Patent Application
20250067786
Reflectometry can be used to detect faults in electrical networks. Some reflectometry methods involve injecting a signal into the electrical network, detecting a reflected signal, and identifying an impedance variation in the electrical network by correlating the injected signal with the reflected signal.
It is desirable to monitor electrical systems such as, but not limited to, overhead line equipment of railways, trolleybuses, trams and the like. Overhead line equipment refers to overhead wires and supporting equipment that vehicles such as trains electrically contact in order to receive electric power. Normally, such vehicles utilise an extending pantograph to contact an overhead wire of overhead line equipment. Components of overhead line equipment is prone to damage, vandalism and/or theft. It is therefore desirable to determine if components of the overhead line equipment are damaged, vandalised or stolen. It is also desirable to monitor other types of electrical systems such as electrical transmission cables, or any other type of electrical system for transmitting electricity over a long distance.
According to a first aspect, there is a provided a method for monitoring an electrical system having multiple electrical ground connection points. The method comprises receiving reflectometry data that has been generated by performing time domain reflectometry on the electrical system, the time domain reflectometry being performed during a measurement time period; determining a measured waveform from the reflectometry data, the measured waveform indicative of a measured state of the system during the measurement time period; obtaining a predetermined baseline waveform indicative of a baseline state of the system during a calibration time period; generating a difference waveform based on the measured waveform and the baseline waveform; and analysing the difference waveform to determine if there is a change of state in the electrical system between the calibration period and measurement period.
The method provides for a precise determination of a change of state of the multiple grounded electrical system. For example, it may be recognised that certain difference waveforms are indicative of particular types of change of state, and optionally, a location of the change of state. As used herein, the term “change of state” may refer to an unauthorised modification/removal of any conductive element of the electrical system, or, an electrical fault, such as a short-circuit. The method enables an operator to determine the nature of the change of state and take appropriate remedial action in a timely fashion. For example, based on the difference waveform, the operator may determine that a particular component needs replacing. The particular component can be arranged for delivery to the fault site even before the fault has been inspected in-situ. The operator may also be able to quickly determine that a conductive element of the electrical system has been stolen, or even that it is in the process of being disconnected in an unauthorised manner, thereby enabling timely remedial action. The method reduces the risk of catastrophic failure of the electrical system, improves fault detection and timely repair, and ensures safe operation.
The term “electrical system” as used herein may refer to an electrical circuit comprising at least one electrical conductor that has multiple ground connection points. The electrical system need not have an electricity supply. Such an electrical system may be monitored in accordance with the techniques disclosed herein to monitor the quality of the earthing of the at least one electrical conductor. For example, the electrical system may comprise a permanently earthed section of a railway line.
Optionally, the time domain reflectometry is performed by transmitting and receiving signals along a grounded conductor having multiple electrical ground connection points.
Optionally, the change of state comprises a change of state of a ground connection of the electrical system.
Optionally, the change of state comprises a change to a conductive structure supporting a conductor of the electrical system.
Where the electrical system is an overhead line comprising a conductive wire, the conductive structure may be a mast supporting the conductive wire. The mast may be connected to ground.
It is desirable to monitor the state of components providing the ground connections. Such components are normally located close to the ground (where earth is acting as ground), so are susceptible to theft. This is particularly the case for permanently earthed sections, since electrical current does not normally flow through the section. Theft of these components renders the electrical system more dangerous due to the reduction of ground connection paths that are available during an electrical system fault such as a short-circuit. Theft of such components have previously been difficult to detect, since their absence may only be noticed in the event of a system fault. Any disruption to any ground connection can be detected based on an analysis of the difference waveform and an alarm can be initiated to immediately notify operators. Furthermore, there is provided the ability to detect permanent and intermittent faults (e.g. arc faults) on ground connections.
Optionally, the reflectometry is one of time domain reflectometry (TDR), Sequence Time Domain Reflectometry (STDR), and Spread Spectrum Time Domain Reflectometry (SSTDR).
Whilst the scope of this disclosure encompasses various types of reflectometry, SSTDR reflectometry is particularly favourable since SSTDR signals can be isolated from activity and noise generated by the monitored electrical system during operation. Therefore, an electrical system can be monitored even during routine operation.
Optionally, analysing the difference waveform comprises determining if there is a change of state based on an amplitude of the difference waveform.
Optionally, analysing the difference waveform comprises determining if there is a change of state based on a location of a peak of the difference waveform.
Optionally, analysing the difference waveform comprises inputting the difference waveform into a trained classifier for determining if there is a change of state in the electrical system.
Use of a trained classifier provides for an improved and efficient determination of the type and/or location of a change of state of the electrical system based on the difference waveform. The trained classifier can be continuously trained to recognise difference waveforms as relating to different types of change of state. The trained classifier can also be trained to recognise certain difference waveforms as not relating to a problematic change of state, for example, where the electrical system is temporarily affected by adverse weather or the movement of any vehicles that are connected to the electrical system. The trained classifier may be an artificial neural network.
Optionally, analysing the difference waveform comprises comparing the difference waveform with two or more stored difference waveforms, wherein each of the two or more stored difference waveforms corresponds to a known change of state.
The comparison may comprise undertaking a computational similarity analysis between the difference waveform and the stored difference waveforms. The stored waveforms difference waveforms may have been generated and stored during an earlier period of time (e.g. the calibration period), such as during controlled simulations of different changes of state. The analysis may involve a comparison of portions of the difference and stored waveforms. Such analysis may involve use of a Fourier transform function to obtain a representation of the waves in the frequency domain. It is possible to store a large number of waveforms corresponding to a variety of known changes of state. The comparison provides for a determination of a change of state of the system without requiring visual inspection of the electrical system itself.
Optionally, the reflectometry data is received by a local monitoring system, and is transmitted to a server which is remote to the local monitoring system, and wherein the difference waveform is analysed by the server.
The step of analysing may require a significant amount of processing power. Furthermore, it may be desirable to utilise a single processing unit (e.g. of the server) for undertaking the analysis of difference waveforms for multiple different electrical systems. This may be efficient where, for example, the analysis utilises a database of stored waveforms, or where the analysis utilises an artificial neural network.
Optionally, analysing the difference waveform comprises determining one or both of a location within the electrical system, and a type of any change of state based on the difference waveform.
It is advantageous that not only the presence of a fault/change of state is detected, but rather that a type and/or location of the fault can be ascertained. An operator can therefore take appropriate responsive action in a quick manner.
Optionally, the method further comprises generating an alarm when a change of state is determined, and generating an indication of the one or both of the location and type of change of state.
Optionally, the method further comprises obtaining a plurality of sets of reflectometry data of the electrical system, wherein each set of reflectometry data is obtained under a different simulated change of state of the system, and wherein each of the plurality of sets of reflectometry data comprise a simulated difference waveform corresponding to the respective simulated change of state of the system, and wherein each simulated difference waveform is stored as training data.
Obtaining and storing reflectometry data under simulated states of change enables the monitoring method to be tailored for particular electrical systems that are to be monitored, thereby improving the accuracy of the determination of changes of state. Obtaining and storing reflectometry data may be undertaken during the calibration time period.
Optionally, the method further comprises training a classifier by inputting the plurality of sets of reflectometry data and corresponding simulated change of state of the system into the classifier.
According to a further aspect there is provided an apparatus for monitoring an electrical system, the apparatus comprising a processing unit configured to undertake the method described herein.
Optionally, the apparatus further comprises a reflectometry generator configured to be connected to a conductor of the electrical system, the reflectometry generator configured to obtain test reflectometry data of the electrical system.
The reflectometry generator is one of a TDR, STDR, and SSTDR reflectometry generator. Where the electrical system is an overhead line comprising a conductive wire supported by a plurality of conductive masts, the conductive masts providing a ground connection point, then the reflectometry generator may be connected between one of the conductive masts and the conductive wire.
Optionally, the processing unit is located on a server remote to the electrical system.
According to a further aspect there is provided a computer-readable storage medium comprising instructions for execution by a processor, wherein the instructions are configured to cause the processor to perform the method described herein.
The skilled person will appreciate that except where mutually exclusive, a feature described in relation to any one of the aspects, examples or embodiments described herein may be applied to any other aspect, example, embodiment or feature. Further, the description of any aspect, example or feature may form part of or the entirety of an embodiment of the invention as defined by the claims. Any of the examples described herein may be an example which embodies the invention defined by the claims and thus an embodiment of the invention.
The invention will now be described, by way of example only, with reference to the accompanying drawings, in which:
Disclosed herein are systems and methods suitable for monitoring electrical systems. The disclosed systems and methods are particularly suitable for detecting and locating changes to a permanently earthed sections (PES), which are often part of the Overhead Line Equipment (OLE) utilised by railways, tramlines, trolleybuses, and the like. Permanently earthed sections are sections of an OLE which are not live and may be utilised for sections of the line where there is human contact due to maintenance. Alternatively, permanently earthed sections may be utilised where it is not possible to place equipment that is required for live sections due to environmental factors or space considerations, such as under low bridges. PES sections enable a pantograph of a train to maintain contact with a section of overhead wire which is not electrically powered. PES sections typically cross several spans of OLE, each span being separated by a supporting mast. It is desirable to monitor changes to PES as a result of cable cuts, or components such as jumpers and droppers being removed due to theft. The disclosed systems and methods provide for an operator to be notified when there is such a change to the electrical system, thereby helping to guard against the theft of a component and/or providing an immediate alarm when safety-critical damage has occurred. The methods and systems of the present disclosure are particularly advantageous in that an operator is sometimes able to determine precisely what type of change has occurred, and the location of the change, thereby enabling prompt remedial action. Furthermore, if the electrical system is a permanently earthed system, then the operator is able to detect any faults or changes of state of any ground connections.
With reference to
With reference to
Where the reflectometry generator 207 is a Spread Spectrum Time Domain Reflectometry (SSTDR) generator, the transmitted and received signals may be correlated in the following manner. First, a digital message of the transmitted signal may be encoded with a bit sequence (e.g. a pseudorandom noise code) to obtain an encoded message. The process of encoding may include performing a bitwise exclusive-OR of the bit sequence with the digital message. Subsequently, a digital to analogue converter (not shown) may convert the encoded digital message to an analogue signal. The analogue signal may be modulated (e.g. with a sine wave having a modulation frequency) to produce a modulated signal. The modulated analogue signal may be transmitted, by the SSTDR generator, into the electrical system 209. The SSTDR generator may include a receiver for receiving the reflection of the transmitted signal. The received signal may be extracted by demodulating the reflected signal using the sine wave having the modulation frequency, converting the demodulated signal to an analogue signal using an analogue to digital converter (not shown) and, then decoding the reflected signal (e.g. by performing a bitwise exclusive-OR of the demodulated received signal with the bit sequence). This results in a digitised version of the received signal, which can be processed further. The digitised received signal may subsequently be correlated (e.g. cross-correlated) with the transmitted signal to obtain the measured waveform signal. The cross-correlation may be undertaken by a correlation circuit implemented on a field-programmable gate array (FPGA) (not shown).
The analysis of the measured waveform signal may comprise determining a change of state based on an amplitude of the difference waveform and/or a location of a peak of the difference waveform.
The control unit 203 may control operation of the reflectometry generator 203 and/or provide a user interface for notifying an operator when/where a change of state in the electrical system has been detected.
The difference waveform may be input into a trained classifier, such as an artificial neural network, for determining if there is a change of state in the electrical system. Other suitable machine learning techniques may be utilised to analyse the difference waveform. The trained classifier may be trained to recognise the specific type of change of state in the electrical system that is indicated by a particular difference waveform. The training data for the neural network may include a database of difference waveforms and corresponding changes to an electrical system that each difference waveform indicates. The trained classifier may be specific to a particular design of electrical system, or be common to multiple designs of electrical system. Particular difference waveforms may be recognised via the trained classifier to indicate particular changes of state for different types of electrical systems which are being monitored. The trained classifier may be continuously trained based on changes of state and corresponding difference waveform that are detected from multiple monitored electrical systems.
In some examples, the processing unit 202 and storage device 205 are located remotely to the PES 209 and reflectometry generator 207. For example, the processing unit 202 and storage device may be embodied in one or more servers, or a cloud computing environment. Multiple different reflectometry generators monitoring different electrical systems 209 may be connected to the same processing unit 202. In these instances, a connection 211 to the reflectometry generator 207 may be a network such as the internet. A user may control the processing unit and be notified of a change of state in the electrical system 209 via the control unit 203, which may embodied as a computer terminal accessing the processing unit 202 via the internet or other communications network. The control unit 203 may comprise software including a web browser for enabling a user to control the analysis module 204 or other aspects of the monitoring apparatus 201 via the internet.
With reference to
The systems and methods discussed herein utilise a baseline waveform of an electrical system. The baseline waveform depends on the state of an electrical system when there is no abnormal fault or change of state. For permanently earthed systems, the baseline waveform takes account of the earthing arrangement of the electrical system. The baseline waveform is typically obtained using reflectometry methods at a time when the electrical system is in a baseline condition—i.e. no faults or abnormal changes. The difference waveform is obtained based on a difference between the baseline waveform and a newly obtained measured waveform, and analysis of the difference waveform provides an indication that there has been a change in the earthing system arrangement. For example, with continued reference to
The following examples relate to testing that has been undertaken to demonstrate the functionality of the discussed systems and methods for monitoring electrical systems. The testing has been undertaken on a test rig simulating a permanently earthed section (PES) of an overhead line equipment (OLE) for a train line. SSTDR reflectometry has been used for monitoring in these tests. The tests demonstrated that the discussed methods and systems enable observation of not only changes in a main contact wire of an OLE, but also, changes to structures providing earth bonding in a PES such as dropper links across insulators in series with a ground connection, bondings between a mast and a cantilever of an OLE, and masts or other rigid structural elements of the OLE.
With reference to
The test rig comprises three masts 602a, 602b, 602c that are electrically grounded. An electrical contact wire 603 is connected between masts 602a, 602c, and (via a substantially horizontal cantilever 611) mast 602c. The electric contact wire 603 is therefore grounded via the masts 602a, 602b, 602c. The electrical contact wire 603 is additionally suspended from droppers 612, 613 extending from a messenger wire 614. An isolator 605 impedes current along the messenger wire 614, and is bypassed by a jumper 605. Another isolator 607 impedes current along a cantilever 615. Another jumper 608 bypasses the isolator 607. A return path 610 is connected between masts 602a and 602b. The structural features which are supported by the masts 602a, 602b, 602c are at the same electrical potential. The components supported by the masts 602a, 602b, 602c, and the ground connections enabled by the masts themselves, make up an electrical system which is being monitored. An SSTDR reflectometry generator 601 is connected in between mast 602a and the electrical contact wire 603. During each test, the SSTDR reflectometry generator transmitted a signal across the contact cable 603 and the mast 602a, and received a corresponding reflection. The transmitted and reflected signals were correlated to obtain the waveforms discussed below. For each test, the same frequency was used to transmit the SSTDR signals for obtaining both baseline and measured waveforms, although different frequencies were utilised for different tests.
With continued reference to
It can be observed from
With reference to
It will be understood that the invention is not limited to the examples and embodiments above-described and various modifications and improvements can be made without departing from the concepts described herein. Except where mutually exclusive, any of the features may be employed separately or in combination with any other features and the disclosure extends to and includes all combinations and sub-combinations of one or more features described herein.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2110851.9 | Jul 2021 | GB | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/GB2022/051879 | 7/20/2022 | WO |
