JP2018179598A - Diagnostic method, diagnostic device, and program - Google Patents

Abstract

A diagnostic method, a diagnostic device, and a program capable of accurately diagnosing the insulation of electric power equipment are provided. A diagnostic method according to an embodiment has the following steps. That is, a process in which a signal obtained by a sensor for detecting discharge from a diagnosis target device is divided into a plurality of frequency bands by a processing unit. Then, the analysis unit calculates a predetermined first parameter value based on the signal for each frequency band divided by the processing unit, and stores the calculated first parameter value and the calculated first parameter value in advance in a database. calculating the similarity of the partial discharge signal to the second parameter value; [Selection drawing] Fig. 1

Description

  Embodiments of the present invention relate to a diagnostic method, a diagnostic device, and a program.

  Electric power equipment is an important equipment that supports social infrastructure, and long-term stable operation is required. In order to ensure stable operation, it is necessary to grasp the deterioration of power facilities and to carry out maintenance and renewal on a criminal basis. In a power receiving device such as a switchgear, when partial discharge occurs, the insulation performance may be significantly reduced, which may lead to a ground fault. Therefore, partial discharge detection is generally performed as a method for detecting a decrease in insulation performance.

  The method of detecting the current flowing to the ground line detects all the electrical signals including partial discharge signals emitted from many devices, so it may be difficult to diagnose partial discharge correctly due to the influence of noise. . The method of detecting an electromagnetic wave may be affected by reflection in an electric device or an electromagnetic noise, and correct diagnosis may be difficult.

  In these conventional techniques, it is determined whether or not the peak recognized as a partial discharge signal is synchronized with the commercial voltage in order to prevent erroneous diagnosis, and the signal level determined to be out of synchronization and the time from the synchronization signal Insulation diagnostic devices have been proposed that use different differences to determine discharge signals and reduce false diagnostics. In addition, there is also proposed an electrical equipment insulation diagnosis apparatus provided with a method of discriminating between a broadcast wave and a discharge signal by having measuring means capable of measuring a plurality of frequency bands.

  However, in the method of determining the discharge signal by synchronization with the power supply frequency, it may not be possible to accurately capture the partial discharge phenomenon including noise synchronized with the power supply. Moreover, in the method having a plurality of frequency measurement means, there is a possibility of misdiagnosis in measurement in a place where there are many radio noises, which is disadvantageous in cost.

JP, 2008-045977, A JP-A-9-292433

  The problem to be solved by the present invention is to provide a diagnostic method, a diagnostic device, and a program capable of accurately performing insulation diagnosis of a power device.

  The diagnostic method of the embodiment has the following steps. That is, a process in which a processing unit divides a signal acquired by a sensor for detecting discharge from a device to be diagnosed into a plurality of frequency bands. Then, based on the signal of each frequency band divided by the processing unit, the analysis unit calculates a predetermined first parameter value, and stores the calculated first parameter value in a database in advance. Calculating the similarity with the second parameter value of the partial discharge signal.

FIG. 1 is a block diagram showing a schematic functional configuration of a diagnostic device of a first embodiment. 4 is a flowchart showing details of processing by the diagnostic device of the first embodiment. FIG. 5 is a schematic view showing a concept of processing of dividing a signal into a plurality of frequency bands in a processing unit of the first embodiment. The graph which shows the example which the process part divided | segmented into the several frequency band the signal obtained with the sensor of 1st Embodiment. The graph which shows a part of result of parameter calculation which the analysis part of 1st Embodiment performs. Schematic which shows the structure of the parameter information stored in the database of 1st Embodiment. FIG. 7 is a block diagram showing a schematic functional configuration of a diagnostic device of a second embodiment. The flowchart which shows the detail of the processing by the diagnostic device of a 2nd embodiment. Schematic which shows the signal for demonstrating the content of the process by the determination part of 2nd Embodiment. Schematic which shows the example of a structure of the database of 2nd Embodiment. FIG. 8 is a block diagram showing a schematic functional configuration of a diagnostic device of a third embodiment.

  Hereinafter, a diagnostic method, a diagnostic device, and a program of the embodiment will be described with reference to the drawings.

First Embodiment
FIG. 1 is a block diagram showing a schematic functional configuration of a diagnostic device according to the first embodiment. As illustrated, the diagnostic device 1 includes a sensor 101, a processing unit 102, an analysis unit 103, a display unit 105, and a database 200.
The sensor 101 detects a partial discharge signal. The sensor 101 outputs the detected signal.
The processing unit 102 processes the signal detected by the sensor 101. Specifically, the processing unit 102 divides the signal acquired by the sensor 101 into a plurality of frequency bands.
The analysis unit 103 analyzes the signal processed by the processing unit 102.
Specifically, the analysis unit 103 calculates a predetermined parameter value (referred to as a first parameter value) based on the signal of each frequency band divided by the processing unit 102. Then, the analysis unit 103 calculates the similarity between the calculated first parameter value and the parameter value (referred to as a second parameter value) related to the partial discharge signal stored in advance in the database.
The display unit 105 displays the processing result of each unit up to the previous stage.
The database 200 holds data necessary for analysis processing and the like.

  In addition, as the sensor 101, for example, a method of acquiring a signal of a ground wiring (the same applies to a part electrically grounded) may be used. By doing this, for example, it is possible to obtain the potential of the panel surface of the switchboard to perform diagnosis. Further, even when the sensor is not installed in advance, a power failure and opening of the board are not necessary.

  FIG. 2 is a flowchart showing details of processing by the diagnostic device 1. In addition, the apparatus which the diagnostic apparatus 1 makes an object of diagnosis is an electric power apparatus which handles alternating current power supply. The power device is supplied with commercial AC power of a predetermined frequency (for example, 50 Hz or 60 Hz). The power device to be diagnosed is, for example, a power distribution device such as a switch gear. The following description will be made along the flowchart.

First, in step S <b> 1, the processing unit 102 divides the signal acquired by the sensor 101 into cycles of AC power supplied to the diagnosis target device. That is, the processing unit 102 divides the signal acquired by the sensor 101 into sections of one cycle from phase 0 degrees to 360 degrees.
Next, in step S2, the processing unit 102 divides the signal divided in step S1 into a plurality of frequency bands. In this step, the signal is divided into at least two frequency bands.

Next, in step S3, the analysis unit 103 performs predetermined processing for each of the frequency bands divided in step S2 to calculate values of a plurality of predetermined parameters.
Next, in step S4, the analysis unit 103 compares the value of the parameter group calculated in step S3 with the value of the parameter group stored in advance in the database. Specifically, the analysis unit 103 calculates the similarity between the calculated value of the parameter group and the value of the parameter group stored in the database 200. In the database 200, values of parameter groups in the case where the signal is a signal corresponding to partial discharge are stored in advance. Also, the similarity calculated in this step is, for example, defined in advance as a norm in a multidimensional space formed by the parameter group.

  When it is determined that the similarity is high based on the degree of similarity calculated in step S4 and a predetermined threshold, it can be estimated that the type of discharge has been detected.

  FIG. 3 is a schematic view showing a concept of processing of dividing a signal into a plurality of frequency bands in the processing unit 102. As shown in FIG. In the figure, the horizontal axis is frequency (in Hertz (Hz)), and a logarithmic scale is used. The vertical axis is a value representing the strength of the signal. In the figure, a graph shown by a solid line represents a signal acquired by the sensor 101. The processing unit 102 appropriately divides this signal into a plurality of frequency bands.

  The signal acquired by the sensor 101 includes a partial discharge signal by the device to be diagnosed and noise in the environment in which the device to be diagnosed is installed. The frequency components of the signal acquired by the sensor 101 are widely distributed in the range of several hundred kilohertz (kHz) to about 1 gigahertz (GHz). Partial discharge signals occur in a relatively narrow band within this wide range. The number of divisions when the processing unit 102 divides the signal into frequency bands is at least two. Here, the larger the number of divisions, the higher the S / N ratio (signal-to-noise ratio) obtained. However, if the number of divisions is too large, the time required for processing (calculation, etc.) for diagnosis becomes long, and disadvantages such as not being suitable for on-site diagnosis (location of equipment to be diagnosed) or cost increase obtain. Therefore, the number of divisions is preferably in the range of approximately 2 to 8.

In the case where the frequency band to be divided is fixedly determined in advance, the way of dividing the band is arbitrary, for example, as follows.
When the division number is 2, the first frequency band is less than 100 MHz, and the second frequency band is greater than or equal to 100 MHz and less than 1 GHz.
When the division number is 4, the first frequency band is less than 1 MHz, the second frequency band is 1 MHz or more and less than 10 MHz, and the third frequency band is 10 MHz or more and less than 100 MHz, and the fourth frequency band is 100 MHz. More than and less than 1 GHz.
When the division number is 8, the first frequency band is 100 kHz or more and less than 316.2 kHz, and the second frequency band is 316.2 kHz or more and less than 1 MHz, and the third frequency band is 1 MHz or more and less than 3.162 MHz As the fourth frequency band is 3.162 MHz or more and less than 10 MHz, the fifth frequency band is 10 MHz or more and less than 31.62 MHz, and the sixth frequency band is 31.62 MHz or more and less than 100 MHz. The eighth frequency band is 316.2 MHz or more and less than 1 GHz, where the frequency band is 100 MHz or more and less than 316.2 MHz.
Note that the frequency band may be divided other than the division illustrated above.

  FIG. 4 is a graph showing an example in which the signal obtained by the sensor is divided into a plurality of frequency bands. In any of the figures (A), (B1) and (B2), the horizontal axis is time, and the vertical axis is a voltage level. The figure (A) represents the signal which the sensor 101 acquired. The figure (B1) and the figure (B2) show the result of dividing the signal of the figure (A) into two frequency bands. The signal shown in the figure (B1) is a signal of a band whose frequency is higher than the signal shown in the figure (B2). As an example of the external discharge, as a result of acquiring and analyzing the creeping discharge and the resin internal discharge, respectively, it is clear that the frequency bands possessed by the signals from these two discharges are different. That is, it is known that different types of discharge correspond to components of signals in different frequency bands. That is, it is possible to estimate the type of partial discharge by the processing unit 102 dividing the signal into a plurality of frequency bands.

  FIG. 5 is a graph showing a part of the result of parameter calculation performed by the analysis unit 103. In the figures (A) and (B), the horizontal axis is the phase (in degrees) of the AC power supply of the device to be diagnosed, and the vertical axis is the signal strength. This graph represents the result of analysis of the signal output from the processing unit 102 by the analysis unit 103. This graph is a result of plotting one discharge as one point in a plane represented by two axes of phase-signal intensity and separating the plotted points into two clusters. The figure (A) respond | corresponds to insulator internal discharge, and the figure (B) respond | corresponds to a corona discharge. As shown in the figure, the insulator internal discharge in FIG. 6A is distributed in a phase range of approximately 10 degrees to 70 degrees and in a range of 190 degrees to 250 degrees. Also, the signal strength is also distributed within a predetermined range as shown in the figure. On the other hand, the corona discharge in FIG. 6B is distributed in the range of approximately 50 degrees to 90 degrees and in the range of 250 degrees to 310 degrees as phases. Further, the signal intensity is relatively high in the discharge in the range of 50 degrees to 90 degrees, and the signal intensity is relatively low in the discharge in the range of 250 degrees to 310 degrees.

  The parameters illustrated in FIG. 5 are only a part of the values calculated by the analysis unit 103. The analysis unit 103 may calculate another value as a parameter. Further, the analysis unit 103 may calculate yet another parameter based on the distribution shown in FIG. In addition, the analysis unit 103 may calculate parameters representing the other discharge characteristics other than the insulator internal discharge and the corona discharge.

  As parameters calculated by the analysis unit 103, for example, maximum signal strength (maximum sensor voltage) for each half cycle (that is, from 0 degrees to 180 degrees and from 180 degrees to 360 degrees) in the distribution shown in FIG. 5 And average sensor voltage, minimum phase, maximum phase, average phase, skewness representing the shape of distribution, kurtosis representing the shape of distribution, and the like. Also, the difference between the above parameters in the previous half cycle (from phase 0 degree to 180 degrees) and in the later half cycle (phase from 180 degrees to 360 degrees) may be calculated as a parameter.

  FIG. 6 is a schematic diagram showing the configuration of parameter information stored in the database 200. As shown in FIG. As illustrated, parameter information in the database 200 holds information on parameters regarding the discharge signal for each type of discharge. The parameter information is written in a magnetic hard disk drive or a semiconductor memory as data in the form of a two-dimensional table. Each row in the table corresponds to the type of discharge. In the illustrated example, the first row of data corresponds to "insulator internal discharge" and the second row corresponds to "corona discharge". Also, each column of the table corresponds to the type of parameter. In the illustrated example, the table includes, as parameter types, maximum sensor voltage for each half cycle, average sensor voltage, minimum phase, maximum phase, average phase, distribution skewness, distribution kurtosis, and the like. The skewness is an index that represents the asymmetry of the distribution. Also, the kurtosis is an index that represents the peak of the distribution. And this table holds the information (numerical value etc.) of the parameter for each type of discharge.

  As described above with reference to FIG. 2, the analysis unit 103 calculates the values of the parameter group calculated based on the signal detected by the sensor 101 and the values of the parameter group stored in the database 200 in advance. Calculate the similarity of This similarity is defined in advance as, for example, a norm in a multi-dimensional space formed by the parameter group. As an example, a weight value is applied to the absolute value of the difference between the values of each parameter, and the sum is obtained to obtain the degree of similarity. Alternatively, the difference between the values of each parameter may be squared, the weight value may be calculated, the sum may be determined, and the degree of similarity may be calculated. When the degree of similarity is made higher as the numerical value of the degree of similarity is larger, the reciprocal or the reciprocal of the calculation result may be taken as appropriate. Also, the similarity may be determined by another calculation method.

  As described above, according to the present embodiment, the signal detected by the sensor is divided into a plurality of frequency bands, the parameter value is calculated for each frequency band, and the discharge type stored in the database in advance is stored. Find the similarity to the parameter value. This makes it possible to estimate the discharge with high accuracy, to prevent a false diagnosis or to reduce a false diagnosis.

Second Embodiment
Next, a second embodiment will be described. In addition, the description below may be abbreviate | omitted about the matter which is common in the matter already demonstrated by previous embodiment. In the following, matters specific to the present embodiment will be mainly described.

FIG. 7 is a block diagram showing a schematic functional configuration of the diagnostic device according to the present embodiment. As illustrated, the diagnostic device 2 includes a sensor 101, a processing unit 102, an analysis unit 103, a determination unit 104, a display unit 105, and a database 200. That is, the diagnosis device 2 includes the determination unit 104 in addition to the configuration of the diagnosis device 1 according to the previous embodiment. The functions of the sensor 101, the processing unit 102, the analysis unit 103, the display unit 105, and the database 200 are as described in the first embodiment. However, the configuration of the database in the present embodiment will be further described later.
The determination unit 104 makes a determination on the signal analyzed by the analysis unit 103.

FIG. 8 is a flowchart showing details of processing by the diagnostic device 2. The following description will be made along the flowchart.
The processes in steps S1 to S4 are the same as described in FIG.

  Next, in step S5, the determination unit 104 extracts a frequency band in which the degree of similarity calculated in step S4 exceeds a predetermined threshold. Then, the determination unit 104 compares the signal of the frequency band extracted based on the threshold value with the signal of the specific frequency band. Here, the specific frequency band is a predetermined specific frequency band among the frequency bands divided by the processing unit 102. As the "specific frequency band", one including a band from 100 MHz to 1 GHz is preferably selected. In addition, when the band from 100 MHz to 1 GHz is further divided into a plurality of bands, preferably, the band with the highest frequency among the plurality of bands is selected as the “specific frequency band”. That is, the specific frequency band preferably includes at least a part of the band from 100 MHz to 1 GHz. Since all discharge signals include signals of a specific frequency band, it is possible to prevent or reduce false diagnosis by making a determination based on such a signal of the specific frequency band.

  Then, as a result of the above comparison, in both of the frequency band extracted in this step and the above specific frequency band, signals of levels (voltages) equal to or higher than a predetermined threshold are detected at the same time (or almost the same time). If it is (that is, when both are synchronized), the signal is determined to be a partial discharge signal. Here, “almost the same time” means the case where the time difference is less than or equal to a minute predetermined value. As an example, this “predetermined value” is preferably set to half the period corresponding to the center frequency of the frequency band to be determined. By doing so, it is possible to determine the partial discharge signal with high accuracy. For example, when the center frequency of the frequency band to be determined is 1 MHz, the corresponding period is 1 μsec (microsec). Therefore, the threshold of the time difference for determination is set to 0.5 μsec of that half. However, regardless of the frequency of the signal to be determined, a predetermined time difference threshold may be set.

  As a result of the above comparison, if the above conditions are not satisfied, the signal is considered not to be a partial discharge signal and is excluded. Further, the determination unit 104 resynthesizes the signal determined to be a partial discharge signal and the signal of the above-mentioned specific frequency band.

Next, in step S6, the determination unit 104 calculates parameters again for the signal recombined in step S5, and compares it with the parameters stored in the parameter information (see FIG. 6) of the database 200. Specifically, the determination unit 104 calculates the degree of similarity between the two parameter groups. The calculation method of similarity is as already described. The determination unit 104 calculates the degree of similarity with the detected signal for each type of discharge. Note that the determination unit 104 may estimate that the discharge type with the highest degree of similarity corresponds to the detected signal.
Then, the display unit 105 displays and outputs the result of the determination process by the determination unit 104 on a screen or the like.

  That is, the determination unit 104 includes a first signal included in a signal of a specific frequency band that is a specific band among a plurality of frequency bands and the similarity calculated by the analysis unit 103 exceeds a predetermined threshold, and a plurality of frequency bands. The second signal is compared with a second signal which is included in the signal of one of the frequency bands other than the specific frequency band and the similarity calculated by the analysis unit 103 exceeds a predetermined threshold. Then, when the first signal and the second signal are in synchronization, the determination unit 104 determines the first signal and the second signal as the discharge signal, and excludes other than the discharge signal to recombine it. Generate Then, the determination unit 104 calculates a predetermined parameter value (referred to as a third parameter value) for this recombined signal. Then, determination unit 104 calculates the degree of similarity between the calculated third parameter value of the recombined signal and the parameter value (second parameter value) regarding the partial discharge signal stored in advance in the database.

  FIG. 9 is a schematic view showing a signal for explaining the contents of processing by the determination unit 104. As shown in FIG. In each of the figures (A), (B) and (C), the horizontal axis is time (in microseconds), and the vertical axis is the signal level. The figure (A) shows the signal of a specific frequency band. The figure (B) shows the signal of one frequency band other than a specific frequency band. The same figure (C) shows the signal of the result of having performed the process of time synchronization based on the signal of (A) and (B).

  In the signal of the specific frequency band in the figure (A), there are two time intervals in which the signal level exceeds a predetermined threshold, and these time intervals are hatched in the figure. These time zones are referred to as time interval A1 and time interval A2 for convenience. Further, in the signal of the frequency band in FIG. 5B, there are three time intervals in which the signal level exceeds a predetermined threshold, and these time intervals are hatched in the figure. These time zones are referred to as time interval B1, time interval B2, and time interval B3 for the sake of convenience. The determination unit 104 checks the presence or absence of synchronization in a time zone exceeding the threshold between the signal of (A) and the signal of (B). As a result, the determination unit 104 determines that the time interval A1 and the time interval B1 are synchronized. At this time, the determination unit 104 determines the presence or absence of synchronization based on whether or not the start time and the end time of each of the time interval A1 and the time interval B1 match after allowing an error within a predetermined range. Do. As a result, the determination unit 104 re-synthesizes the signal shown in (A) and the signal shown in (B), and outputs the signal shown in (C). In FIG. 6C, signals of both frequency bands are synthesized again in the above-mentioned synchronized time interval (that is, time intervals A1 and B1). Further, in the other time zones (all time intervals other than the synchronized time interval) in FIG.

  As a result of the above-described processing by the determination unit 104, the determination unit 104 outputs the following signal as to the signal recombined (the same figure (C)). That is, the determination unit 104 is a signal in only a time interval considered to be a discharge, and a frequency band in which the discharge is detected (that is, a frequency indicated by (A) and (B) respectively. Output a signal of only band).

  FIG. 10 is a schematic view showing a configuration example of a database in the present embodiment. As shown, the database 200 includes shipping test data 202, design data 203, abnormal discharge data 204, and trend data of sensor signals in addition to the parameter information (parameter information 201 of the discharge signal) shown in FIG. And 205 are stored.

The shipping test data 202 holds data representing the date of manufacture of the device to be diagnosed and data of the result of the shipping test.
The design data 203 holds information on the type of insulating material used in the device to be diagnosed, information on a portion in the device where the insulating material is used, and the like.
The abnormal discharge data 204 holds parameters and the like obtained from sensor signals when various abnormal discharges occur.
Trend data 205 of the sensor signal is data representing the trend of another sensor signal mounted near the sensor.

The determination unit 104 can grasp what the insulating material used for the device to be diagnosed is by referring to the design data 203 described above. In addition, it is possible to grasp where the insulating material is used in which part of the device to be diagnosed. Therefore, the determination unit 104 can estimate the likelihood of occurrence of discharge, the type of discharge, and the like for each portion of the device to be diagnosed as a determination material by combining the various characteristics and the like of the insulating material.
Further, the determination unit 104 can estimate the degree of influence of the noise signal by referring to the trend data 205 of the sensor signal. In addition, based on the estimation result of noise, it is possible to reduce misdiagnosis of discharge due to excessive noise.

Third Embodiment
Next, a third embodiment will be described. In addition, the description below may be abbreviate | omitted about the matter which is common in the matter already demonstrated by previous embodiment. In the following, matters specific to the present embodiment will be mainly described.

  FIG. 11 is a block diagram showing a schematic functional configuration of a diagnosis apparatus according to the present embodiment. As illustrated, the diagnostic device 11 includes a plurality of sensors 1101-1, 1101-2, 1101-3, ..., a processing unit 102, an analysis unit 1103, a determination unit 1104, and a display unit 105. And a database 200. That is, the feature of the diagnostic device 11 is that it includes a plurality of sensors.

The plurality of sensors 1101-1, 1101-2, 1101-3,... Are respectively installed at a plurality of different places of the diagnosis target device. Each of these sensors 1101-1, 1101-2, 1101-3,... Detects and outputs a signal at a grounded point.
The processing unit 102 performs the same process as the process described in the first embodiment or the like on the signal output from each of the sensors 1101-1, 1101-2, 1101-3,. Then, the processing unit 102 outputs the result of dividing each signal of the sensors 1101-1, 1101-2, 1101-3 ... into a plurality of frequency bands.

  The analysis unit 1103 performs the same process as the process described in the first embodiment and the like. However, the analysis unit 1103 compares the strengths of the signals obtained from the sensors 1101-1, 1101-2, 1101-3,. Further, the analysis unit 1103 refers to the information of the installation place of the sensors 1101-1, 1101-2, 1101-3,. The information of the installation place is stored in advance in the database 200, for example. Then, the analysis unit 1103 determines the order of the sensors 1101-1, 1101-2, 1101-3,... In descending order of signal strength for each estimated discharge.

  The determination unit 1104 performs the same process as the process described in the second embodiment. However, the determination unit 1104 compares the strengths of the signals obtained from the sensors 1101-1, 1101-2, 1101-3,. Further, similarly to the analysis unit 103, the determination unit 1104 refers to the information on the installation places of the sensors 1101-1, 1101-2, 1101-3,. Then, the determination unit 1104 determines the order of the sensors 1101-1, 1101-2, 1101-3, ... in descending order of signal strength for each estimated discharge.

  As described above, the analysis unit 1103 and the determination unit 1104 respectively select one of the sensors 1101-1, 1101-2, 1101-3,... Based on the signal strength (intensity) for each estimated discharge. Output order information. Generally, the shorter the distance from the discharge source to the sensor, the stronger the signal detected by the sensor. Therefore, the information output by the analysis unit 1103 and the determination unit 1104 is information for estimating the occurrence position of each discharge.

  Also, instead of estimating the distance from the discharge source to the sensor based on the signal strength, the analysis unit 1103 and the determination unit 1104 use the sensor from the discharge source based on the time (timing) at which the signal was detected. You may make it estimate the distance to. In other words, the analysis unit 1103 and the determination unit 1104 use the fact that the sensor closer to the discharge generation source detects the discharge signal earlier. Then, it is possible to estimate that the sensor that has detected the signal of discharge most quickly is close to the discharge source.

That is, in the present embodiment, a plurality of sensors are provided at a plurality of locations of the device to be diagnosed.
Each of the analysis unit 1103 and the determination unit 1104 may estimate the generation position of the partial discharge in the device to be diagnosed based on the intensity of the signal from each of the sensors. Further, each of the analysis unit 1103 and the determination unit 1104 may estimate the generation position of the partial discharge in the device to be diagnosed based on the timing of the signal from each of the sensors.

In the above embodiments, when the processing unit 102 divides a signal into a plurality of frequency bands, an example has been described in which band divisions are fixedly set in advance. However, the division of the frequency band may be determined based on the signal detected by the sensor. Also, the delimitation of the frequency band may be changed adaptively. The processing unit 102 analyzes, for example, the frequency spectrum of the signal detected by the sensor. Then, it is estimated that the point at which the frequency peak appears is a signal due to discharge. Then, the processing unit 102 divides the frequency band such that each of the plurality of detected peaks is approximately at the center frequency. Then, the processing unit 102 divides the signal passed from the sensor into these frequency bands. As described above, by determining the number of divisions using the peak frequency, diagnosis can be performed at high speed, and diagnosis can be performed by highlighting a strong frequency band.
That is, the processing unit 102 determines the number of frequency bands to be divided according to the number of peaks of the frequency spectrum of the signal of the sensor.

In each of the above embodiments, the analysis unit (103, 1103) and the determination unit (104, 1104) calculate parameter values based on the signals divided for each frequency band, and use these parameter values and the database 200 in advance. The similarity between the registered parameter values was calculated. Then, the analysis unit (103, 1103) and the determination unit (104, 1104) estimate whether or not the signal represents discharge based on the calculated similarity. Instead, the analysis unit (103, 1103) or the determination unit (104, 1104) may determine the presence or absence of the partial discharge signal by comparing with a statistical parameter based on a Gaussian distribution. The noise contained in the signal detected by the sensor follows a Gaussian distribution. However, when the partial discharge signal is superimposed, the signal has a Gaussian distribution and exhibits a different distribution. For example, the signal does not distribute continuously from intensity 0, and does not have a predetermined intermediate value. Therefore, when the discharge signal is included in the signal acquired by the sensor, the analysis unit (103, 1103) or the determination unit (104, 1104) can distinguish the partial discharge signal from the noise.
That is, the analysis unit (103, 1103) and the determination unit (104, 1104) separate from the information representing the characteristic of the discharge stored in the database 200, and the statistical parameter (parameter representing noise) determined based on the Gaussian distribution. The comparison is performed to determine the presence or absence of a significant difference indicated by the partial discharge signal.
In addition, partial discharges are performed using the information representing the characteristics of the discharge stored in the database 200 by the analysis unit (103, 1103) and the determination unit (104, 1104) in combination with information on statistical parameters of Gaussian distribution representing noise. The presence or absence of a signal may be determined.

  According to at least one embodiment described above, the processing unit divides the signal acquired by the sensor into a plurality of frequency bands. Further, the analysis unit calculates a predetermined parameter value based on the signal for each frequency band divided by the processing unit, and the calculated parameter value and the parameter value related to the partial discharge signal stored in the database in advance. Calculate the degree of similarity with With this configuration, it is possible to realize a diagnosis method, a diagnosis apparatus, and the like with less erroneous diagnosis due to noise and the like.

  Note that at least a part of the functions of the diagnostic device in the embodiment described above may be realized by a computer. In that case, a program for realizing this function may be recorded in a computer readable recording medium, and the program recorded in this recording medium may be read and executed by a computer system. Here, the “computer system” includes an OS and hardware such as peripheral devices. The term "computer-readable recording medium" means portable media such as flexible disks, magneto-optical disks, ROMs, CD-ROMs, DVD-ROMs, and USB memories, and storage devices such as hard disks incorporated in computer systems. It means that. Furthermore, “computer-readable recording medium” dynamically holds a program for a short time, like a communication line in the case of transmitting a program via a network such as the Internet or a communication line such as a telephone line. It may also include one that holds a program for a certain period of time, such as volatile memory in a computer system that becomes a server or client in that case. The program may be for realizing a part of the functions described above, or may be realized in combination with the program already recorded in the computer system.

  While certain embodiments of the present invention have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the invention. These embodiments can be implemented in other various forms, and various omissions, replacements, and modifications can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the invention described in the claims and the equivalents thereof as well as included in the scope and the gist of the invention.

1, 2, 11 ... diagnostic device, 101, 1101-1, 1101-2, 1101-3 ... ... sensor, 102 ... processing unit, 103 ... analysis unit, 104 ... determination unit, 105 ... display unit, 200 ... Database, 201 ... Parameter information of discharge signal, 202 ... Shipping test data, 203 ... Design data, 204 ... Abnormal discharge data, 205 ... Trend data of sensor signal, 1103 ... Analysis unit, 1104 ... Determination unit

Claims (9)

A process in which a processing unit divides a signal acquired by a sensor for detecting discharge from a device to be diagnosed into a plurality of frequency bands;
The analysis unit calculates a predetermined first parameter value based on the signal for each frequency band divided by the processing unit, and the calculated first parameter value and a part stored in advance in a database Calculating the similarity with the second parameter value relating to the discharge signal;
A diagnostic method having: A first signal included in a signal of a specific frequency band which is a specific band among the plurality of frequency bands, and the similarity calculated by the analysis part exceeding a predetermined threshold, and the plurality of frequency bands. Comparing with the second signal which is included in the signal of one frequency band other than the specific frequency band of the above and the similarity calculated by the analysis unit exceeds a predetermined threshold, and the first signal and the second signal And the second signal are determined as discharge signals, and a predetermined third parameter value is calculated for a recombined signal that is recombined by excluding the signals other than the discharge signal, Calculating the similarity between the calculated third parameter value of the recombined signal and the second parameter value of the partial discharge signal stored in the database in advance;
The diagnostic method according to claim 1, further comprising The specific frequency band includes at least a part of a band from 100 MHz to 1 GHz.
The diagnostic method according to claim 2. The number of frequency bands when the processing unit divides the signal into the plurality of frequency bands corresponds to the number of peaks of the frequency spectrum of the signal.
The diagnostic method according to any one of claims 1 to 3. A plurality of the sensors are provided at a plurality of locations of the device to be diagnosed, and a generation position of partial discharge in the device to be diagnosed is estimated based on intensities of signals from the respective sensors
The diagnostic method according to any one of claims 1 to 4. A plurality of the sensors are provided at a plurality of locations of the device to be diagnosed, and a generation position of partial discharge in the device to be diagnosed is estimated based on timings of signals from each of the sensors;
The diagnostic method according to any one of claims 1 to 5. Providing the sensor at a position for acquiring a signal of a grounding wire of the device to be diagnosed or a portion to be electrically grounded;
The diagnostic method according to any one of claims 1 to 6. A sensor for detecting discharge from a device to be diagnosed;
A processing unit that divides a signal acquired by the sensor into a plurality of frequency bands;
A predetermined first parameter value is calculated based on the signal for each frequency band divided by the processing unit, and the calculated first parameter value and a partial discharge signal related to a partial discharge signal stored in a database in advance An analysis unit that calculates the degree of similarity with the two parameter values;
Diagnostic device comprising: Computer,
A processing unit that divides a signal acquired by a sensor for detecting discharge from a device to be diagnosed into a plurality of frequency bands;
A predetermined first parameter value is calculated based on the signal for each frequency band divided by the processing unit, and the calculated first parameter value and a partial discharge signal related to a partial discharge signal stored in a database in advance An analysis unit that calculates the degree of similarity with the two parameter values;
Program for functioning as a diagnostic device provided with

Images