Partial discharge inspection support method and partial discharge inspection support device

Abstract

To provide a partial discharge inspection method and partial discharge inspection device useful to inspect the presence / absence of occurrence of partial discharge.SOLUTION: A partial discharge inspection method comprises: a step of acquiring an electric signal including a plurality of pulse signals measured by a sensor; a step of extracting each one of the pulse signals from the acquired electric signal; a step of acquiring time component information which indicates a time component of each one of the extracted pulse signals, frequency component information which indicates a frequency component, and signal intensity information which indicates signal intensity; and a step of generating a map which is plotted for each pulse signal from which the acquired time component information, frequency component information, and signal intensity information are extracted. The generated map is used for inspecting the presence / absence of occurrence of partial discharge from a power apparatus.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] An embodiment of the present invention is a partial discharge inspection support Methods and partial discharge testing support Regarding the device. [Background technology]

[0002] Electric power facilities are facilities that support social infrastructure, and include electric power devices such as inverter devices and switchgears.

[0003] Electric power equipment has, for example, an insulating portion on its surface, and the insulating portion (its insulating properties) deteriorates over time. The deterioration of the insulating portion can cause dielectric breakdown (failure of the electric power equipment).

[0004] Incidentally, it is known that when insulation deteriorates, partial discharge occurs in electric power equipment in response to the deterioration, and checking for the occurrence of such partial discharge is useful for diagnosing the insulation of electric power equipment (diagnosing the deterioration of insulation).

[0005] When checking for the occurrence of partial discharge, electrical signals generated in response to the operation of power equipment are measured. These electrical signals may contain pulse signals caused by noise (hereinafter referred to as noise signals) in addition to pulse signals generated based on the partial discharge (hereinafter referred to as partial discharge signals).

[0006] This makes it difficult to properly check for the occurrence of partial discharge based on the pulse signal contained in the electrical signal, and there is a need for an insulation diagnosis technique that is useful for such testing. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Patent No. 6702573 Summary of the Invention [Problem to be solved by the invention]

[0008] Therefore, the problem to be solved by the present invention is to provide a partial discharge inspection method useful for inspecting whether or not partial discharge occurs in electric power equipment. support Methods and partial discharge testing support The present invention aims to provide a device. [Means for solving the problem]

[0009] In accordance with an embodiment Partial discharge inspection support device performs Partial Discharge Inspection support The method includes the steps of acquiring an electrical signal including a plurality of pulse signals measured by a sensor attached to electric power equipment, extracting each of the plurality of pulse signals from the acquired electrical signal, acquiring time component information indicating a time component, frequency component information indicating a frequency component, and signal intensity information indicating a signal intensity for each of the extracted plurality of pulse signals, and creating a map in which the acquired time component information, frequency component information, and signal intensity information are plotted for each of the extracted pulse signals, and the created map is used to inspect for the occurrence of partial discharge from the electric power equipment. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 10 is a diagram for explaining a TF map in a comparative example of the embodiment. [Figure 2] FIG. 10 is a diagram for explaining a TF map in a comparative example of the present embodiment. [Figure 3] FIG. 2 is a block diagram showing an example of the functional configuration of the partial discharge inspection device according to the present embodiment. [Figure 4] FIG. 1 is a diagram showing an example of a hardware configuration of a partial discharge inspection device. [Figure 5] 10 is a flowchart showing an example of a processing procedure of a partial discharge inspection device. [Figure 6] FIG. 10 is a diagram showing an example of an electrical signal. [Figure 7] FIG. 10 is a diagram showing an example of a TFV map. [Figure 8]FIG. 10 is a diagram for explaining an example of a process for inspecting whether or not partial discharge occurs in an electric power device. [Figure 9] FIG. 10 is a diagram for explaining an example of a process for inspecting whether or not partial discharge occurs in an electric power device. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments will be described with reference to the drawings. The partial discharge inspection device according to this embodiment is used to inspect whether or not partial discharge occurs in electric power equipment in response to deterioration of an insulating portion of the electric power equipment.

[0012] When checking whether partial discharge has occurred in electric power equipment, the electric signal generated in response to the operation of the electric power equipment is measured. This electric signal includes not only a pulse signal generated based on partial discharge (hereinafter referred to as partial discharge signal) but also a pulse signal caused by noise (hereinafter referred to as noise signal) (i.e., noise is occurring in the electric signal).

[0013] In this case, it is possible to remove noise signals from an electrical signal (pulse signal) by applying a wavelet transform, which can obtain the frequency characteristics of a waveform along the time axis, to the electrical signal. However, in the case of inverter-driven power equipment (such as an inverter panel), it is difficult to remove noise signals because of the large amount of noise caused by switching.

[0014] Therefore, there is a method (hereinafter referred to as a comparative example of this embodiment) in which the occurrence of partial discharge from power equipment such as an inverter panel is inspected using a TF map created by performing time-frequency analysis. Below, a brief description will be given of this comparative example of this embodiment.

[0015] First, the TF map used in the comparative example of this embodiment is created by plotting (mapping) time component information indicating the time components and frequency component information indicating the frequency components of each of a plurality of pulse signals included in an electric signal generated in response to the operation of electric power equipment.

[0016] In the comparative example of this embodiment, when the multiple plots corresponding to the combination of time component information and frequency component information in the TF map created in this way have a discrete nature (i.e., a set of partial discharge signal plots and a set of noise signal plots are discretely arranged), it is possible to estimate that at least the electrical signal used to create the TF map contains both a partial discharge signal and a noise signal, and to determine that a partial discharge has occurred in the electric equipment (i.e., to detect the occurrence of a partial discharge).

[0017] However, in the comparative example of this embodiment described above, there are cases where it is not possible to detect the occurrence of partial discharge from the electric power equipment even if the TF map is referenced.

[0018] Here, Fig. 1 shows an example of a TF map (hereinafter referred to as a first T-F map) created based on electrical signals (multiple pulse signals) generated in response to the operation of electric power equipment that is not experiencing partial discharge, and Fig. 2 shows an example of a TF map (hereinafter referred to as a second T-F map) created based on electrical signals (multiple pulse signals) generated in response to the operation of electric power equipment that is experiencing partial discharge.

[0019] For example, when the time component information and frequency component information of a partial discharge signal and a noise signal differ, the TF map has multiple plots that are discrete as described above, and therefore, by referring to the TF map, it is possible to determine that a partial discharge has occurred in the power equipment. However, when a set of plots of a partial discharge signal and a set of plots of a noise signal overlap in the same area (i.e., the set of plots is not discretely arranged), as in the first T-F map shown in Figure 1 and the second T-F map shown in Figure 2, the same TF map is created regardless of whether a partial discharge has occurred, and therefore, it is not possible to determine that a partial discharge has occurred even by referring to the second T-F map, for example.

[0020] That is, in the comparative example of the present embodiment described above, if the time component information and frequency component information of the partial discharge signal and the noise signal overlap, it may not be possible to properly check whether or not a partial discharge has occurred in the electric power equipment.

[0021] Taking these circumstances into consideration, the partial discharge inspection device according to this embodiment has a configuration for appropriately inspecting whether or not partial discharge is occurring in electric power equipment.

[0022] The partial discharge inspection device according to this embodiment will be described below. Fig. 3 is a block diagram showing an example of the functional configuration of the partial discharge inspection device according to this embodiment.

[0023] Here, when the partial discharge inspection device 10 according to this embodiment is used to inspect whether or not partial discharge occurs in electric power equipment 20, the electric power equipment 20 includes, for example, an inverter panel (inverter device) configured to control the voltage supplied to a motor or the like arranged downstream. In the following, a case will be described assuming that the electric power equipment 20 is an inverter panel, but the electric power equipment 20 may also be equipment (power receiving and distribution equipment such as switchgear) configured, for example, by a circuit breaker, disconnector, rectifier, or transformer. Furthermore, the electric power equipment 20 may be any equipment that may generate partial discharge due to deterioration of its insulation, such as a power transformer, gas-insulated switchgear, generator, or reactor.

[0024] The electric power device 20 (inverter panel) has a structure in which the surface is covered with an insulating part, and the electric power device 20 has a sensor 21 attached thereto.

[0025] 3 shows only one electric power device 20 for convenience, there may be a plurality of electric power devices 20. When there are a plurality of electric power devices 20, the same number of sensors 21 as the plurality of electric power devices 20 may be prepared, or one detachable sensor 21 may be attached to the plurality of electric power devices 20 in sequence.

[0026] 3, the partial discharge inspection device 10 includes an electric signal acquisition unit 11, a pulse signal extraction unit 12, a time component information acquisition unit 13, a frequency component information acquisition unit 14, a signal intensity information acquisition unit 15, a map creation unit 16, and an inspection unit 17. The partial discharge inspection device 10 is connected to a sensor 21 attached to an electric power device 20.

[0027] Here, the sensor 21 is configured to measure an electrical signal generated in response to the operation of the electric power equipment 20 to which the sensor 21 is attached. The electrical signal acquiring unit 11 acquires the electrical signal measured by the sensor 21. The electrical signal acquired by the electrical signal acquiring unit 11 includes a plurality of pulse signals. Note that, when the electric power equipment 20 is generating a partial discharge, the plurality of pulse signals include a pulse signal generated based on the partial discharge (partial discharge signal) and a pulse signal caused by noise different from the partial discharge signal (noise signal). On the other hand, when the electric power equipment 20 is not generating a partial discharge, the plurality of pulse signals include only a noise signal.

[0028] The pulse signal extracting unit 12 extracts each of the plurality of pulse signals from the electrical signal acquired by the electrical signal acquiring unit 11 .

[0029] The time component information acquiring unit 13 acquires time component information indicating the time component of each of the plurality of pulse signals extracted by the pulse signal extracting unit 12 .

[0030] The frequency component information acquisition unit 14 acquires frequency component information indicating the frequency components of each of the plurality of pulse signals extracted by the pulse signal extraction unit 12 .

[0031] The signal strength information acquisition unit 15 acquires signal strength information indicating the signal strength of each of the plurality of pulse signals extracted by the pulse signal extraction unit 12 .

[0032] That is, in this embodiment, time component information, frequency component information, and signal intensity information for each pulse signal are read from the electrical signal.

[0033] The map creation unit 16 creates a map in which the time component information acquired by the time component information acquisition unit 13, the frequency component information acquired by the frequency component information acquisition unit 14, and the signal intensity information acquired by the signal intensity information acquisition unit 15 are plotted for each pulse signal extracted by the pulse signal extraction unit 12.

[0034] In the comparative example of this embodiment described above, a two-dimensional TF map is created by plotting time component information and frequency component information. However, in this embodiment, a three-dimensional map is created in which signal intensity information is plotted in addition to time component information and frequency component information. Furthermore, in this embodiment, "plotting time component information, frequency component information, and signal intensity information" refers to arranging points (plots) on the map that correspond to combinations of the time component information (time components indicated by the time component information), the frequency component information (frequency components indicated by the frequency component information), and the signal intensity information (signal intensity indicated by the signal intensity information). Hereinafter, for convenience, the map created in this embodiment will be referred to as a TFV map.

[0035] The inspection unit 17 inspects whether or not a partial discharge has occurred in the electric power equipment 20, based on the TFV map created by the map creation unit 16. Note that in this embodiment, "inspecting whether or not a partial discharge has occurred in the electric power equipment 20" includes determining whether or not the partial discharge has occurred.

[0036] Fig. 4 shows an example of the hardware configuration of the partial discharge inspection device 10 shown in Fig. 2. The partial discharge inspection device 10 includes a CPU 101, a nonvolatile memory 102, a RAM 103, and a communication device 104. The partial discharge inspection device 10 also includes a bus 105 that interconnects the CPU 101, the nonvolatile memory 102, the RAM 103, and the communication device 104.

[0037] The CPU 101 is a processor for controlling the operation of each component in the partial discharge inspection device 10. The CPU 101 may be a single processor or may be configured with multiple processors. The CPU 101 executes various programs loaded from the nonvolatile memory 102 to the RAM 103. In this embodiment, the programs executed by the CPU 101 include a partial discharge inspection program 103a.

[0038] 3, some or all of the electrical signal acquiring unit 11, pulse signal extracting unit 12, time component information acquiring unit 13, frequency component information acquiring unit 14, signal strength information acquiring unit 15, map creating unit 16, and inspection unit 17 are realized by software, for example, when the CPU 101 (i.e., the computer of the partial discharge inspection apparatus 10) executes the partial discharge inspection program 103a. This partial discharge inspection program 103a may be distributed by being stored in a computer-readable storage medium, or may be downloaded to the partial discharge inspection apparatus 10 via a network.

[0039] Here, it has been described that some or all of the units 11 to 17 are realized by software, but some or all of the units 11 to 17 may also be realized by hardware such as an IC (Integrated Circuit), or may be realized by a combination of software and hardware.

[0040] The nonvolatile memory 102 is a storage medium used as an auxiliary storage device. The RAM 103 is a storage medium used as a main storage device. Although only the nonvolatile memory 102 and the RAM 103 are shown in Fig. 4, the partial discharge inspection device 10 may also include other storage devices such as an HDD (Hard Disk Drive) and an SSD (Solid State Drive).

[0041] The communication device 104 is a device configured to perform wired or wireless communication with an external device. Note that, although the external device in this embodiment includes, for example, the sensor 21 attached to the power device 20, the partial discharge inspection apparatus 10 may be configured to be able to communicate with an external device such as a server device via the communication device 104.

[0042] Although not shown in FIG. 4, the partial discharge inspection device 10 may further include an input device such as a mouse or a keyboard, and a display device such as a display.

[0043] Next, an example of a processing procedure of the partial discharge inspection device 10 will be described with reference to the flowchart of Fig. 5. Note that the processing shown in Fig. 5 may be executed at a timing instructed by, for example, an administrator of the electric power device 20 (or the partial discharge inspection device 10), or may be executed automatically at a predetermined timing (time).

[0044] 5 is executed, the sensor 21 measures an electrical signal generated in response to the operation of the electric power equipment 20. For example, if the sensor 21 is a transient earth potential sensor (TEV sensor), the sensor 21 is attached to, for example, the panel surface of the electric power equipment 20 (inverter panel), and measures an electrical signal representing the potential of the panel surface of the electric power equipment 20. The electrical signal measured by the sensor 21 is output (transmitted) from the sensor 21 to the partial discharge inspection device 10 via, for example, a cable.

[0045] As described above, the electrical signal acquiring unit 11 acquires the electrical signal output from the sensor 21 to the partial discharge inspection device 10 (step S1). Note that Fig. 6 shows an example of the electrical signal acquired in step S1, and the electrical signal includes a plurality of pulse signals.

[0046] Next, the pulse signal extractor 12 extracts a pulse signal from the electrical signal acquired in step S1 (step S2).

[0047] In this case, pulse signal extraction unit 12 applies a predetermined low-pass filter to the electrical signal (raw waveform) acquired in step S1, for example, and sets the average value of the signal values ​​(e.g., voltage values) that have passed through the low-pass filter as a threshold. Pulse signal extraction unit 12 detects the maximum value of the signal values ​​that exceed such a threshold as the peak value of the pulse signal, and can extract a signal within a predetermined time range that includes the detected peak value as the pulse signal.

[0048] The process of extracting the pulse signal described here is an example, and in step S2, the pulse signal may be extracted by executing other processes.

[0049] After the process of step S2 is executed, it is determined whether or not all pulse signals have been detected from the electrical signals acquired in step S1 (step S3).

[0050] If it is determined that all pulse signals have not been detected (NO in step S3), the process returns to step S2 and is repeated. By repeatedly executing the process of step S2 in this manner, each of the multiple pulse signals is extracted from the electrical signal acquired in step S1.

[0051] On the other hand, if it is determined that all pulse signals have been detected (YES in step S3), the processes of steps S4 to S7 are executed for each of the plurality of pulse signals extracted in step S2. In the following description, the pulse signal that is the target of the processes of steps S4 to S7 is referred to as the target pulse signal.

[0052] First, the time component information acquisition unit 13 acquires time component information indicating the time component of the target pulse signal (step S4). Note that the time component indicated by the time component information in this embodiment includes a weighted average of the time width of the target pulse signal. This weighted average T of the time width of the target pulse signal is calculated by the following equation (1).

number

[0053] In the above formula (1), t i (i=1,2,...,N) represents the sampling time for the target pulse signal, and t0 represents the time of the center of gravity of the target pulse signal. i is the sampling time t i 4 shows the signal output (signal strength) of the target pulse signal at

[0054] In step S4, time component information is acquired that indicates, as a time component, the weighted average T of the time width of the target pulse signal calculated by the above-mentioned equation (1).

[0055] Next, the frequency component information acquisition unit 14 acquires frequency component information indicating the frequency components of the target pulse signal (step S5). Note that the frequency components indicated by the frequency component information in this embodiment include a weighted average of the results of applying a Fourier transform to the target pulse signal (hereinafter referred to as the weighted average of the Fourier transform results for the target pulse signal). This weighted average F of the Fourier transform results for the target pulse signal is calculated using the following equation (2).

number

[0056] The Fourier transform is an analytical method for converting a time domain expression into a frequency domain expression. i (i=1,2,…,N) represents the frequency domain after Fourier transform. i is the frequency domain f i This represents the signal strength (frequency strength) at

[0057] In step S5, frequency component information is acquired that indicates, as a frequency component, the weighted average F of the Fourier transform result for the target pulse signal calculated by the above-mentioned equation (2).

[0058] Next, the signal strength information acquiring unit 15 acquires signal strength information indicating the signal strength of the target pulse signal (step S6). Note that the signal strength indicated by the signal strength information in this embodiment includes the maximum absolute value V of the signal value (voltage value) of the target pulse signal.

[0059] The time component information (time component of the target pulse signal), frequency component information (frequency component of the target pulse), and signal intensity information (signal intensity of the target pulse signal) described here are merely examples, and in this embodiment, it is sufficient to acquire information on the time component, frequency component, and signal intensity of the target pulse signal used to create a TFV map.

[0060] Furthermore, in FIG. 5, the processing is described as being executed in the order of steps S4 to S6, but the order of the processing of steps S4 to S6 may be changed, or the processing of steps S4 to S6 may be executed in parallel.

[0061] When the processing of steps S4 to S6 described above is executed, the map creation unit 16 plots the time component information (weighted average T of the time width of the target pulse signal), frequency component information (weighted average F of the Fourier transform result for the target pulse signal), and signal strength information (maximum absolute value V of the signal value of the target pulse signal) acquired in steps S4 to S6 on a TFV map format prepared in advance (step S7).

[0062] As described above, the processes of steps S4 to S7 are performed for all pulse signals extracted in step S2. This creates a TFV map in which the time component information, frequency component information, and signal intensity information acquired by performing the processes of steps S4 to S6 are plotted for all pulse signals.

[0063] 7 shows an example of a TFV map (TFV plot). In this embodiment, the TFV map is assumed to be a map in which the horizontal axis represents time components and the vertical axis represents frequency components, and the signal strength (voltage value) is expressed by changing colors (i.e., a map that three-dimensionally represents time component information, frequency component information, and signal strength information).

[0064] Returning to FIG. 5 again, the inspection unit 17 inspects whether or not a partial discharge has occurred in the electric power device 20 based on the TFV map created as described above (step S8).

[0065] An example of the processing of step S8 will be described below. If the TFV map created by performing the processing of steps S4 to S7 described above for each pulse signal is referred to as a target TFV map, the processing of step S8 is performed by comparing the target TFV map with a comparison TFV map. Note that the comparison TFV map is assumed to be stored in advance inside the partial discharge inspection device 10 (inspection unit 17).

[0066] Specifically, the comparative TFV map is a TFV map (hereinafter referred to as a first comparative TFV map) that plots time component information, frequency component information, and signal intensity information for each of a plurality of pulse signals included in an electrical signal measured by a sensor 21 attached to the electric power equipment 20 when no partial discharge is occurring. In other words, the first comparative TFV map is a TFV map created based on an electrical signal that does not include a partial discharge signal (a TFV map of the electric power equipment 20 when no partial discharge is occurring). In this case, in step S8, the presence or absence of partial discharge in the electric power equipment 20 can be checked by evaluating the difference between the target TFV map and the first comparative TFV map.

[0067] Here, Fig. 8 shows an example of the target TFV map, and Fig. 9 shows an example of the first comparative TFV map. In this case, a clear difference is observed between the target TFV map shown in Fig. 8 and the first comparative TFV map shown in Fig. 9, mainly in terms of signal strength, and therefore, in step S8, it is determined that a partial discharge has occurred in the power equipment 20 (i.e., the electrical signal measured by the sensor 21 contains a partial discharge signal). On the other hand, if no difference is observed between the target TFV and the first comparative TFV map, it is determined in step S8 that a partial discharge has not occurred in the power equipment 20 (i.e., the electrical signal measured by the sensor 21 does not contain a partial discharge signal).

[0068] That is, in this embodiment, the occurrence of partial discharge can be inspected based on the change over time from the first comparative TFV map (a TFV map created in advance when the electric power equipment 20 is not generating partial discharge) stored in advance inside the partial discharge inspection device 10 to the target TFV map.

[0069] Here, the first comparative TFV map created in advance when the electric power equipment 20 was not experiencing partial discharge has been described as being used, but the comparative TFV map may also be a TFV map of the electric power equipment 20 in a state where partial discharge is occurring (hereinafter referred to as a second comparative TFV map). Note that the second comparative TFV map corresponds to a past TFV map created based on an electrical signal measured by a sensor attached to the electric power equipment 20 experiencing partial discharge.

[0070] In this case, the inspection unit 17 can inspect whether or not partial discharge occurs in the electric device 20 by evaluating the similarity between the target TFV map and the second comparative TFV map.

[0071] Specifically, for example, if the similarity between the target TFV map and the second comparison TFV map is equal to or greater than a predetermined value, the inspection unit 17 can determine that partial discharge is occurring in the electric equipment 20, and if the similarity is less than the predetermined value, the inspection unit 17 can determine that partial discharge is not occurring in the electric equipment 20. The similarity between the target TFV map and the second comparison TFV map can be calculated based on, for example, the arrangement of multiple plots in the target TFV map and the arrangement of multiple plots in the second comparison TFV map.

[0072] The second comparative TFV map is assumed to be stored in advance in, for example, a database that manages information about partial discharges that have occurred in the past. In this embodiment, a configuration can be adopted in which the presence or absence of partial discharges is inspected using such a second comparative TFV map.

[0073] In addition, although the present description has been given of inspecting the occurrence of partial discharge from the electric equipment 20 by comparing the target TFV map with the comparative TFV map (first or second comparative TFV map), the processing of step S8 described above may be performed using a statistical model generated based on a technology known as machine learning or artificial intelligence (AI).

[0074] The statistical model is assumed to be generated in advance by executing a learning process using previously created TFV maps such as the first or second comparative TFV map described above (i.e., a TFV map created from an electrical signal that does not include a partial discharge signal or a TFV map created from an electrical signal that includes a partial discharge signal) as learning data. This makes it possible to construct a statistical model that, for example, inputs a target TFV map, identifies TFV maps (plots) similar to the target TFV map, and outputs an inspection result indicating whether or not a partial discharge is occurring in the power equipment 20 based on the identification result.

[0075] It is to be noted that the TFV map is likely to change depending on the measurement environment of the electrical signal (for example, the installation position of the sensor 21, the length of the cable connecting the sensor 21 and the partial discharge inspection device 10, etc.). Therefore, by performing a learning process using past TFV maps created based on electrical signals measured in various measurement environments, it is possible to construct a statistical model that can flexibly respond to changes in the measurement environment.

[0076] Furthermore, the statistical model may be, for example, a neural network. However, if the TFV map is treated as three-dimensional information and the three-dimensional information is input to the statistical model, a convolutional neural network (CNN), which is generally considered suitable for recognizing three-dimensional information, may be used. However, the statistical model may also be, for example, a fully connected neural network or a recurrent neural network. Furthermore, the statistical model may be a random forest or may be generated by another machine learning algorithm.

[0077] 8 and 9, for example, a difference is recognized mainly in signal strength between the TFV map of the electric power equipment 20 in a state where a partial discharge is occurring and a TFV map of the electric power equipment 20 in a state where a partial discharge is not occurring, and therefore the processing of step S8 described above may be executed based on the signal strength (voltage value). Specifically, for example, the signal strength indicated by the signal strength information plotted on the TFV map of the electric power equipment 20 in a state where a partial discharge is not occurring may be set as a threshold, and if the signal strength indicated by the signal strength information plotted on the target TFV map exceeds the threshold, it may be determined that a partial discharge is occurring in the electric power equipment 20.

[0078] According to the example shown in Fig. 9, the maximum value (maximum voltage value) of the signal strength (information) plotted on the TFV map of the electric power equipment 20 in a state where no partial discharge is occurring is about 4 V. In contrast, the maximum value of the signal strength plotted on the TFV map of the electric power equipment 20 in a state where a partial discharge is occurring as shown in Fig. 8 is about 6 V, so it is possible to determine that a partial discharge is occurring in the electric power equipment 20 based on the TFV map shown in Fig. 8.

[0079] Furthermore, in the TF maps shown in FIGS. 1 and 2 described above, the plots are not discrete, making it difficult to determine whether a partial discharge has occurred in the electric power equipment 20. However, the TFV map in this embodiment is created by plotting signal intensity information in addition to time component information and frequency component information.

[0080] In this way, a TFV map that takes into account signal strength information in addition to time component information and frequency component information is considered to be able to express the discreteness of multiple plots when partial discharges occur. Specifically, for example, in the TFV map shown in FIG. 8 described above, plots corresponding to partial discharge signals with high signal strength are discretely arranged in an area where plots corresponding to noise signals with low signal strength are arranged. For this reason, in this embodiment, the presence or absence of partial discharges may be inspected based on whether or not multiple plots in the target TFV map have discreteness (are discretely arranged).

[0081] Here, a plurality of methods for inspecting whether or not partial discharge has occurred in the electric power equipment 20 has been described, but in this embodiment, two or more of these methods may be combined to inspect whether or not partial discharge has occurred in the electric power equipment 20. That is, in this embodiment, for example, the inspection results based on at least two or more of the above-described plurality of methods may be comprehensively taken into consideration to inspect whether or not partial discharge has occurred in the electric power equipment 20 (to determine whether or not partial discharge has occurred).

[0082] In the example shown in FIG. 5, the process shown in FIG. 5 is shown to end when the process of step S8 is executed, but the result of the process of step S8 (i.e., the inspection result of whether or not a partial discharge has occurred in the power equipment 20) is output from the inspection unit 17. In this case, the inspection result may be transmitted to, for example, an external device (e.g., a server device) of the partial discharge inspection apparatus 10, or may be displayed on a display device (display) provided in the partial discharge inspection apparatus 10. Furthermore, the inspection result may be held (accumulated) inside the partial discharge inspection apparatus 10.

[0083] As described above, the partial discharge inspection device 10 according to this embodiment acquires an electrical signal including a plurality of pulse signals measured by the sensor 21 attached to the electric power equipment 20, extracts each of the plurality of pulse signals from the acquired electrical signal, acquires time component information indicating the time component, frequency component information indicating the frequency component, and signal strength information indicating the signal strength of each of the extracted plurality of pulse signals, and creates a target TFV map in which the acquired time component information, frequency component information, and signal strength information are plotted for each pulse signal.

[0084] Each of the multiple pulse signals included in the electrical signal can be extracted, for example, by applying a low-pass filter to the electrical signal and extracting the signal value based on the signal value exceeding the average value of the signal values ​​that have passed through the low-pass filter. The time component indicated by the time component information includes, for example, a weighted average of the time width of the pulse signal. The frequency component indicated by the frequency component information includes, for example, a weighted average of the results of applying a Fourier transform to the pulse signal. The signal intensity indicated by the signal intensity information includes, for example, the maximum absolute value of the signal value of the pulse signal.

[0085] In this embodiment, the above-described configuration makes it possible to realize a partial discharge inspection device 10 that is useful for inspecting whether or not partial discharge has occurred.

[0086] Specifically, the partial discharge inspection device 10 according to this embodiment is configured to inspect whether or not partial discharge has occurred in the electric power equipment 20, based on the target TFV map described above.

[0087] In this case, the occurrence of partial discharge is checked, for example, by comparing the target TFV map with the above-mentioned first comparative TFV map (for example, a TFV map created in advance when no partial discharge is occurring). That is, in this embodiment, the occurrence of partial discharge in the electric equipment 20 can be checked based on the change over time of the TFV map (i.e., the transition over time from the TFV map of the electric equipment 20 when it is new and no partial discharge has occurred to the TFV map of the current electric equipment 20).

[0088] With this configuration, even in a case where the time component information and frequency component information of a partial discharge signal and a noise signal are similar to each other as in the comparative example of this embodiment, and therefore a set of multiple plots does not have discreteness in the TF map in which the time component information and frequency component information are plotted, it is possible to appropriately check for the occurrence of partial discharge by using the TFV map.

[0089] Note that this embodiment may be configured to check for the occurrence of partial discharge by comparing the target TFV map with a second comparative TFV map (a past TFV map created based on an electrical signal measured by a sensor 21 attached to the power equipment 20 in which partial discharge is occurring). In other words, in this embodiment, the occurrence of partial discharge in the power equipment 20 may be checked based on whether the target TFV map is similar to a past TFV map stored in a database or the like that manages information about past partial discharges.

[0090] Furthermore, this embodiment may be configured to check for the occurrence of partial discharge based on a target TFV map. Therefore, this embodiment may be configured to check for the occurrence of partial discharge using a statistical model (e.g., a convolutional neural network or a random forest) generated by learning a previously created TFV map, or to check for the occurrence of partial discharge based on whether the signal strength indicated by the signal strength information plotted on the target TFV map is equal to or greater than a predetermined value (threshold value), or to check for the occurrence of partial discharge based on whether multiple plots corresponding to combinations of time component information, frequency component information, and signal strength information on the target TFV map have discreteness.

[0091] That is, in this embodiment, in order to inspect for the occurrence of partial discharge, it is only necessary to create a TFV map in which time component information, frequency component information, and signal intensity information of a plurality of pulse signals extracted from an electrical signal are plotted, and various processes can be performed as a process for inspecting for the occurrence of partial discharge using the TFV map.

[0092] Furthermore, in the present embodiment, the partial discharge inspection device 10 has been described as having the function of automatically inspecting for the occurrence of partial discharge using a TFV map. However, the partial discharge inspection device 10 may be configured to be used for inspecting for the occurrence of partial discharge. That is, the partial discharge inspection device 10 according to the present embodiment may be configured to display (output) the TFV map created by the map creation unit 16 on a display device or the like provided in the partial discharge inspection device 10. In this case, for example, a manager of the electric power equipment 20 may determine the occurrence of partial discharge in the electric power equipment 20 by referring to the TFV map displayed on the display device. Furthermore, the TFV map created by the map creation unit 16 may be transmitted to an external device (such as a server device) or may be stored (accumulated) within the partial discharge inspection device 10.

[0093] As described above, when the partial discharge inspection device 10 does not inspect for the occurrence of partial discharge, the partial discharge inspection device 10 may be configured to omit the inspection unit 17. In other words, the partial discharge inspection device 10 according to this embodiment only needs to be configured to have at least the function of creating a TFV map used to inspect for the occurrence of partial discharge from the electric power equipment 20.

[0094] In the TF map used in the comparative example of this embodiment described above, plots corresponding to combinations of time component information and frequency components of pulse signals corresponding to the time component information are arranged, as shown in Figures 1 and 2. In contrast, the TFV map of this embodiment adds signal intensity (voltage value) to the TF map used in the comparative example of this embodiment, making it possible to more appropriately represent the characteristics of partial discharge signals contained in electrical signals (for example, to clearly indicate the discreteness between partial discharge signals and noise signals), thereby improving the accuracy of inspection for the occurrence of partial discharges.

[0095] In addition, in this embodiment, the sensor 21 is a TEV sensor, and the sensor 21 is described as measuring an electrical signal representing the potential on the panel surface of the electric power equipment 20 (inverter panel). However, the electrical signal may be any signal representing a physical quantity generated in response to the operation of the electric power equipment 20. Specifically, the electrical signal may be a signal representing a current flowing through a grounding wire connected to the electric power equipment 20, in which case a current transformer or the like can be used as the sensor 21. Furthermore, the electrical signal may be a signal representing an electromagnetic wave generated in response to the operation of the electric power equipment 20, in which case a TEV sensor or the like can be used as the sensor 21. Furthermore, the electrical signal may be a signal representing a sound wave generated in response to the operation of the electric power equipment 20, in which case an AE (Acoustic Emission) sensor or the like can be used as the sensor 21. In this embodiment, even when an electrical signal representing a physical quantity other than the potential on the panel surface of the electric power equipment 20 is measured, a TFV map can be created in which a combination of time component information, frequency component information, and signal intensity information obtained from the electrical signal (a plurality of pulse signals extracted from the electrical signal) is plotted for each pulse signal, and the TFV map can be used to check whether or not partial discharge is occurring in the electric power equipment 20.

[0096] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents. [Explanation of symbols]

[0097] 10...Partial discharge inspection device, 11...Electrical signal acquisition unit, 12...Pulse signal extraction unit, 13...Time component information acquisition unit, 14...Frequency component information acquisition unit, 15...Signal strength information acquisition unit, 16...Map creation unit, 17...Inspection unit, 20...Power equipment, 21...Sensor, 101...CPU, 102...Non-volatile memory, 103...RAM, 103a...Partial discharge inspection program, 104...Communication device

Claims

1. A partial discharge inspection support method executed by a partial discharge inspection support device, comprising: acquiring an electrical signal including a plurality of pulse signals measured by a sensor attached to the electric power equipment; extracting each of the plurality of pulse signals from the acquired electrical signal; acquiring time component information indicating a time component, frequency component information indicating a frequency component, and signal intensity information indicating a signal intensity of each of the extracted plurality of pulse signals; creating a map in which the acquired time component information, frequency component information, and signal intensity information are plotted for each of the extracted pulse signals; Equipped with The created map is used to check whether or not partial discharge occurs in the power equipment. Partial discharge inspection support method.

2. 2. The partial discharge inspection support method according to claim 1, wherein the extracting step includes a step of applying a low-pass filter to the acquired electrical signal and extracting each of a plurality of pulse signals included in the electrical signal based on a signal value that exceeds an average value of signal values ​​that have passed through the low-pass filter.

3. 2. The partial discharge inspection support method according to claim 1, wherein the time component indicated by the time component information includes a weighted average of the time width of the pulse signal.

4. 2. The partial discharge inspection support method according to claim 1, wherein the frequency component indicated by the frequency component information includes a weighted average of a result of applying a Fourier transform to the pulse signal.

5. 2. The partial discharge inspection support method according to claim 1, wherein the signal strength indicated by the signal strength information includes a maximum absolute value of the signal value of the pulse signal.

6. 2. The partial discharge inspection support method according to claim 1, further comprising the step of inspecting whether or not the partial discharge has occurred based on the created map.

7. 7. The partial discharge inspection support method according to claim 6, wherein the step of inspecting includes a step of inspecting whether or not the partial discharge has occurred based on a change over time in the created map.

8. 7. The partial discharge inspection support method according to claim 6, wherein the inspecting step includes a step of inspecting whether or not a partial discharge has occurred by comparing the created map with a past map created based on an electrical signal measured by a sensor attached to an electric device in which a partial discharge is occurring.

9. 7. The partial discharge inspection support method according to claim 6, wherein the inspecting step includes a step of inspecting whether or not the partial discharge has occurred using a statistical model generated by learning a map created in the past.

10. The partial discharge inspection support method according to claim 9, wherein the statistical model is a convolutional neural network.

11. The partial discharge inspection support method according to claim 9 , wherein the statistical model is a random forest.

12. 7. The partial discharge inspection support method according to claim 6, wherein the inspecting step includes a step of inspecting whether or not the partial discharge has occurred based on whether or not the signal strength indicated by the signal strength information plotted on the created map is equal to or greater than a predetermined value.

13. 2. The partial discharge inspection support method according to claim 1, further comprising the step of displaying the created map on a display device provided in the partial discharge inspection support device.

14. The partial discharge inspection support method according to any one of claims 1 to 13, wherein the electrical signal is a signal representing the potential of a panel surface of the electric power device.

15. The partial discharge inspection support method according to any one of claims 1 to 13, wherein the electrical signal is a signal representing a current flowing through a grounding wire connected to the power device.

16. The partial discharge inspection support method according to any one of claims 1 to 13, wherein the electrical signal is a signal representing an electromagnetic wave generated in response to an operation of the power equipment.

17. The partial discharge inspection support method according to any one of claims 1 to 13, wherein the electric signal is a signal representing a sound wave generated in response to an operation of the electric power equipment.

18. a first acquiring means for acquiring an electric signal including a plurality of pulse signals measured by a sensor attached to the electric power equipment; an extracting means for extracting each of the plurality of pulse signals from the acquired electrical signal; a second acquiring means for acquiring time component information indicating a time component, frequency component information indicating a frequency component, and signal intensity information indicating a signal intensity of each of the extracted plurality of pulse signals; a generating means for generating a map in which the acquired time component information, frequency component information, and signal intensity information are plotted for each of the extracted pulse signals; Equipped with The created map is used to check whether or not partial discharge occurs in the power equipment. Partial discharge inspection support device.

Images