WO2014006662A1 - Method for diagnosing an electric device - Google Patents

Method for diagnosing an electric device Download PDF

Info

Publication number
WO2014006662A1
WO2014006662A1 PCT/JP2012/004398 JP2012004398W WO2014006662A1 WO 2014006662 A1 WO2014006662 A1 WO 2014006662A1 JP 2012004398 W JP2012004398 W JP 2012004398W WO 2014006662 A1 WO2014006662 A1 WO 2014006662A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
set
electric device
method according
electric
Prior art date
Application number
PCT/JP2012/004398
Other languages
French (fr)
Inventor
Francesco Baldini
Original Assignee
Mitsubishi Heavy Industries, Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Mitsubishi Heavy Industries, Ltd. filed Critical Mitsubishi Heavy Industries, Ltd.
Priority to PCT/JP2012/004398 priority Critical patent/WO2014006662A1/en
Publication of WO2014006662A1 publication Critical patent/WO2014006662A1/en

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/12Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
    • G01R31/1227Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/28Testing of electronic circuits, e.g. by signal tracer
    • G01R31/2832Specific tests of electronic circuits not provided for elsewhere
    • G01R31/2836Fault-finding or characterising
    • G01R31/2846Fault-finding or characterising using hard- or software simulation or using knowledge-based systems, e.g. expert systems, artificial intelligence or interactive algorithms

Abstract

The invention relates to a method for diagnosing an electric device, preferably a high-voltage device, comprising the steps of acquisition of measured values of operational parameters of the electric device to obtain a first set of data, which first set of data is analogue data, amplifying the first set of data and converting the amplified first set of data from analogue to digital to obtain a second set of data from the first set of data, filtering the second set of data by means of predetermined filtering processes to obtain a third set of data from the second set of data, processing the third set of data by means of at least one statistical algorithm to obtain a set of statistical features of the third set of data, these statistical features of the third set of data forming a fourth set of data, and applying a neural network processing step to the fourth set of data by utilizing a neural network in order to classify the fourth set of data so as to obtain a fifth set of data that characterizes an diagnostic status of the electric device.

Description

METHOD FOR DIAGNOSING AN ELECTRIC DEVICE

The invention relates to a method for diagnosing an electric device, preferably a high voltage device.

The process of determining the diagnostic status of an electric device is a very crucial task. Typically, electric devices, especially high voltage devices, are prone to defects or failures which can severely influence the operation and the lifetime of the device. Usually, a severe damage is the result of smaller defects or failures leading to this damage. However, if these smaller, not quite so severe, defects go unnoticed very costly or ultimate failure of the electric devices can result.

An approach to avoid severe damages resulting from smaller defects is to determine the diagnostic status of an electric device which preferably comprises information of defects or failures of the electric device. One example of small defects or failures is partial discharges which affect the insulation in an electric device. Even though partial discharges are considered small defects, they can after a while lead to the failure of the entire insulation which can result in a very severe damage of the electric device. In order to detect small defects such as partial discharges extensive measurements are necessary which typically have to be carried out on particular components of the electric device so that diagnosing the electric device results in a down time in which the electric device cannot operate.

Concerning the example of partial discharges, detailed visual inspections are usually necessary to detect voids or cracks in the insulation of the electric device caused by the partial discharges. If the electric device is a motor or a generator, this kind of diagnosing method would usually result in a complete dewinding and rewinding of the coils of the generator or motor yielding extremely high costs, extensive inspection time and a long down-time of the generator or motor.

Online diagnose methods for electric devices which can assess the diagnostic status of an electric device during normal operation are known. However, the known methods do not render the same quality of the results of the diagnosis as it is usual for other methods which are not online, usually due to a very high noise level and the difficult and comprehensive processing of data.

It is an object of the present invention to provide an improved method for diagnosing an electric device, preferably a high-voltage device.

According to the present invention, the diagnosis method comprises the following steps: measured values of operational parameters of the electric device are acquired to obtain a first set of data, which first set of data is analogue data. The first set of data is amplified and the amplified first set of data is converted from analogue to digital to obtain a second set of data from the first set of data. The second set of data is filtered by means of predetermined filtering processes to obtain a third set of data from the second set of data. The third set of data is processed by means of at least one statistical algorithm to obtain a set of statistical features of the third set of data. These statistical features of the third set of data form a fourth set of data on which a neural network processing step is applied by utilizing a neural network in order to classify the fourth set of data so as to obtain a fifth set of data that characterizes a diagnostic status of the electric device. Preferably, the above described steps are performed consecutively in the above described order.

The term "electric device" refers to any device which produces electricity, is powered by electricity or conducts electricity. High voltage devices are electric devices which operate or function at high voltages or produce high voltages.

Diagnosing signifies the process of attempting to determine or identify a diagnostic status of the electric device. A diagnostic status of the electric device could be related to any kind of information of failures or defects of the electric device. Concerning a failure or a defect, the diagnostic status preferably comprises information about the severity, location, duration or any other kind of significant information of the failure or defect. In particular, the diagnostic status of the electric device may refer, for instance, to an aging status of insulation elements of the electric device.

The invention comprises the step of the acquisition of measured values of operational parameters of the electric device in order to obtain a first analogue set of data. An operational parameter is defined by characterizing the operation of the electric device. The acquisition is preferably achieved by measuring values of an operational parameter of the electric device over a certain period of time. In particular, an input to the electric device is provided. The response of the electric device to the provided input is measured by measuring the operational parameters of the electric device.

The first set of data is amplified and converted from analogue to digital to obtain a second set of data. The second set of data is filtered by means of predetermined filtering processes to obtain a third set of data from the second set of data. The filtering step can comprise analogue filtering processes as well as digital filtering processes. Digital filtering is preferably applied in order to delete parts of the second set of data which can be misleading for the further diagnosis process, especially for the classification of the fourth set of data. Concerning the digital filtering process, it is preferably ensured that a reference between the measured values of the operational parameters and the provided input to the electric device is determined. Further preferably, the filtering processes are conducted automatically. Preferably, the filtering processes comprise noise filtering, waveform superimposing or disturbance and interference suppression of abnormal signal waveforms.

The third set of data is processed by means of at least one statistical algorithm to obtain a set of statistical features of the third set of data. Preferably, each applied statistical algorithm renders one statistical feature providing specific information of the third set of data so that the obtained set of features, representing a fourth set of data, characterizes the third set of data.

A neural network processing step is applied to the fourth set of data in order to classify the fourth set of data so as to obtain a fifth set of data that characterizes a diagnostic status of the electric device. The fifth set of data can be a signal distribution, a signal pattern or any other kind of data. Classifying signifies assigning the fourth set of data to at least one of previously set classes. Each of these classes preferably represents a diagnostic status of the electric device so that after the classification step the diagnostic status of the electric device is determined. Preferably, the diagnostic status of the electric device comprises failures of the device so that the fifth set of data characterizes a classification of a type of failures of the electric device. By this way, corrective actions can be planned long before a severe damage to the electric device occurs, resulting in reduced unscheduled down-time of the electric device.

In a preferred embodiment, a neural network is established by training a learning algorithm on a set of test data prior to applying a neural network processing step on the fourth set of data. The establishment of the neural network is based on a data driven technique using test data for training. The test data preferably comprises a plurality of stored samples of data patterns characterizing diagnostic statuses of the electric device. More preferably, the test data comprises input data given to the neural network. This input data simulates various diagnostic statutes of the electric device. Furthermore, the test data includes the answer at which the network should arrive, i.e. resulting class to which the neural network processing step should classify the input data.

By feeding the network in this way, a learning algorithm of the neural network is forced to adapt by rebuilding its logical structures. The network will thus develop classification capabilities which go beyond the test data set. This capability, namely to classify unknown data which has not been part of the test data, is the generalization capability of the network and is one of the major advantages of neural networks, since it will almost never be possible to feed a network all possible patterns but only on a selection.

Concerning the amount of data in the test data, a compromise has to be found between a too small data set, not allowing a comprehensive leaning effect of the network, and a too big data set, having the risk of overfitting the neural network, i.e. that the network is overly specialized on the training set of data and loses its capability of generalizing. After establishing a neural network with the test set of data, the classification process does not only rely on comparing the actual set of data with the stored samples of data patterns but also on the generalization abilities of the network.

According to a further embodiment of the invention, the operational parameters of the electric device comprise an electric voltage. Further preferably, the input provided to the electric device is also a voltage. The response of the electric device to the input voltage in form of a voltage is measured. Preferably, the amplitude of the input voltage follows a repetitive time behavior so that the pattern of the input voltage is repeated every cycle, as it is the case in voltages with a sinus or square waveform. The measured values of the operational parameters comprise the response of the electric device to the input voltage. When filtering the measured processed values it is important to keep the reference phase between the input voltage and the values of the operational parameters in form of a voltage.

Preferably, the electric voltage is measured at electrical connection ports of the electric device. Typically, connection ports are weak links in an electric device, prone to effects such as partial discharges.

Further preferably, the values of the operational parameters of the electric device comprise portions of the electric voltage, said portions of the electric voltage resulting from partial discharges within the electric device and wherein the fifth set of data characterizes a classification of the diagnostic status of electric insulation elements of the electric device with respect to said partial discharges. The values of the operational parameter preferably comprise the input voltage applied to the electric device and the response of the electric device superimposed on the input voltage. The response of the electric device in particular results from partial discharge phenomena in the electric device. Usually, the partial discharges occur within electric insulation elements of the electric device and are defined as localized dielectric breakdowns of a small portion of an electrical insulation of the electric device. Partial discharges usually begin within voids, cracks or inclusions within an insulation element. Typically, partial discharge phenomena occur under high voltage stress. The amplitude of the partial discharge processes is usually much smaller than the amplitude of the input voltage. This makes the amplification of the first set of data very important. When applying an input voltage to the electric device, partial discharges are preferably triggered by this input voltage resulting in the portions of the measured electric voltage. The portions of the measured electric voltage comprise information about the partial discharges induced by the input voltage which is given to the electric device.

The fifth set of data obtained by the method of the invention preferably characterizes a classification of the diagnostic status of electric insulation elements of the electric device with respect to the partial discharges. Such a diagnostic status could comprise the location and the size of defects, such as voids or cracks, in the electric insulation elements. Further preferably, the fifth set of data is additionally postprocessed for evaluation in order to obtain final diagnostic data from it. Such diagnostic data can comprise information about the general aging status of the insulation elements of the electric device or information about an actual approximation of a residual lifetime of at least parts of the electric device, preferably the insulation elements of the electric device. This information can be extremely important for maintenance. Estimating the residual lifetime of parts of the electric device, in particular the insulation elements, by means of the present diagnosing method eliminates the need for opening electric devices and inspecting them visually or by means of any other more complicated and elaborate diagnosis method.

Preferably, the portions of the electric voltage are extracted from the measured electric voltage by means of averaging the measured voltage and superimposing a compensating voltage pattern. Typically, the input voltage is applied to the electric device for a certain number of cycles of the input voltage. Consequently, the measured voltage displays the same time behavior as the input voltage for the same number of cycles. Even though all of the cycles of the input voltage have the exact same shape of amplitude, the response of the electric device to the input voltage in form of portions of the electric voltage resulting from partial discharge processes will never be exactly the same. As a consequence, an average distribution of the cycles of the data is calculated which displays for every phase the averaged measured value. This average distribution represents the averaged measured values of the operational parameter of the electric device over one phase cycle. Preferably, the average distribution is an average voltage distribution of the averaged measured electric voltage of the electric device over one cycle. The resulting average distribution is representative for the electric device for the certain period of time in which the measurement took place. In particular, the average distribution is representative for the partial discharges occurring in the electric device so that it corresponds to a partial discharge pulse pattern. In order to extract the information about the partial discharges from the input voltage serving as a trigger for the partial discharge processes and as a carrier voltage, a compensating voltage pattern is superimposed with the measured electric voltage. Compensating signifies that the compensating voltage pattern displays the reverse pattern of the input voltage. After superimposing these two voltages with mutually reverse patterns the resulting voltage at every degree equals zero. However, the portions of the electric voltage resulting from the partial discharges are not cancelled out and are therefore extracted as a characteristic distribution formed by a pulse pattern.

According to a further embodiment, the averaging step is performed on the third set of data. Further preferably, the average distribution is normalized, i.e. each value is divided by the maximum absolute value recorded for the corresponding distribution. In such a way, a normalized average distribution is obtained which is related to the type of failures to be classified.

According to another preferred embodiment of the invention, an average distribution of the third set of data is transformed into an Rn vector so as to perform phase analysis, amplitude analysis or repetition rate analysis. A statistical estimator is a mathematical operator which is applied to data, typically in the form of distributions, in order to extract specific information. In particular, the set of statistical estimators comprises Kurtosis, Skewness, Cross Correlation, Asymmetry, Weibull Alpha or Weibull Beta operators. Each statistical estimator renders at least one statistical feature which describes features of the third set of data, especially the distribution of the third set of data. By applying a number of statistical estimators to the third set of data a set of statistical features is obtained. This set of statistical features characterizes the third set of data. Most preferably, the set of statistical features represents a fingerprint of the third set of data.

In a further preferred embodiment, the statistical estimators to be applied to the third set of data are customized to the electrical device by selecting them according to characteristics of the electric device. Preferably, the statistical estimators are adapted in order to fit the specific electric device. Customizing the set of statistical estimators as well as the statistical estimators to the specific electric device is very crucial since only after this step all of the necessary and desired information can be extracted from the measured values of operational operators. Selecting the right kind of statistical estimators greatly improves the ability of the diagnosis method to discriminate between different distributions of the third set of data. Since the set of statistical features obtained by the application of the selected statistical estimators is fed into the neutral network for classification, selecting suitable estimators greatly improves the classification step of the neural network. Preferably, various extensive measurements are conducted to characterize the electric device in great detail. These characteristics of the electric device as well as the kind of information which is desired to be extracted by the diagnosis method determine the selection of the statistical estimators. By applying this customization step, the method is not only able to generally classify diagnostic statuses of the electric device which would render only very rough and not sufficiently detailed information. In contrast, by customizing the statistical estimators the method is able to provide detailed, specific and reliable information about the diagnostic status of the electric device. The quality and reliability of the resulting diagnostic statuses equal those achieved by visual inspection methods but with much less effort and related costs.

In a further embodiment, each piece of data in the set of statistical features forming the fourth set of data shows a specific type of failure and/or a specific approximation of the residual lifetime of the electric device. This is related to the fact that each piece of data of the fourth set of data originates from one statistical estimator rendering specific information. Further preferably, selecting suitable statistical estimators renders a fourth set of data which includes all of the necessary and desired information about the diagnostic status of the electric device.

According to another preferred embodiment of the invention, each piece of data in the set of statistical features forming the fourth set of data is analyzed by means of the neural network. This neural network concludes a diagnosis of the specific type of failures and/or the specific approximation of a residual lifetime of the electric device. Preferably, the neural network yields information about defects in the insulation elements of the electric device. By the classification step of the normal network, the defects in the insulation elements can be exactly determined, such as their size and their location.

Preferably, the diagnosis concluded by the neural network is performed after an appropriate amount of data sets is collected under real operation conditions of the electric device. This amount of data sets is typically employed to characterize the specific electric device and then select and adapt suitable statistical estimators for processing the third set of data.

In another preferred embodiment of the invention, the diagnosis is performed in intervals to determine a countermeasure of the specific type of failures recognized. These time intervals can be chosen according to maintenance schedules or due to conclusions from previous diagnostic statuses of the electric device. By regularly employing the above method, a diagnostic status of the electric device can be monitored and the change of the diagnostic status can be determined so that a countermeasure and the appropriate time frame for applying the countermeasure can be determined. A countermeasure can be any corrective or predictive measure to influence the diagnostic status of the electric device. In the preferable case that the diagnoses aims to determine defects of insulation elements of an electric device or the aging status of the insulation elements, an appropriate time frame for replacing the insulation elements as a countermeasure can be concluded.

According to a further embodiment of the invention, the method of the invention is applied to an electric device which is a component of a wind turbine generator. Such electrical devices preferably comprise generators, transformers or high voltage cables. Especially concerning off-shore wind turbine generators, the method of the invention enables a remote diagnosis of the diagnostic status of the electric devices in the wind turbine. As a consequence, less down-town time of a wind turbine generator results due to severe failures of insulation elements of electric devices in the wind turbine generator. The monitoring and the maintenance of the electric devices can be achieved substantially more cost and time effective.

The invention will be described below with reference to the following schematic figures:
Figure 1 is a cross sectional view of an insulation element. Figure 2 is a measured voltage on an electric device. Figure 3 is a flow chart of a method according to the invention.

Figure 1 shows a cross-sectional view of an insulation element 11 which is located between two conductors 12, 13. Within the insulating material 14 of the insulation element 11 voids 15, 16 are formed. Voids are hollow spaces within the insulating material which are filled with gas. The dielectric constant of the voids 15, 16 is considerably less than the dielectric constant of the surrounding insulating material 14. This fact results in electric fields across the voids 15, 16 which are essentially higher than along an equally great distance within the insulating material 14. If the voltage across a void 15, 16 exceeds the electric breakdown voltage level for the gas within the void, a partial discharge process will occur which bridges a part of the insulation element 11. Once begun, partial discharges can affect the insulation elements severely by progressively deteriorating the insulating material 14 electrically and physically. The effect of partial discharges usually cumulates so that conducting discharge channels are formed. This process typically results ultimately in the electrical breakdown of the entire insulation element 11 which in turn can lead to a complete failure or destruction of the electric device.

In figure 2 values of a measured electric voltage 17 are shown. In figure 2 these measured values of the measured electric voltage 17 form a pulse pattern over two phase cycles 19. Typically, the acquisition of data comprises hundreds of cycles. As it can be seen in figure 2, the measured electric voltage 17 is composed of a sine voltage pattern 17a, which has been applied to the electric device to trigger partial discharge processes, and portions 17b of the electric voltage 17 which result from partial discharges within insulation elements of the electric device. The amplitudes 18b of these portions 17b of the electric voltage 17 are typically significantly smaller than the amplitudes 18a of the sine voltage pattern 17a.

Figure 3 shows a flow chart of a method 10 according to the present invention. Preferably, the displayed steps of the flow chart are conducted consecutively from the upper end of the flow chart towards its lower end. The first step 20 comprises the acquisition of measured values of operational parameters of the electric device for obtaining an analogue first set of data 21. In the second step 22 the first set of data 21 is amplified and converted from analogue to digital to obtain a second set of data 23. In the third step 24 the second set of data 23 is filtered by means of predetermined filtering processes so that a third set of data 25 from the second set of data 23 is obtained. The third set of data 25 is processed in a fourth step 26 by applying at least one statistical algorithm for obtaining a set of statistical features which from a fourth set of data 27. A neural network processing step is applied in a fifth step 28 to the fourth set of data 27 by utilizing a neural network in order to classify the fourth set of data 27. In this way, a fifth set of data 29 is obtained characterizing a diagnostic status of the electric device.

Claims (20)

  1. A method for diagnosing an electric device, preferably a high-voltage device, comprising the steps of

    a) acquiring measured values of operational parameters of the electric device to obtain a first set of data, said first set of data is analogue data,
    b) amplifying the first set of data and converting the amplified first set of data from analogue to digital to obtain a second set of data from the first set of data,
    c) filtering the second set of data by means of predetermined filtering processes to obtain a third set of data from the second set of data,
    d) processing the third set of data by means of at least one statistical algorithm to obtain a set of statistical features of the third set of data, said statistical features of the third set of data forming a fourth set of data, and
    e) applying a neural network processing step to the fourth set of data by utilizing a neural network in order to classify the fourth set of data so as to obtain a fifth set of data that characterizes a diagnostic status of the electric device.
  2. The method according to claim 1,
    wherein prior to step e) a neural network is established by training a learning algorithm on a set of test data.
  3. The method according to claim 1,
    wherein the operational parameters of the electric device comprise an electric voltage.
  4. The method according to claim 3,
    wherein the electric voltage is measured at electrical connection ports of the electric device.
  5. The method according to claim 2,
    wherein the test data comprises a plurality of stored samples of data patterns characterizing diagnostic statuses of the electric device.
  6. The method according to claim 1,
    wherein the fifth set of data characterizes a classification of a type of failures of the electric device.
  7. The method according to claim 3,
    wherein the measured values of the operational parameters of the electric device comprise portions of the electric voltage, said portions of the electric voltage resulting from partial discharges within the electric device, wherein the fifth set of data characterizes a classification of the status of electric insulation elements of the electric device with respect to said partial discharges.
  8. The method according to claim 7,
    wherein the portions of the electric voltage are extracted from the measured electric voltage by means of averaging the measured voltage and superimposing a compensating voltage pattern.
  9. The method according to claim 7,
    wherein the partial discharges occur within said electric insulation elements of the electric device.
  10. The method according to claim 7,
    comprising an additional step of postprocessing for evaluation of the fifth set of data so as to obtain final diagnostic data from the fifth set of data, preferably aging status information of the insulation elements of the electric device or information about an actual approximation of a residual lifetime of at least parts of the electric device.
  11. The method according to claim 1,
    wherein an average distribution of the third set of data is transformed by means of the statistical estimators into an Rn vector so as to perform phase analysis, amplitude analysis or repetition rate analysis.
  12. The method according to claim 11,
    wherein a set of statistical estimators comprises Kurtosis, Skewness, Cross Correlation, Asymmetry, Weibull Alpha or Weibull Beta operators.
  13. The method according to claim 11,
    wherein the statistical estimators to be applied are customized to the electrical device by selecting them according to characteristics of the electric device.
  14. The method according to claim 7,
    wherein processing of the third set of data results in a normalized average distribution related to the type of failures to be classified.
  15. The method according to claim 12,
    wherein each piece of data in the set of statistical features forming the fourth set of data shows a specific type of failures and/or a specific approximation of the residual lifetime of the electric device.
  16. The method according to claim 1,
    wherein each piece of data in the set of statistical features forming the fourth set of data is analyzed by means of the neural network which concludes a diagnosis of the specific type of failures and/or the specific approximation of a residual lifetime of the electric device.
  17. The method according to claim 16,
    wherein the diagnosis is performed after an appropriate amount of data sets is collected under real operation conditions of the electric device.
  18. The method according to claim 16,
    wherein the diagnosis is performed in intervals to determine a countermeasure of the specific type of failures recognized.
  19. The method according to claim 1,
    wherein the filtering processes comprise noise filtering, waveform superimposing or disturbance and interference suppression of abnormal signal waveforms.
  20. The method according to claim 1,
    wherein the method is applied to an electric device, such as a generator, a transformer or a high voltage cable, said electric device preferably being a component of a wind turbine.
PCT/JP2012/004398 2012-07-06 2012-07-06 Method for diagnosing an electric device WO2014006662A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2012/004398 WO2014006662A1 (en) 2012-07-06 2012-07-06 Method for diagnosing an electric device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2012/004398 WO2014006662A1 (en) 2012-07-06 2012-07-06 Method for diagnosing an electric device

Publications (1)

Publication Number Publication Date
WO2014006662A1 true WO2014006662A1 (en) 2014-01-09

Family

ID=46548780

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2012/004398 WO2014006662A1 (en) 2012-07-06 2012-07-06 Method for diagnosing an electric device

Country Status (1)

Country Link
WO (1) WO2014006662A1 (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016070139A1 (en) * 2014-10-30 2016-05-06 Hansell Jerritt Harold System, method and apparatus for grid location
US9380545B2 (en) 2011-08-03 2016-06-28 Astrolink International Llc System and methods for synchronizing edge devices on channels without carrier sense
US9438312B2 (en) 2013-06-06 2016-09-06 Astrolink International Llc System and method for inferring schematic relationships between load points and service transformers
US9647994B2 (en) 2011-06-09 2017-05-09 Astrolink International Llc System and method for grid based cyber security
US10001514B2 (en) 2013-06-13 2018-06-19 Astrolink International Llc System and method for detecting and localizing non-technical losses in an electrical power distribution grid
US10079765B2 (en) 2014-10-30 2018-09-18 Astrolink International Llc System and methods for assigning slots and resolving slot conflicts in an electrical distribution grid
US10097240B2 (en) 2013-02-19 2018-10-09 Astrolink International, Llc System and method for inferring schematic and topological properties of an electrical distribution grid
US10459411B2 (en) 2011-04-15 2019-10-29 Astrolink International Llc System and method for single and multizonal optimization of utility services delivery and utilization

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090177420A1 (en) * 2005-05-20 2009-07-09 Daniel Fournier Detection, localization and interpretation of partial discharge

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090177420A1 (en) * 2005-05-20 2009-07-09 Daniel Fournier Detection, localization and interpretation of partial discharge

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
IBRAHIM Y SHURRAB ET AL: "Partial discharge on-line monitoring of outdoor insulators", ELECTRICAL INSULATION (ISEI), CONFERENCE RECORD OF THE 2012 IEEE INTERNATIONAL SYMPOSIUM ON, IEEE, 10 June 2012 (2012-06-10), pages 391-394, XP032211442, DOI: 10.1109/ELINSL.2012.6251496 ISBN: 978-1-4673-0488-7 *
KRANZ H-G: "DIAGNOSIS OF PARTIAL DISCHARGE SIGNALS USING NEURAL NETWORKS AND MINIMUM DISTANCE CLASSIFICATION", IEEE TRANSACTIONS ON ELECTRICAL INSULATION, 31968 1, vol. 28, no. 6, 1 December 1993 (1993-12-01), pages 1016-1024, XP002073089, DOI: 10.1109/14.249375 *
None
TAO HONG ET AL: "Detection and Classification of Partial Discharge Using a Feature Decomposition-Based Modular Neural Network", IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 50, no. 5, 1 October 2001 (2001-10-01), XP011025365, ISSN: 0018-9456 *

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10459411B2 (en) 2011-04-15 2019-10-29 Astrolink International Llc System and method for single and multizonal optimization of utility services delivery and utilization
US9647994B2 (en) 2011-06-09 2017-05-09 Astrolink International Llc System and method for grid based cyber security
US10356055B2 (en) 2011-06-09 2019-07-16 Astrolink International Llc System and method for grid based cyber security
US9380545B2 (en) 2011-08-03 2016-06-28 Astrolink International Llc System and methods for synchronizing edge devices on channels without carrier sense
US9848446B2 (en) 2011-08-03 2017-12-19 Astrolink International Llc System and methods for synchronizing edge devices on channels without carrier sense
US10097240B2 (en) 2013-02-19 2018-10-09 Astrolink International, Llc System and method for inferring schematic and topological properties of an electrical distribution grid
US9438312B2 (en) 2013-06-06 2016-09-06 Astrolink International Llc System and method for inferring schematic relationships between load points and service transformers
US10001514B2 (en) 2013-06-13 2018-06-19 Astrolink International Llc System and method for detecting and localizing non-technical losses in an electrical power distribution grid
US9853498B2 (en) 2014-10-30 2017-12-26 Astrolink International Llc System, method, and apparatus for grid location
US10079765B2 (en) 2014-10-30 2018-09-18 Astrolink International Llc System and methods for assigning slots and resolving slot conflicts in an electrical distribution grid
US10020677B2 (en) 2014-10-30 2018-07-10 Astrolink International Llc System, method, and apparatus for grid location
CN107003346A (en) * 2014-10-30 2017-08-01 艾斯通林克国际有限责任公司 System, the method and apparatus of grid position
WO2016070139A1 (en) * 2014-10-30 2016-05-06 Hansell Jerritt Harold System, method and apparatus for grid location

Similar Documents

Publication Publication Date Title
ES2302335T3 (en) Control of partial internal downloads in an electric transformer.
Cavallini et al. A new methodology for the identification of PD in electrical apparatus: properties and applications
US7254520B2 (en) Testing of wire systems and end devices installed in industrial processes
Behjat et al. Diagnosing shorted turns on the windings of power transformers based upon online FRA using capacitive and inductive couplings
JP4223949B2 (en) Detection method of foreign matter in fluid
US7676333B2 (en) Method and apparatus for analyzing partial discharges in electrical devices
EP0909390A3 (en) Statistical pattern analysis methods of partial discharge measurments in high voltage insulation
CH700936B1 (en) Method and system for determining a remaining before failure of an electric motor system time.
Setayeshmehr et al. On-line monitoring and diagnoses of power transformer bushings
JP2009115505A (en) Winding inspection device and inspection method
Landry et al. An improved vibration analysis algorithm as a diagnostic tool for detecting mechanical anomalies on power circuit breakers
US20080088314A1 (en) Methods and apparatus for analyzing partial discharge in electrical machinery
KR100541996B1 (en) A dignosis method of insulation of electric apparatus
Aschenbrenner et al. On line PD measurements and diagnosis on power transformers
JP2009213346A (en) Methods and systems for detecting rotor field ground faults in rotating machinery
Stone The use of partial discharge measurements to assess the condition of rotating machine insulation
US7076404B2 (en) Method and apparatus for the analysis and monitoring of the partial discharge behavior of an electrical operating device
JP3953087B2 (en) Insulation inspection device and insulation inspection method
US9046563B2 (en) Arcing event detection
US6686746B2 (en) Method and apparatus for monitoring integrity of wires or electrical cables
JP2014211455A (en) Method for determining and evaluating display of eddy current of crack particularly in object to be tested comprising conductive material
Lee et al. An online groundwall and phase-to-phase insulation quality assessment technique for AC-machine stator windings
JP6023449B2 (en) Automatic test system for digital display system
EP2270483A1 (en) Method and device for monitoring a zone of metal
Guastavino et al. Tree growth monitoring by means of digital partial discharge measurements

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12737901

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase in:

Ref country code: DE

122 Ep: pct app. not ent. europ. phase

Ref document number: 12737901

Country of ref document: EP

Kind code of ref document: A1