speci fic order, this is not limiting and changes can be made to the order .

[0250] For example , steps S510 and S5120 may be performed in reverse order .

[0251] As another example , step S508 may be performed contemporaneously to at least one of step S510 , step S502 ( i f the obtention of a new first frequency response is triggered) , step S512 and step S504 ( i f the obtention of a new subsequent frequency response is triggered) . In other words , the generation of the alert noti fication need not interrupt the obtention of the first frequency response and / or the subsequent frequency response ( s ) .

[0252] Although the above example embodiments describe the processing operations to be comprising a detection of triggers for repeating the obtention of the first frequency response and the subsequent frequency response , this is not limiting as one or both of the steps S510 and S512 may be omitted .

[0253] For example , the controller 140 may be configured to cause the VNA 110 to obtain a first frequency response ( step S502 ) when the monitoring device 10 is powered for the first time . Additionally or alternatively, the controller 140 may be configured to cause the VNA 110 to obtain a subsequent frequency response ( S504 ) when a predetermined event is detected ( e . g . the loss of electrical power in the electric power system) , or when a predetermined time has elapse since the first frequency response was obtained .

[0254] In other examples , step S508 may be omitted .

[0255] For example , instead of generating an alert noti fication, the controller 140 may be configured to implement a network protection function to disconnect or otherwise protect elements of the electric power system from ef fects that may arise from the detected change in state . It would be apparent to the person skilled in the art that the controller 140 may perform both the alert noti fication and other functions ( such as protection relay or loT integration) when a change in state of the electric power system is detected .

[0256] Although the above example of a monitoring device for a multiphase electric power system comprises four network interface connections it would be understood that the number of network interface connections is not limited thereto , and may instead comprise two , three , five or more network interface connections .

[0257] Although the above example of a monitoring device for a multiphase electric power system comprises a switch for selectively coupling the network interface connections to the VNA, this is non- limiting . Alternatively or additionally, the VNA may comprise a plurality of connection points , each to be coupled to a di f ferent network interface connection . For example , the VNA may comprises four channels , each coupled to a di f ferent network interface connection .

[0258] In some examples , the VNA with a plurality of connection points and the switch may be used in combinations . For example , the VNA may comprise two channel , and the switch may be configured to connect the two channels to two of the four network interface connections . As will be apparent to the person skilled in the art , any of the above-described example embodiments describing processing operations may be embodied by a computer program comprising computer-readable instructions , which, when executed by a computer or by one or more processor, causes the computer or the one or more processor to implement the method defined by processing operations . Said computer program may be stored in a transitory or non-transitory computer-readable storage medium or carried by a signal .

Claims

CLAIMS1. A monitoring device (10) for an electric power system, comprising a vector network analyser (110) , a network interface connection (120) and a controller (140) , wherein:• the network interface connection (120) is configured to- electrically couple the vector network analyser (110) with the electric power system at a single location thereof and- filter out a network power frequency of the electric power system from signals received by the vector network analyser (110) ;• the vector network analyser (110) is configured to perform a spectral frequency analysis of the electric power system on at least one frequency range that is higher than the network power frequency, to obtain a frequency response of the electric power system;• the controller (140) is configured to, while the network interface connection (120) electrically couples the vector network analyser (110) with the electric power system,- cause the vector network analyser (110) to obtain a first frequency response,- cause the vector network analyser (110) to obtain at least one subsequent frequency response, and- detect a change in state of the electric power system based on a comparison of the at least one subsequent frequency response with the first frequency response; whereby the monitoring device (10) results to be a single device that monitors and analyses the electric power system, radially and in an asynchronous polling mode, from the single location where the vector network analyser (110) is electrically coupled with said electric power system, to detect faults and state variations of said electric power system.

2. The monitoring device of claim 1, wherein the change in state indicates a fault or a modification to connections to the electric power system.

3. The monitoring device according to claim 1 or 2, wherein the network interface connection (120) comprises a band-pass or high-pass filter, preferably a T-filter, to filter out the network power frequency.

4. The monitoring device according to any of claims 1 to3, wherein the controller (140) is configured to:• detect a change in state of the electric power system based on a deviation in amplitude and / or a frequency shift between the first frequency response and the at least one subsequent frequency response.

5. The monitoring device according to any of claims 1 to4, wherein the controller (140) is configured to:• determine, from the first frequency response, a first value of at least one electrical characteristic.

6. The monitoring device of claim 5, wherein the controller (140) is configured to:• determine, from the at least one subsequent frequency response, a second value of the at least one electrical characteristic, and• detect the change in state of the electric power system based on a comparison of each first value and the corresponding second value.

7. The monitoring device of claim 6, wherein the at least one electrical characteristic comprises a frequency at which the frequency response has a minimum return loss; and wherein the comparison of each first value and the corresponding second value comprises a comparison of a first frequency at which the first frequency response has a minimum return loss and a secondfrequency at which the at least one subsequent frequency response has a minimum return loss .8 . The monitoring device according to any of claims 5 to 7 , wherein the controller ( 140 ) is configured to cause the vector network analyser ( 110 ) to obtain the at least one subsequent frequency response on one or more second frequency ranges , the one or more second frequency ranges being determined based on the first value of at least one electrical characteristic .

9. The monitoring device according to any of claims 1 to 8 , wherein the controller ( 140 ) is configured to , upon detecting a predetermined trigger, cause the vector network analyser ( 110 ) to obtain the first frequency response again, and to detect the change in state of the electric power system using the new first frequency response .10 . The monitoring device according to any of claims 1 to 9 , wherein the controller ( 140 ) is configured to , upon detecting the change in state of the electric power system, cause the generation of an alert noti fication, or to directly command a protection relay .11 . The monitoring device according to any of claims 1 to 10 , wherein the electric power system is a low voltage energy distribution electric power system or a medium voltage energy distribution electric power system, wherein the network interface connection ( 120 ) is configured to electrically couple the vector network analyser ( 110 ) with either the low voltage energy distribution electric power system or the medium voltage energy distribution electric power system .

Images