JP2015175769A - Insulation abnormality diagnostic device, insulation abnormality diagnostic method and program for insulation abnormality diagnosis - Google Patents

Abstract

PROBLEM TO BE SOLVED: To more exactly diagnose an insulation abnormality of an insulation object included in electric equipment without depending on installation environment of an acoustic sensor in the electric equipment.SOLUTION: An insulation abnormality diagnostic device comprises: a reference value calculation part 15 which calculates a reference value of intensity based on an intensity value representing the intensity of a signal for every frequency to be obtained by performing filter processing and envelope detection processing to an output signal of an acoustic sensor 10; and an index value calculation part 17 which calculates an index value for determining presence/absence of an insulation abnormality by calculating the intensity to the reference value for a signal of a frequency component which is n-times of a power frequency, and calculates a properly corrected index value so as to suppress an influence of variation of sound collection situations by installation environment of the acoustic sensor 10 by calculating the reference value from an acoustic signal actually detected at a site, and calculating the intensity of the signal to the reference value as the index value.

Description

  The present invention relates to an insulation abnormality diagnosis device, an insulation abnormality diagnosis method, and an insulation abnormality diagnosis program, and is particularly suitable for use in an apparatus that diagnoses an insulation abnormality by analyzing an acoustic signal of a discharge sound detected by an acoustic sensor. Is.

  In general, the insulation resistance value of an insulator is as high as that of a new product, and tends to decrease with age. Therefore, measuring the insulation resistance value of an insulator included in an electrical facility is also useful for grasping an appropriate replacement time of the electrical facility, and is often performed in a periodic inspection. However, measurement of the insulation resistance value can usually be performed only after the electricity is stopped, and since the measurement instrument is brought into contact with the electric circuit, the equipment operation rate is reduced due to a power failure. There is a risk of electric shocks due to incorrect operation.

  Therefore, it is desired from the viewpoints of improving the operating rate of electrical equipment and improving the workability of inspection work so that the insulation resistance value of the insulator can be measured in an energized state without contact. 2. Description of the Related Art Conventionally, a technique has been proposed in which a discharge generated when an insulation resistance of an insulator in an electrical facility is deteriorated is regarded as an acoustic signal and an insulation abnormality is diagnosed by performing discharge detection (see, for example, Patent Documents 1 to 4). ).

  The corona discharge detection device described in Patent Document 1 includes a sound detection unit that detects sound generated when corona discharge occurs, and a high-pass signal that extracts high-frequency components from the detection signal of the sound detection unit and removes ambient noise. From the filter, the rectifier and low-pass filter that detect the envelope of the signal after noise removal by rectification and extract the strong and weak components of the corona discharge sound, and the content of the frequency component twice the power frequency in the strong and weak components And a central processing unit for determining the sound generation site and its appearance.

  In the insulation deterioration diagnosis device described in Patent Document 2, air vibration due to partial discharge sound caused by insulation deterioration of AC electrical equipment is detected using an acoustic-light conversion sensor, and ultrasonic waves that are characteristic of discharge sound are detected by a high-pass filter. After extracting the components, a Fourier series having a frequency twice the power supply frequency is obtained by envelope detection. Then, based on the relationship between the frequency component value twice the power supply frequency in the discharge sound and the discharge charge amount obtained and stored in advance, the amount of discharge charge is converted into a corresponding discharge charge amount to determine the insulation deterioration state.

  In the partial discharge detector described in Patent Document 3, an ultrasonic sensor (microphone) that collects ultrasonic waves, a bandpass filter, a detection circuit, and a filter circuit that passes only twice the frequency of the commercial power supply frequency, By removing the frequency other than the specific frequency, only the information on the vibration wave caused by the partial discharge is displayed on the display device so that the discharge can be easily determined.

  In the insulation abnormality diagnosis device described in Patent Document 4, ultrasonic waves associated with partial discharges generated in the power receiving panel are detected by an acoustic sensor, and only a specific frequency component of the output signal of the acoustic sensor is extracted through a filter circuit. To do. The extracted signal is subjected to envelope detection by a detection circuit, a signal having a frequency component twice the power supply frequency is extracted, and an insulation abnormality of the power receiving panel is diagnosed based on the extracted signal.

  As described in these Patent Documents 1 to 4, pulse components in a specific frequency region are separated by processing a high-pass filter or a band-pass filter on an acoustic signal collected by an acoustic sensor such as a microphone. Can be extracted. Furthermore, by performing envelope detection using FFT, etc., and extracting signals with frequency components that are twice the power supply frequency, noise from non-discharge pulses due to environmental sounds such as insect wing sounds and rain dripping sounds can be detected. As can be removed.

  On the other hand, the discharge sound intensity in a specific frequency range obtained by fast Fourier transform of the discharge sound detected from the insulator of the electric device, the discharge sound intensity of the insulator of the electric device in the energized state, and the surface of the insulator A technique has also been proposed in which the surface insulation resistance value of an insulator is obtained based on a relational expression stored in advance with respect to the relation with the insulation resistance value (see, for example, Patent Document 5). According to the technique described in Patent Document 5, it is possible to measure the insulation resistance value in a non-contact state while energizing the electric device.

JP-A-9-233679 JP 2000-137053 A JP 2001-305178 A JP 2005-147890 A JP 2012-168158 A

  However, none of these techniques described in Patent Documents 1 to 5 considers the influence of an environment in which an acoustic sensor such as a microphone is installed in electrical equipment on detection of discharge sound. There has been a problem that an insulation abnormality cannot be accurately diagnosed by detection.

  That is, while the acoustic sensor is installed at a fixed position of the electrical equipment, it is uncertain where the insulation abnormality occurs in the insulator included in the electrical equipment. Therefore, the distance between the location where the discharge sound is generated due to the insulation abnormality and the acoustic sensor differs depending on the situation, and the intensity of the discharge sound detected by the acoustic sensor is affected by the difference in the distance.

  In addition, in electrical equipment where wiring and parts are densely installed in a narrow space, such as a switchboard, for example, sound is transmitted between the location where the discharge noise is generated due to insulation abnormality and the acoustic sensor. There may be an interfering object that disturbs. Also in this case, the intensity of the discharge sound detected by the acoustic sensor is affected by the presence or absence of the interference.

  In other words, the index value obtained by processing the acoustic signal collected by an acoustic sensor such as a microphone is the distance between the location where the discharge sound is generated and the acoustic sensor, and the presence or absence of an interfering object between the acoustic sensor. The index value may change even in the same physical state (discharge state or insulation resistance value) due to the influence of the installation environment of the acoustic sensor. However, the technologies described in Patent Documents 1 to 5 do not take into account such an influence of the installation environment, and an accurate index value cannot be obtained. As a result, there has been a problem that an insulation abnormality of an insulator that may occur in an actual site cannot be accurately diagnosed.

  The present invention has been made to solve such a problem, and more accurately diagnoses an insulation abnormality of an insulator included in an electrical facility regardless of the installation environment of the acoustic sensor in the electrical facility. The purpose is to be able to.

  In order to solve the above-described problem, in the present invention, based on an intensity value representing the intensity of a signal for each frequency obtained by performing filter processing and envelope detection processing on an output signal of an acoustic sensor, By calculating the reference value and calculating the strength of the frequency component signal n times the power frequency (n is one or more integers greater than or equal to 1) relative to the reference value, an insulation abnormality of the insulator has occurred. An index value for determining whether or not there is is obtained.

  According to the present invention configured as described above, the intensity of the signal of the frequency component n times the power supply frequency is not used as an index value as it is, but the acoustic signal actually detected at the site where the acoustic sensor is installed is used. A reference value is calculated based on the signal strength for each frequency obtained by processing, and the signal strength with respect to the reference value is calculated as an index value. Therefore, the calculated index value is a value that is appropriately corrected so as to suppress the influence of the variation in the sound collection state due to the installation environment of the acoustic sensor in the electrical equipment. Thus, by using this index value to determine whether or not an insulation abnormality of the insulator has occurred, the insulation abnormality of the insulator contained in the electrical equipment can be determined regardless of the installation environment of the acoustic sensor in the electrical equipment. Diagnosis can be made more accurately.

It is a block diagram which shows the structural example of the diagnostic system provided with the insulation abnormality diagnostic apparatus by 1st Embodiment. It is a figure which shows the process example of a high pass filter. It is a figure which shows the envelope obtained when envelope detection is performed with respect to the discharge pulse. It is a figure which shows the envelope obtained when an envelope detection is performed with respect to a non-discharge pulse. It is a figure for demonstrating the process which calculates | requires the median value by 1st Embodiment. It is a figure which shows the reference value calculated for every frequency from the output signal of an FFT process part in 1st Embodiment. It is a figure which shows the envelope obtained as a result of subtracting a reference value from the intensity value of the signal for every frequency. It is a figure which shows the example of the index value computed from the various acoustic signals detected by the acoustic sensor. It is a block diagram which shows the structural example of the diagnostic system provided with the insulation abnormality diagnostic apparatus by 2nd Embodiment. It is a figure which shows the process example of the average calculation part in a flame | frame by a 2nd Embodiment, and an average calculation part between frames. It is a block diagram which shows the structural example of the diagnostic system provided with the insulation abnormality diagnostic apparatus by 3rd Embodiment. It is a block diagram which shows the structural example of the diagnostic system provided with the insulation abnormality diagnostic apparatus by 4th Embodiment.

(First embodiment)
DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, a first embodiment of the invention will be described with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration example of a diagnosis system including an insulation abnormality diagnosis apparatus according to the first embodiment. As shown in FIG. 1, the diagnosis system according to the first embodiment includes an acoustic sensor 10, an insulation abnormality diagnosis device 20, and a communication device 30. The acoustic sensor 10, the insulation abnormality diagnosis device 20, and the communication device 30 are all installed in an electrical facility to be diagnosed.

  The acoustic sensor 10 detects a discharge sound generated when a discharge is generated due to an insulation abnormality of an insulator included in an electrical facility, and is configured by a microphone, for example. The insulation abnormality diagnosis apparatus 20 obtains an index value for determining whether or not an insulation abnormality of the insulator has occurred by processing the acoustic signal detected by the acoustic sensor 10. The communication device 30 transmits the index value calculated by the insulation abnormality diagnosis device 20 to a computer installed in a security center or the like.

  In a computer installed in a security center or the like, based on an index value transmitted from the insulation abnormality diagnosis device 20 via the communication device 30, whether or not insulation abnormality has occurred in the insulator included in the electrical equipment is determined. Diagnose. For example, when the index value is greater than a predetermined threshold, it is determined that there is a possibility that an abnormality in insulation has occurred, and an alarm is issued. Alternatively, it may be determined that there is a possibility that an insulation abnormality has occurred when a state where the index value is greater than a predetermined threshold continues for a predetermined time or longer.

  In the first embodiment, detection of an acoustic signal by the acoustic sensor 10 is always performed. On the other hand, the calculation of the index value by the insulation abnormality diagnosis device 20 is performed at predetermined time intervals. Alternatively, the computer is configured such that an index value calculation command can be notified from a computer such as a security center to the insulation abnormality diagnosis device 20 via the communication device 30, and the index is displayed when the calculation command is received by the insulation abnormality diagnosis device 20. A value may be calculated.

  As shown in FIG. 1, the insulation abnormality diagnosis apparatus 20 has, as its functional configuration, a high-pass filter 11, an absolute value processing unit 12, a low-pass filter 13, an FFT processing unit 14, a reference value calculation unit 15, and a specific frequency component extraction unit 16. And an index value calculation unit 17. These functional blocks 11 to 17 can be configured by any of hardware, DSP (Digital Signal Processor), software, or a combination thereof. In the first embodiment, the acoustic signal detected by the acoustic sensor 10 is A / D converted at a sampling frequency of 96 kHz, for example. The insulation abnormality diagnosis device 20 performs digital signal processing on an acoustic signal digitized by A / D conversion.

  The high-pass filter 11 corresponds to a “filter unit” in the claims, and performs a filtering process on the output signal of the acoustic sensor 10 to pass a component having a predetermined frequency or higher. For example, the high-pass filter 11 removes a low-frequency noise component by performing a process of passing a frequency component of 5 kHz or more on the output signal of the acoustic sensor 10. Although the high-pass filter 11 is used here, a band-pass filter may be used.

  FIG. 2 is a diagram illustrating a processing example of the high-pass filter 11. FIG. 2A shows a waveform of an acoustic signal detected by the acoustic sensor 10. FIG. 2B shows the waveform of the output signal of the high-pass filter 11. As shown in FIG. 2, it is possible to separate and extract a pulse component having a predetermined frequency or more from an acoustic signal by performing a high-pass filter process on the acoustic signal.

  The absolute value processing unit 12 converts the output signal of the high pass filter 11 into an absolute value. The low-pass filter 13 performs a low-pass filter process on the output signal of the absolute value processing unit 12 to pass a component having a predetermined frequency or less (for example, 400 Hz or less). The FFT processing unit 14 converts the time-axis signal into a frequency-axis signal by performing FFT processing on the output signal of the low-pass filter 13.

  The absolute value processing unit 12, the low-pass filter 13, and the FFT processing unit 14 correspond to an “envelope detection unit” in the claims. That is, the intensity of the signal for each frequency is detected by performing envelope detection on the output signal of the high-pass filter 11 by the processing of the absolute value processing unit 12, the low-pass filter 13, and the FFT processing unit 14. 3 and 4 are diagrams illustrating a processing example of the envelope detection unit.

  Among these, FIG. 3 shows an envelope obtained when envelope detection is performed on the discharge pulse extracted by the high-pass filter 11 from the acoustic signal whose discharge sound is detected by the acoustic sensor 10. 3A shows the output signal of the high-pass filter 11, and FIG. 3B shows the output signal of the FFT processing unit 14. As can be seen from FIG. 3B, when the discharge pulse is detected by envelope detection, a characteristic peak is generated at a frequency n times (n = 1, 2, 3,...) Of the power supply frequency (50 Hz in FIG. 3). To do.

  On the other hand, FIG. 4 shows an envelope obtained when envelope detection is performed on the non-discharge pulse extracted by the high-pass filter 11 from the acoustic signal in which the insect feather sound is detected by the acoustic sensor 10. 4A shows the output signal of the high-pass filter 11, and FIG. 4B shows the output signal of the FFT processing unit 14. As can be seen from FIG. 4 (b), when a non-discharge pulse due to insect wing noise is detected by envelope detection, no characteristic peak occurs at any frequency.

  The reference value calculation unit 15 calculates a reference value of intensity based on an intensity value representing the intensity of the signal for each frequency (output signal of the FFT processing unit 14) detected by the envelope detection unit. Specifically, the reference value calculation unit 15 sets a moving window of a predetermined size for each frequency signal calculated by the FFT processing unit 14, and signals of a predetermined number of frequency components included in the moving window. A representative value for the strength of the is calculated. Then, the calculated representative value is set as a reference value. The representative value may be any one of an average value, a minimum value, a median value, a maximum value, and the like, but it is preferable to use a median value.

FIG. 5 is a diagram for explaining processing for obtaining a median value related to the intensity of signals of a predetermined number of frequency components included in the moving window. FIGS. 5A and 5B show signal intensity values for each frequency calculated by the FFT processing unit 14. Here, it is shown that the signal intensity values for the frequencies f _{1 to} f ₈ are −62, −63, −70, −63, −67, −69, −64, and −66 [dB], respectively. .

  A moving window having a size including five frequencies is sequentially set for the output signal of the FFT processing unit 14, and a median value regarding the intensity of the signal of the five frequency components included in each moving window is calculated. The result is shown in FIG. When the median value is calculated as the representative value, the moving window is sized so that the number of frequencies included in the moving window is an odd number.

  The intensity values of the frequency component signals included in the first moving window are five values of -62, -63, -70, -63, and -67 [dB]. Therefore, the median value in this case is −63 [dB]. The intensity values of the frequency component signals included in the second moving window are five values of −63, −70, −63, −67, and −69 [dB]. Therefore, the median value in this case is −67 [dB]. Similarly, the median value is obtained for the third and subsequent moving windows.

  FIG. 6 is a diagram illustrating a reference value (median value) calculated for each frequency from the output signal of the FFT processing unit 14 illustrated in FIG. In FIG. 6, an envelope indicated by reference numeral 61 is a signal strength value for each frequency (output signal of the FFT processing unit 14), and an envelope indicated by reference numeral 62 is a reference for each frequency obtained from the output signal of the FFT processing unit 14. Value.

  The specific frequency component extraction unit 16 extracts a frequency component n times the power supply frequency (n is an integer of 1 or more) from the output signal of the FFT processing unit 14. In the first embodiment, n = 2 and a frequency component that is twice the power supply frequency (100 Hz when the power supply frequency is 50 Hz) is extracted. The reason why the frequency component of 100 Hz is extracted is that the peak component is larger than the other frequency components other than n = 2 and the characteristic of the discharge pulse appears most strongly.

The index value calculation unit 17 uses the strength of the reference value calculated by the reference value calculation unit 15 for the signal of the frequency component of 100 Hz extracted by the specific frequency component extraction unit 16 to determine whether an insulation abnormality of the insulator has occurred. It is calculated as an index value for determining whether or not. For example, in the example illustrated in FIG. 5, if the frequency f ₄ is 100 Hz, the reference value calculation unit 15 calculates the reference value from the intensity value (−63 [dB]) of the signal of the frequency f ₄ . By subtracting the median value (−67 [dB]), a value of 4 [dB] is set as an index value.

  FIG. 7 shows an envelope obtained as a result of subtracting the reference value for each frequency indicated by reference numeral 62 from the intensity value of the signal for each frequency indicated by reference numeral 61 in FIG. However, if the result of subtraction is a negative value, the value is zero. In the envelope shown in FIG. 7, the intensity value at a frequency of 100 Hz is an index value calculated by the index value calculation unit 17. Here, the reference value for each frequency is calculated by the reference value calculation unit 15, but the reference value used by the index value calculation unit 17 is only the reference value of 100 Hz. Therefore, the reference value calculation unit 15 may calculate the reference value only for a signal having a frequency component n times the power supply frequency extracted by the specific frequency component extraction unit 16.

  FIG. 8 is a diagram illustrating an example of index values calculated for each sound by detecting various sounds by the acoustic sensor 10. In FIG. 8, sample data (1-1) to (1-5) are acoustic signals obtained when a discharge sound is detected by the acoustic sensor 10. The sample data (2-1) to (2-5) are acoustic signals obtained when the acoustic sensor 10 detects non-discharge sounds such as shutter opening / closing sounds, insect feather sounds, and rain dripping sounds. is there.

  The example shows an index value of 100 Hz calculated by the first embodiment from these ten kinds of acoustic signals. That is, the numerical values are shown when the index value is calculated by correcting the signal intensity value of 100 Hz based on the reference value calculated by the reference value calculation unit 15. On the other hand, the comparative example shows an index value of 100 Hz calculated by the conventional method from the same 10 kinds of acoustic signals. That is, the numerical value is shown when the 100 Hz signal intensity value calculated by the FFT processing unit 14 is used as an index value as it is.

  In FIG. 8, the index value calculated from the sample data (1-1) to (1-5) regarding the discharge sound and the index calculated from the sample data (2-1) to (2-5) regarding the non-discharge sound. The distance between sets with values is also shown. The inter-set distance is a distance between a set formed by a plurality of index values obtained from the discharge sound and a set formed by a plurality of index values obtained from the non-discharge sound. Here, the distance between sets is calculated by the Mahalanobis distance.

  As can be seen from FIG. 8, the inter-set distance related to the index value calculated by applying the first embodiment is 6.8, and the inter-set distance related to the index value calculated by the conventional method is 1.4. The value of the distance between sets is larger when the first embodiment is applied, which shows that it is advantageous for the detection of discharge sound.

  As described above in detail, in the first embodiment, based on the intensity value representing the intensity of the signal for each frequency obtained by performing the filtering process and the envelope detection process on the output signal of the acoustic sensor 10, An index value for determining whether or not an insulation abnormality of the insulator has occurred by calculating a strength reference value and calculating a strength with respect to the reference value for a signal having a frequency component n times the power frequency. I want to ask.

  According to the first embodiment configured as described above, the reference value is calculated based on the intensity of the signal for each frequency obtained by processing the acoustic signal actually detected at the site where the acoustic sensor 10 is installed. The signal strength with respect to the reference value is calculated as an index value. Therefore, the calculated index value is a value that is appropriately corrected so as to suppress the influence of the variation in the sound collection situation due to the installation environment of the acoustic sensor 10 in the electrical equipment. Thus, by determining whether or not an insulation abnormality of the insulator has occurred using this index value, the insulation abnormality of the insulator included in the electrical equipment is determined regardless of the installation environment of the acoustic sensor 10 in the electrical equipment. Can be diagnosed more accurately.

  Also, as shown in FIG. 8, according to the first embodiment, the distance between the sets of the index value obtained from the discharge sound and the index value obtained from the non-discharge sound can be increased. Thereby, it can be made easy to determine whether or not the discharge sound is generated based on the index value. That is, since it is possible to clearly distinguish whether or not the index value calculated by the index value calculation unit 17 indicates a discharge sound by a predetermined threshold value, erroneous determination can be reduced.

  In the first embodiment, the index value is calculated by the insulation abnormality diagnosis device 20 and transmitted to a computer such as a security center by the communication device 30. Therefore, the presence or absence of insulation abnormality of the insulator can be diagnosed using the index value sent from the site without actually going to the site where the electrical equipment is installed and collecting the acoustic signal.

(Second Embodiment)
Next, a second embodiment of the present invention will be described with reference to the drawings. FIG. 9 is a block diagram illustrating a configuration example of a diagnosis system including an insulation abnormality diagnosis device according to the second embodiment. In FIG. 9, components having the same reference numerals as those shown in FIG. 1 have the same functions, and thus redundant description is omitted here.

  As shown in FIG. 9, the insulation abnormality diagnosis device 20 according to the second embodiment includes a second reference value calculation unit 15A instead of the reference value calculation unit 15 shown in FIG. The second reference value calculation unit 15A includes an intra-frame average calculation unit 21 and an inter-frame average calculation unit 22 as its functional configuration. Further, the insulation abnormality diagnosis device 20 according to the second embodiment includes an index value calculation unit 17A instead of the index value calculation unit 17.

  The intra-frame average calculation unit 21 calculates an average value of signal strength (hereinafter referred to as an intra-frame average value) for each frame divided in predetermined time units for the output signal of the absolute value processing unit 12. The inter-frame average calculation unit 22 calculates an average value of intra-frame average values (hereinafter referred to as an inter-frame average value) calculated for m frames (m is an arbitrary integer of 2 or more) by the intra-frame average calculation unit 21 calculate.

  FIG. 10 is a diagram illustrating a processing example of the intra-frame average calculation unit 21 and the inter-frame average calculation unit 22. The waveform shown in FIG. 10 shows the output signal of the absolute value processing unit 12. The intra-frame average calculation unit 21 divides the output signal at predetermined time intervals and sets a frame as a processing unit. Each time unit indicated by m = 0, 1, 2,... is a frame. One frame includes N sampling values indicated by t = 0 to N−1. The intra-frame average calculating unit 21 calculates an average value of the N sampling values for each frame.

  The inter-frame average calculation unit 22 calculates an average value of the m intra-frame average values calculated for each frame by the intra-frame average calculation unit 21. In the second embodiment, the inter-frame average value calculated by the inter-frame average calculation unit 22 is used as a reference value for a signal having a frequency component that is n times the power supply frequency.

  The index value calculation unit 17A calculates an index value by calculating the strength with respect to the reference value calculated by the interframe average calculation unit 22 for a signal having a frequency component n times the power supply frequency. Specifically, the index value is obtained by subtracting the reference value calculated by the second reference value calculation unit 15A from the intensity value of the 100 Hz signal extracted by the specific frequency component extraction unit 16.

  Note that the reference value calculation (update) timing by the second reference value calculation unit 15A can be arbitrarily set. For example, A) every frame, B) about once a day, C) periodic inspection of the electrical equipment, D) installation of the insulation abnormality diagnosis device 20 in the electrical equipment, etc. are conceivable. Or, it is configured such that a reference value calculation command can be notified from a computer such as a security center to the insulation abnormality diagnosis device 20 via the communication device 30, and the insulation abnormality diagnosis device 20 receives this calculation instruction. Alternatively, the reference value may be calculated.

  However, since it is not preferable to update the reference value too frequently, it is preferable to set the timing of B) to E). When the reference value is calculated at the timings B) to E), a storage unit is provided after the inter-frame average calculation unit 22, and the reference value is stored in the storage unit until the next update. In the case of D), the reference value is calculated only once and is not updated. The index value calculation unit 17A calculates the index value using the reference value stored in the storage unit. The index value calculation timing by the index value calculation unit 17A may be matched with the reference value calculation timing (except for the cases of C and D) by the second reference value calculation unit 15A.

  Also in the second embodiment configured as described above, the reference value is calculated based on the intensity of the signal obtained by processing the acoustic signal actually detected at the site where the acoustic sensor 10 is installed. Is calculated as an index value. Thus, by determining whether or not an insulation abnormality of the insulator has occurred using this index value, the insulation abnormality of the insulator included in the electrical equipment is determined regardless of the installation environment of the acoustic sensor 10 in the electrical equipment. Can be diagnosed more accurately.

(Third embodiment)
Next, a third embodiment of the present invention will be described with reference to the drawings. FIG. 11 is a block diagram illustrating a configuration example of a diagnosis system including an insulation abnormality diagnosis device according to the third embodiment. Note that in FIG. 11, those given the same reference numerals as those shown in FIG. 1 have the same functions, and therefore redundant description is omitted here.

  As shown in FIG. 11, the insulation abnormality diagnosis device 20 according to the third embodiment includes a third reference value calculation unit 15B instead of the reference value calculation unit 15 shown in FIG. The second reference value calculation unit 15B includes a DC component extraction unit 31 and an average calculation unit 32 as functional configurations. In addition, the insulation abnormality diagnosis device 20 according to the third embodiment includes an index value calculation unit 17B instead of the index value calculation unit 17.

  The DC component extraction unit 31 extracts a DC component for each frame divided by a predetermined time unit from the output signal of the FFT processing unit 14. The frame is the same as the frame described in the second embodiment. The DC component extracted for each frame is an intensity value of a frequency component of 0 Hz obtained when the output signal of the high-pass filter 11 is divided in units of frames and a signal in each frame is subjected to FFT processing.

  The average calculation unit 32 calculates the average value of the DC components extracted by the DC component extraction unit 31 for m frames (m is an arbitrary integer equal to or greater than 2). In the third embodiment, the average value calculated by the average calculation unit 32 is set as a reference value for a signal having a frequency component n times the power supply frequency.

  The index value calculation unit 17B obtains an index value by calculating the strength with respect to the reference value calculated by the average calculation unit 32 for a signal having a frequency component of n times the power supply frequency. Specifically, the index value is obtained by subtracting the reference value calculated by the third reference value calculation unit 15B from the intensity value of the 100 Hz signal extracted by the specific frequency component extraction unit 16.

  Also in the third embodiment configured as described above, the reference value is calculated based on the intensity of the DC component for each frame obtained by processing the acoustic signal actually detected at the site where the acoustic sensor 10 is installed. The signal strength with respect to the reference value is calculated as an index value. Thus, by determining whether or not an insulation abnormality of the insulator has occurred using this index value, the insulation abnormality of the insulator included in the electrical equipment is determined regardless of the installation environment of the acoustic sensor 10 in the electrical equipment. Can be diagnosed more accurately.

  The reference value calculation (update) timing by the third reference value calculation unit 15B can also be set arbitrarily. For example, A) every frame, B) about once a day, C) during periodic inspection of electrical equipment, D) when installing insulation abnormality diagnosis device 20 in electrical equipment, and E) insulation abnormality diagnosis device 20 at security center The reference value can be calculated when a reference value calculation command is received from the computer. When the reference value is calculated at the timings B) to E), a storage unit is provided after the average calculation unit 32. The index value calculation unit 17B calculates the index value using the reference value stored in the storage unit. You may make it match | combine the calculation timing of the index value by this index value calculation part 17B with the calculation timing (except the case of C and D) of the reference value by the 3rd reference value calculation part 15B.

(Fourth embodiment)
Next, a fourth embodiment of the present invention will be described with reference to the drawings. FIG. 12 is a block diagram illustrating a configuration example of a diagnosis system including an insulation abnormality diagnosis device according to the fourth embodiment. In FIG. 12, components having the same reference numerals as those shown in FIG. 1 have the same functions, and thus redundant description is omitted here.

  As shown in FIG. 12, the insulation abnormality diagnosis device 20 according to the second embodiment includes an index value calibration unit 18 instead of the reference value calculation unit 15 shown in FIG. The index value calibration unit 18 includes a calibration value calculation unit 41, a calibration value storage unit 42, and a calibration unit 43 as its functional configuration. Further, the insulation abnormality diagnosis apparatus 20 according to the third embodiment includes an index value calculation unit 17C instead of the index value calculation unit 17.

  The index value calculation unit 17C calculates an index value for each frame divided in predetermined time units. The frame is the same as the frame described in the second embodiment. The index value calculated for each frame is the intensity of a signal having a frequency component of 100 Hz obtained when the output signal of the high-pass filter 11 is divided in units of frames and the signal in each frame is FFT processed by the FFT processing unit 14. Value.

  The calibration value calculation unit 41 calculates an average value of the index values calculated for m frames (m is an arbitrary integer equal to or greater than 2) by the index value calculation unit 17C as a calibration value for the index value. For example, the process of the calibration value calculation unit 41 is performed in a state where it is confirmed that the electrical equipment is normal (when the electrical equipment is newly installed or immediately after the electrical equipment is inspected and maintained). . Then, the calibration value calculated by the calibration value calculation unit 41 is stored in the calibration value storage unit 42.

  The calibration unit 43 stores the calibration value in the calibration value storage unit 42 and then calibrates the index value calculated by the index value calculation unit 17C with the calibration value stored in the calibration value storage unit 42. That is, the calibration unit 43 receives a command for calculating an index value every predetermined time interval (for example, every frame interval or about once a day) or from a computer such as a security center via the communication device 30. The index value calculated by the index value calculation unit 17C is calibrated with the calibration value stored in the calibration value storage unit 42. Specifically, the final index value is obtained by subtracting the calibration value stored in the calibration value storage unit 42 from the intensity value of the 100 Hz signal calculated by the index value calculation unit 17C.

  According to the fourth embodiment configured as described above, the calibration value is calculated based on the index value calculated when the electrical equipment is in a normal state, and the intensity value of the signal calibrated by the calibration value. Is calculated as the final index value. Thus, by determining whether or not an insulation abnormality of the insulator has occurred using this index value, the insulation abnormality of the insulator included in the electrical equipment is determined regardless of the installation environment of the acoustic sensor 10 in the electrical equipment. Can be diagnosed more accurately.

  Although an example in which the index value calibration unit 18 is provided instead of the reference value calculation unit 15 has been described here, the index value calibration unit 18 may be provided in addition to the reference value calculation unit 15. In this case, the index value calculated for each frame divided by the index value calculation unit 17C by a predetermined time unit is the output signal of the high-pass filter 11 divided by the frame unit, and the signal in each frame is FFT processed by the FFT processing unit 14. This is a value corrected based on the reference value calculated by the reference value calculation unit 15 with respect to the intensity value of the signal of the frequency component of 100 Hz obtained at the time of processing.

  In addition to the second reference value calculation unit 15A described in the second embodiment or the third reference value calculation unit 15B described in the third embodiment, an index value calibration unit 18 may be provided. . In this case, the index value calculated for each frame divided by the index value calculation unit 17C by a predetermined time unit is the output signal of the high-pass filter 11 divided by the frame unit, and the signal in each frame is FFT processed by the FFT processing unit 14. It is a value corrected based on the reference value calculated by the second reference value calculation unit 15A or the third reference value calculation unit 15B with respect to the intensity value of the signal of the frequency component of 100 Hz obtained when processing is performed. .

  In the first to fourth embodiments, the index value calculation units 17, 17 </ b> A, 17 </ b> B, and 17 </ b> C are n times the power supply frequency among the output signals of the FFT processing unit 14 (n is an integer of 1 or more). For example, the example in which the index value is calculated for the signal of the frequency component of n = 2) has been described, but the present invention is not limited to this. For example, a plurality of index values are calculated for a signal having a frequency component n times (n is an integer of 1 or more) of the power supply frequency, and the final index value is obtained by combining the plurality of index values. It may be.

  For example, assuming n = 1, 2, 4, index values are calculated for signals of frequency components of 50 Hz, 100 Hz, and 200 Hz, respectively, and one index value is finally calculated by adding three index values. By doing in this way, an index value strong against noise can be obtained. That is, even if noise is mixed in the frequency component of 100 Hz, since the index values calculated for the frequencies of 50 Hz and 200 Hz are added, the ratio of noise is relatively reduced to minimize the influence. Can do.

  In the first to fourth embodiments, the index value calculated by the insulation abnormality diagnosis device 20 is transmitted to a computer such as a security center, and the computer diagnoses whether or not an insulation abnormality has occurred in the computer. Although described, the present invention is not limited to this. For example, the insulation abnormality diagnosis device 20 may perform diagnosis up to insulation abnormality by threshold determination based on the index value, and transmit the diagnosis result to a computer such as a security center.

  Moreover, although the said embodiment demonstrated the example which determines the presence or absence of an insulation abnormality based on the index value itself calculated by the insulation abnormality diagnostic apparatus 20, this invention is not limited to this. For example, the insulation resistance value is calculated on the basis of the index value calculated by the insulation abnormality diagnosis device 20 and a relational expression or a reference table determined in advance for the relationship between the index value and the insulation resistance value of the insulator. The presence or absence of insulation abnormality may be determined based on the value.

  When calculating the insulation resistance value, the index value is calculated in parallel by any of the plurality of methods in the first to fourth embodiments, and further, based on the above-described relational expression or reference table from those index values. The insulation resistance value may be calculated respectively. Of the plurality of insulation resistance values calculated in this way, for example, based on the smallest insulation resistance value (that is, a value indicating the highest possibility that an insulation abnormality has occurred), The presence or absence may be determined. Alternatively, an average value of a plurality of insulation resistance values may be calculated, and the presence / absence of an insulation abnormality may be determined based on the average value.

  In addition, when calculating the insulation resistance value, if the insulation resistance value is monitored over time and it is detected that the insulation resistance value is decreasing, an insulation abnormality may occur in the future. It may be possible to make a predictive diagnosis.

  In addition, the above-described first to fourth embodiments are merely examples of specific embodiments for carrying out the present invention, and the technical scope of the present invention should not be construed as being limited thereto. It will not be. That is, the present invention can be implemented in various forms without departing from the gist or the main features thereof.

10 Acoustic sensor 11 High pass filter (filter part)
12 Absolute value processor (envelope detector)
13 Low-pass filter (envelope detector)
14 FFT processor (envelope detector)
15, 15A, 15B Reference value calculation unit 16 Specific frequency component extraction unit 17, 17A, 17B, 17C Index value calculation unit 18 Index value calibration unit 20 Insulation abnormality diagnosis device 21 In-frame average calculation unit 22 Inter-frame average calculation unit 30 Communication Device 31 DC component extraction unit 32 Average calculation unit 41 Calibration value calculation unit 42 Calibration value storage unit 43 Calibration unit

Claims (8)

A filter unit that passes a component of a predetermined frequency or higher by performing a filtering process on an output signal of an acoustic sensor that detects a discharge sound generated when a discharge occurs due to an insulation abnormality of an insulator;
By performing envelope detection on the output signal of the filter unit, an envelope detection unit that detects the intensity of the signal for each frequency, and
A reference value calculation unit for calculating a reference value of intensity based on an intensity value representing the intensity of the signal for each frequency detected by the envelope detection unit;
Among the output signals of the envelope detector, the intensity of the frequency component that is n times the power frequency (n is one or more integers equal to or greater than 1) with respect to the reference value calculated by the reference value calculator An insulation abnormality diagnosis apparatus comprising: an index value calculation unit that obtains an index value for determining whether or not an insulation abnormality of the insulator has occurred by calculating   The reference value calculation unit sets a moving window of a predetermined size for the signal for each frequency detected by the envelope detection unit, and a representative value related to the intensity of a signal of a predetermined number of frequency components included in the moving window The insulation abnormality diagnosis apparatus according to claim 1, wherein the representative value is set as the reference value by calculating.   3. The insulation abnormality diagnosis apparatus according to claim 2, wherein the representative value is a median value.   4. The reference value calculation unit according to claim 1, wherein the reference value calculation unit calculates the reference value only for a signal having a frequency component that is n times the power supply frequency (n is one or more integers equal to or greater than 1). The insulation abnormality diagnostic device according to any one of the above. A second reference value calculation unit is provided instead of the reference value calculation unit, and further includes an absolute value processing unit that converts the output signal of the filter unit into an absolute value, and the second reference value calculation unit includes:
For the output signal of the absolute value processing unit, an intra-frame average calculation unit that calculates an intra-frame average value of the signal strength for each frame divided in predetermined time units,
The intra-frame average value is calculated by further averaging the intra-frame average value calculated for m frames (m is an arbitrary integer equal to or greater than 2) by the intra-frame average calculation unit. And an inter-frame average calculation unit that serves as a reference value for a signal having a frequency component that is n times the power supply frequency,
The index value calculation unit calculates, as the index value, an intensity with respect to the reference value calculated by the second reference value calculation unit for a signal having a frequency component n times the power supply frequency. The insulation abnormality diagnosis device according to Item 1. A third reference value calculation unit is provided instead of the reference value calculation unit, and the third reference value calculation unit includes:
For the output signal of the envelope detection unit, a DC component extraction unit that extracts a DC component for each frame divided in predetermined time units;
An average value of DC components extracted for m frames (m is an arbitrary integer equal to or greater than 2) by the DC component extraction unit is calculated, and the average value is calculated for a signal having a frequency component n times the power supply frequency. An average calculation unit as a reference value,
The index value calculation unit calculates, as the index value, an intensity with respect to the reference value calculated by the third reference value calculation unit for a signal having a frequency component of n times the power supply frequency. The insulation abnormality diagnosis device according to Item 1. In place of or in addition to the reference value calculation unit, the second reference value calculation unit, or the third reference value calculation unit, an index value calibration unit is provided,
The index value calculation unit calculates the index value for each frame divided by a predetermined time unit,
The index value calibration unit calculates a calibration value for calculating an average value of index values calculated for m frames (m is an arbitrary integer equal to or greater than 2) by the index value calculation unit as a calibration value for the index value. And
The calibration unit that calibrates the index value calculated by the index value calculation unit with the calibration value calculated by the calibration value calculation unit. Insulation abnormality diagnosis device.   The index value calculation unit calculates a plurality of index values for a signal having a frequency component of n times (n is an integer of 1 or more) of the power supply frequency among the output signals of the envelope detection unit. The insulation abnormality diagnosis apparatus according to claim 1, wherein a final one index value is obtained by combining the index values.

Images