Substation equipment defect diagnosis method and system based on sound signal heterogeneous topology

Abstract

The invention relates to the technical field of power equipment state monitoring, and provides a substation equipment defect diagnosis method and system based on sound signal heterogeneous topology, and the method comprises the steps: obtaining a digital signal and an analog signal of to-be-diagnosed equipment; the method comprises the following steps: sampling an analog signal, taking an average value of the maximum amplitude and the minimum amplitude of sampling, taking the average value as an average projection mapping interval of a digital-analog cooperative signal, converting the analog signal into a digital signal, calculating an intersection of the obtained digital signal and the digital signal obtained by converting the analog signal, and calculating the position of a true value in the intersection. The obtained digital signals and analog signals are reserved, otherwise, the digital signals and the analog signals are deleted, and the denoised digital signals and the denoised analog signals are obtained; and performing spectrogram conversion and image recognition on the denoised digital signal and the denoised analog signal to obtain an equipment defect diagnosis result. On the premise of ensuring the integrity of the signal, the noise is effectively removed, and the accuracy of equipment defect diagnosis is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power equipment status monitoring, and in particular relates to a method and system for diagnosing substation equipment defects based on sound signal heterogeneous topology. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] The traditional method of monitoring the status of power equipment in substations is mainly a combination of manual inspection and visual monitoring. This method has the problem of discovery delay and cannot achieve timely prediction of the status of substation equipment. With the development of information technology, online monitoring technology based on voiceprints has gradually been applied to the field of status monitoring and fault diagnosis of power equipment, but voiceprint monitoring is susceptible to environmental noise, the voiceprint monitoring device has insufficient computing power, and the distance between the voiceprint collection device and the substation equipment cannot be too far. CN111192447A proposes a substation voiceprint detection and identification device, method and system based on the Internet of Things, and designs a monitoring scheme including a central processing unit, a memory, a MEMS microphone, an ADC module, an FPGA and other modules. However, the scheme still collects voiceprint data through a traditional sound collection unit and performs data analysis through an FPGA module. The natural high power consumption and low computing power of FPGA cannot effectively overcome the problems of large environmental noise influence and insufficient computing power of the voiceprint monitoring device.

[0004] The current voiceprint defect recognition method for substation equipment mainly collects sound information through a microphone array, converts the sound digital signal into a frequency domain signal through a sound conversion algorithm, removes environmental noise based on the sound frequency domain signal-to-noise ratio parameter, and determines whether the equipment sound is normal based on the equipment normal sound frequency template matching. The traditional voiceprint defect recognition method can achieve preliminary monitoring of the equipment status, but there are still the following problems:

[0005] (1) The effectiveness of noise removal cannot be guaranteed: When collecting information through a single microphone array or vibration sensor, the voiceprint information contains too much environmental noise information, making it difficult to completely remove the noise, and thus, effective noise removal cannot be achieved;

[0006] (2) Signal integrity is difficult to guarantee: The noise removal process often uses the method of converting digital signals into frequency domain signals, but the effect on nonlinear signals is poor. When nonlinear abnormal vibration occurs in the equipment, such signals may be removed as noise, and the signal integrity cannot be guaranteed.

[0007] (3) The sound signal analysis process is complex: The sound signal acquisition and analysis process is controlled by the central processing unit. The data flow passes through the microphone, memory, digital-to-analog converter (ADC), and FPGA module in sequence. Parallel analysis is not possible, and the process efficiency is low.

[0008] (4) The sound monitoring and analysis device has low energy consumption ratio: The computing module of the sound monitoring and analysis device is mainly FPGA. FPGA has the natural characteristic of high power consumption. The higher the computing power, the higher the power consumption. The sound monitoring device cannot provide enough battery due to environmental factors such as the collection location. Therefore, the FPGA analysis module generally selects a low computing power model. There is a contradiction between the high computing power and low power consumption of the sound monitoring and analysis device. Summary of the invention

[0009] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a substation equipment defect diagnosis method and system based on the heterogeneous topology of sound signals, designs a digital-analog collaborative parallel denoising method, designs a digital-analog collaborative signal average projection mapping interval, and realizes preliminary noise removal in the sound collection stage through the intersection of digital signals and analog signals; and converts sound information into a spectrum diagram, adopts an image recognition method to identify the sound information, enhances the recognition ability of nonlinear abnormal signals, reduces the probability of misjudging nonlinear abnormal signals as noise signals, and realizes effective noise removal while ensuring the integrity of the signal, thereby improving the accuracy of equipment defect diagnosis.

[0010] In order to achieve the above object, the present invention adopts the following technical solution:

[0011] A first aspect of the present invention provides a method for diagnosing substation equipment defects based on sound signal heterogeneous topology, which comprises:

[0012] Obtain digital and analog signals from the device to be diagnosed;

[0013] Designing an average projection mapping interval of a digital-analog collaborative signal, and performing digital-analog collaborative parallel denoising on a digital signal and an analog signal, wherein the steps of the digital-analog collaborative parallel denoising include: sampling the analog signal, taking the average of the maximum amplitude and the minimum amplitude of the sampling, taking the average as the average projection mapping interval of the digital-analog collaborative signal, converting the analog signal into a digital signal, calculating the intersection of the acquired digital signal and the digital signal obtained by converting the analog signal, retaining the acquired digital signal and the analog signal at the position with the true value in the intersection, otherwise deleting them, to obtain a denoised digital signal and a denoised analog signal;

[0014] For the denoised digital signal and the denoised analog signal, the spectrum conversion and image recognition are performed respectively to obtain the equipment defect diagnosis results.

[0015] Furthermore, for samples of the analog signal obtained by sampling, samples below the average value are set to 0, and samples above the average value are set to 1, thereby obtaining a converted digital signal.

[0016] Furthermore, for the denoised digital signal, after conversion into a spectrum graph, image recognition is used to determine whether there is any abnormality in the equipment sound, and a first judgment result is obtained; for the denoised analog signal, after conversion into a spectrum graph, image recognition is used to determine whether there is any abnormality in the equipment sound, and a second judgment result is obtained; when the first judgment result is consistent with the second judgment result, the equipment defect diagnosis result is obtained.

[0017] Furthermore, the image recognition uses a trained sound information spectrogram recognition model, and the training steps of the sound information spectrogram recognition model include:

[0018] For the denoised digital signal or the denoised analog signal, a discrete Fourier expansion transform is performed to obtain a spectrum diagram;

[0019] Based on the spectrum graph, the spectrum graph features are calculated, and the equipment defect diagnosis results are marked in the spectrum graph based on the spectrum graph features;

[0020] The spectrum graph is input as a feature graph into the neural network for training to obtain a sound information spectrum graph recognition model.

[0021] Furthermore, the spectrum features include the maximum amplitude in the frequency domain, the minimum amplitude, the median amplitude, the average amplitude, the peak-to-peak amplitude, the main frequency, the vibration entropy, the 50Hz odd-even multiple frequency ratio, the main frequency proportion, the fundamental frequency proportion, the total harmonic distortion rate, the high-to-low frequency ratio and / or the high-frequency proportion.

[0022] A second aspect of the present invention provides a substation equipment defect diagnosis system based on sound signal heterogeneous topology, which includes:

[0023] A sound signal acquisition module, which is configured to: acquire digital signals and analog signals of the device to be diagnosed;

[0024] A merge intersection processing module is configured to: design an average projection mapping interval of a digital-analog collaborative signal, perform digital-analog collaborative parallel denoising on digital signals and analog signals, wherein the steps of the digital-analog collaborative parallel denoising include: sampling the analog signal, taking the average of the maximum amplitude and the minimum amplitude of the sample, taking the average as the average projection mapping interval of the digital-analog collaborative signal, converting the analog signal into a digital signal, calculating the intersection of the acquired digital signal and the digital signal converted from the analog signal, retaining the acquired digital signal and the analog signal for the position with the true value in the intersection, otherwise deleting them, to obtain a denoised digital signal and a denoised analog signal;

[0025] The defect diagnosis module is configured to: perform spectrum conversion and image recognition on the denoised digital signal and the denoised analog signal respectively, and obtain the equipment defect diagnosis result.

[0026] Furthermore, for samples of the analog signal obtained by sampling, samples below the average value are set to 0, and samples above the average value are set to 1, thereby obtaining a converted digital signal.

[0027] Furthermore, for the denoised digital signal, after conversion into a spectrum graph, image recognition is used to determine whether there is any abnormality in the equipment sound, and a first judgment result is obtained; for the denoised analog signal, after conversion into a spectrum graph, image recognition is used to determine whether there is any abnormality in the equipment sound, and a second judgment result is obtained; when the first judgment result is consistent with the second judgment result, the equipment defect diagnosis result is obtained.

[0028] Furthermore, the defect diagnosis module includes:

[0029] The FPGA is configured to: convert the de-noised digital signal into a frequency spectrum, and determine whether the device sound is abnormal through image recognition to obtain a first determination result;

[0030] The central processing unit is configured to: convert the de-noised analog signal into a frequency spectrum, and determine whether the device sound is abnormal through image recognition to obtain a second determination result;

[0031] The result merging and outputting module is configured to obtain the equipment defect diagnosis result when the first judgment result is consistent with the second judgment result.

[0032] Furthermore, the spectrum features include the maximum amplitude in the frequency domain, the minimum amplitude, the median amplitude, the average amplitude, the peak-to-peak amplitude, the main frequency, the vibration entropy, the 50Hz odd-even multiple frequency ratio, the main frequency proportion, the fundamental frequency proportion, the total harmonic distortion rate, the high-to-low frequency ratio and / or the high-frequency proportion.

[0033] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the method for diagnosing substation equipment defects based on heterogeneous topology of sound signals as described above.

[0034] The fourth aspect of the present invention provides a computer device, including a computer-readable storage medium, a processor, and a computer program stored on the computer-readable storage medium and executable on the processor, wherein when the processor executes the program, the steps in the method for diagnosing substation equipment defects based on heterogeneous topology of sound signals as described above are implemented.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] The present invention designs a sound signal acquisition device for digital-analog collaborative acquisition, which uses a digital microphone and an analog microphone for collaborative acquisition, wherein the digital microphone is responsible for collecting digital information of vibration, and the analog microphone module saves the vibration signal of the sound in the form of analog quantity. Finally, through the intersection of the digital microphone and the analog microphone signals, preliminary noise removal can be achieved in the sound acquisition stage.

[0037] The present invention designs a sound information spectrum recognition method, converts sound information into a spectrum, adopts an image recognition method to recognize the sound information, enhances the recognition ability of nonlinear abnormal signals, and reduces the probability of misjudging nonlinear abnormal signals as noise signals.

[0038] The present invention designs a sound signal parallel processing topology structure, removes the digital-to-analog conversion (ADC) module, increases the central processing unit computing task, converts the sound data stream into a digital data stream and an analog data stream, the digital data stream enters the FPGA module, and the analog data stream enters the central processing unit for computing, thereby realizing data stream diversion and enhancing the sound signal parallel analysis capability, while reducing the computing pressure of the FPGA and alleviating the power consumption requirements of the FPGA due to high computing power. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0040] Figure 1 It is a flow chart of a method for diagnosing substation equipment defects based on sound signal heterogeneous topology according to a first embodiment of the present invention;

[0041] Figure 2 is a flow chart of intersection calculation of digital microphone and analog microphone signals in embodiment 1 of the present invention;

[0042] Figure 3 is a voiceprint frequency characteristic diagram of the first embodiment of the present invention;

[0043] Figure 4 is a schematic diagram of a spectrum diagram of Embodiment 1 of the present invention;

[0044] Figure 5 It is a structural diagram of a computer device according to a fourth embodiment of the present invention. DETAILED DESCRIPTION

[0045] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0046] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0047] Embodiment 1

[0048] This embodiment provides a method for diagnosing substation equipment defects based on sound signal heterogeneous topology.

[0049] The present embodiment provides a method for diagnosing substation equipment defects based on sound signal heterogeneous topology, such as Figure 1 As shown, the following steps are included:

[0050] Step 1: Design a sound signal acquisition device for digital-analog collaborative acquisition.

[0051] like Figure 2 As shown, a digital microphone and an analog microphone are used for collaborative collection, wherein the digital microphone is responsible for collecting digital information of vibration, and the analog microphone saves the vibration signal of the sound in the form of analog quantity; finally, the intersection processing unit is combined to calculate the intersection of the digital microphone and analog microphone signals, and the intersection signal is used to denoise the digital signal and the analog signal respectively, so that preliminary noise removal can be achieved in the sound collection stage.

[0052] Specific implementation method:

[0053] Step 101: Design a sound signal acquisition device for digital-analog collaborative acquisition, using a digital microphone and an analog microphone for collaborative acquisition:

[0054] Digital microphones store sound signals in the form of digital data. Specifically, digital microphones have a high-frequency sampling module that can convert vibrations into digital values, convert them into discrete values, and then quantize the discrete values, map them to a certain range of values, and convert the values ​​into binary digital signals to achieve digital data storage;

[0055] The analog microphone acquisition is specifically implemented as follows: the analog microphone has a physical disk that can store vibration signals (analog signals), and the vibration signals are stored in the physical disk in the form of vibration line traces in the form of scratches.

[0056] Step 102: A merging and intersection processing unit is provided behind the digital microphone and the analog microphone to merge the signals of the digital microphone and the analog microphone to obtain an intersection signal.

[0057] Digital signals are discrete digital signals on a time scale and consist of binary digits (0 and 1). They appear as high levels (1) and low levels (0) on the acquisition device. Their information is discontinuous and the signal is expressed in a fixed range, such as 0-1.

[0058] Analog signals are continuous signals on a time scale, and appear as continuous changes in voltage on the acquisition device. The best expression range changes with the strength of the signal and there is no fixed range.

[0059] The main way to find the intersection of the two is to convert the analog signal into discrete samples and perform high-frequency sampling on the analog signal. The sampling frequency is p. The larger the p, the less the analog signal loss. For example, Figure 3 As shown; obtain the maximum and minimum sampling amplitudes, take the average value, and use it as the average projection mapping interval of the digital-analog collaborative signal. The value below the average value is set to 0, and the value above the average value is set to 1; perform "and" calculation on the analog signal converted into a digital signal and the original digital signal collected by the digital microphone, and retain the true value, which is the intersection signal.

[0060] Step 103: For the positions with true values ​​in the intersection signal, both the digital signal and the analog signal are retained, otherwise they are deleted, to obtain a denoised digital signal and a denoised analog signal, thereby obtaining a valid signal after noise removal.

[0061] Step 2: Input the denoised digital signal into the FPGA, and the denoised analog signal into the central processing unit. Both the FPGA and the central processing unit are configured with a sound information spectrum recognition method to determine whether there is an abnormality in the device sound. If the two judgment results output by the FPGA and the central processing unit are consistent, the result is considered reliable and the judgment result is obtained. Otherwise, return to step 1.

[0062] In this embodiment, a sound signal parallel processing topology is designed, the digital-to-analog conversion (ADC) module is removed, the central processing unit computing task is increased, the sound data stream is converted into a digital data stream and an analog data stream, the digital data stream enters the FPGA module, and the analog data stream enters the central processing unit for calculation, thereby realizing data stream diversion, enhancing the parallel analysis capability of the sound signal, and at the same time reducing the computing pressure of the FPGA and alleviating the power consumption requirements of the FPGA due to high computing power.

[0063] The denoised digital data stream and analog data stream obtained in step 1 are used to write the trained sound information spectrum recognition model into the FPGA and central processing unit respectively. The central processing unit is used to infer the analog data stream information, and the FPGA is only used to infer the digital data stream. This design realizes the diversion and parallel inference of analog data and digital data, further improves the inference efficiency, maximizes the use of the computing power of the central processing unit, and reduces the computing pressure of the FPGA.

[0064] Both the FPGA and the central processing unit are equipped with a sound information spectrum recognition method, which converts the sound information (denoised digital signal or denoised analog signal) into a spectrum, and annotates the information according to the image features of the spectrum. The image recognition method is used to identify the sound information, enhance the recognition ability of nonlinear abnormal signals, and reduce the probability of misjudging nonlinear abnormal signals as noise signals.

[0065] Specific implementation:

[0066] (1) Perform discrete Fourier transformation on the sound signal (denoised digital signal or denoised analog signal) obtained in step 1 to obtain a frequency domain spectrum:

[0067]

[0068] Where x(k) is the kth output, k∈(0,N-1), N is the signal length, j is an imaginary number, and e is the base of the natural logarithm.

[0069] (2) Based on the frequency domain analysis, spectrograms of different voiceprint states can be obtained. For a sound signal (a denoised digital signal or a denoised analog signal), there is only one spectrum graph, such as Figure 4 As shown, different voiceprint states are mainly reflected in amplitude, frequency and time. The amplitude is represented by color, the lighter the color, the larger the amplitude. The frequency is the vertical axis and the time is the horizontal axis.

[0070] (3) The spectrum graph and spectrum graph features are input into the trained sound information spectrum graph recognition model to carry out multi-factor comprehensive diagnosis image recognition of defects and determine whether there is any abnormality in the equipment sound.

[0071] In this embodiment, the training process of the sound information spectrogram recognition model includes:

[0072] First, perform discrete Fourier transformation on the sound signal to obtain a spectrum.

[0073] Then, based on the spectrum diagram, 13 features are calculated, including the maximum amplitude, minimum amplitude, median amplitude, average amplitude, peak-to-peak amplitude, main frequency, vibration entropy, 50Hz odd-even multiple ratio, main frequency proportion, fundamental frequency proportion, total harmonic distortion rate, high-to-low frequency ratio, and high-frequency proportion.

[0074] Finally, the equipment status (equipment defect diagnosis results) is annotated in the spectrum graph based on the characteristics of the spectrum graph, and the spectrum graph is input into the neural network as a feature graph for training. A sound information spectrum graph recognition model is constructed, and the spectrum graph is analyzed to determine whether there is any abnormality in the equipment sound.

[0075] The present embodiment provides a substation equipment defect diagnosis method based on a heterogeneous topology of sound signals, designs a sound signal acquisition device for collaborative digital-analog acquisition, and uses a digital microphone and an analog microphone for collaborative acquisition, wherein the digital microphone is responsible for collecting digital information of vibration, and the analog microphone module saves the vibration signal of the sound in the form of an analog quantity. Finally, through the intersection of the digital microphone and the analog microphone signals, preliminary noise removal can be achieved in the sound acquisition stage.

[0076] The present embodiment provides a substation equipment defect diagnosis method based on sound signal heterogeneous topology, designs a sound information spectrum recognition method, converts sound information into a spectrum, uses an image recognition method to recognize the sound information, enhances the recognition capability of nonlinear abnormal signals, and reduces the probability of misjudging nonlinear abnormal signals as noise signals.

[0077] The substation equipment defect diagnosis method based on the sound signal heterogeneous topology provided in this embodiment designs a sound signal parallel processing topology structure, removes the digital-to-analog conversion (ADC) module, increases the central processing unit computing task, converts the sound data stream into a digital data stream and an analog data stream, the digital data stream enters the FPGA module, and the analog data stream enters the central processing unit for computing, thereby realizing data stream diversion, enhancing the parallel analysis capability of the sound signal, and at the same time reducing the computing pressure of the FPGA and alleviating the power consumption requirements of the FPGA due to high computing power.

[0078] The substation equipment defect diagnosis method based on the sound signal heterogeneous topology provided in this embodiment designs a sound signal acquisition device for digital-analog collaborative acquisition from four aspects: sound information acquisition-sound information judgment-result inference calculation-calculation device update, proposes a sound information spectrum diagram recognition method, and constructs a sound signal parallel processing topology structure, which solves the problems of the inability to ensure the effectiveness of noise removal, the difficulty in ensuring signal integrity, and the complexity of the sound signal analysis process, improves the energy consumption ratio of the sound monitoring and analysis device, and alleviates the contradiction between high computing power and low power consumption of the sound monitoring and analysis device.

[0079] Embodiment 2

[0080] This embodiment provides a substation equipment defect diagnosis system based on sound signal heterogeneous topology.

[0081] The substation equipment defect diagnosis system based on sound signal heterogeneous topology provided in this embodiment specifically includes a sound signal acquisition module, a merge intersection processing module and a defect diagnosis module. The sound signal acquisition module includes a digital microphone and an analog microphone, and the defect diagnosis module includes an FPGA, a central processing unit and a result merging output module.

[0082] A sound signal acquisition module, which is configured to: acquire digital signals and analog signals of the device to be diagnosed;

[0083] A merge intersection processing module is configured to: design an average projection mapping interval of a digital-analog collaborative signal, perform digital-analog collaborative parallel denoising on digital signals and analog signals, wherein the steps of the digital-analog collaborative parallel denoising include: sampling the analog signal, taking the average of the maximum amplitude and the minimum amplitude of the sample, taking the average as the average projection mapping interval of the digital-analog collaborative signal, converting the analog signal into a digital signal, calculating the intersection of the acquired digital signal and the digital signal converted from the analog signal, retaining the acquired digital signal and the analog signal for the position with the true value in the intersection, otherwise deleting them, to obtain a denoised digital signal and a denoised analog signal;

[0084] The defect diagnosis module is configured to: perform spectrum conversion and image recognition on the denoised digital signal and the denoised analog signal respectively, and obtain the equipment defect diagnosis result.

[0085] The FPGA is configured to: convert the de-noised digital signal into a frequency spectrum, and determine whether the device sound is abnormal through image recognition to obtain a first determination result;

[0086] The central processing unit is configured to: convert the de-noised analog signal into a frequency spectrum, and determine whether the device sound is abnormal through image recognition to obtain a second determination result;

[0087] The result merging and outputting module is configured to obtain the equipment defect diagnosis result when the first judgment result is consistent with the second judgment result.

[0088] It should be noted here that each module in this embodiment corresponds to each step in Example 1 one by one, and the specific implementation process is the same, which will not be repeated here.

[0089] Embodiment 3

[0090] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the method for diagnosing substation equipment defects based on sound signal heterogeneous topology as described in the first embodiment above are implemented.

[0091] Embodiment 4

[0092] This embodiment provides a computer device, such as Figure 5As shown, it includes a computer-readable storage medium 1003, a processor 1001, a communication interface 1002, and a computer program stored on the computer-readable storage medium 1003 and executable on the processor 1001, wherein the processor 1001, the communication interface 1002, and the computer-readable storage medium 1003 may be connected via a bus or other means. The communication interface 1002 is used to receive and send data, and when the processor 1001 executes the program, the steps in the method for diagnosing substation equipment defects based on heterogeneous topology of sound signals as described in the first embodiment above are implemented.

[0093] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0094] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0095] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0096] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for diagnosing substation equipment defects based on sound signal heterogeneous topology, characterized in that: include: Obtain digital and analog signals from the device to be diagnosed; Designing an average projection mapping interval of a digital-analog collaborative signal, and performing digital-analog collaborative parallel denoising on a digital signal and an analog signal, wherein the steps of the digital-analog collaborative parallel denoising include: sampling the analog signal, taking the average of the maximum amplitude and the minimum amplitude of the sampling, taking the average as the average projection mapping interval of the digital-analog collaborative signal, converting the analog signal into a digital signal, calculating the intersection of the acquired digital signal and the digital signal obtained by converting the analog signal, retaining the acquired digital signal and the analog signal at the position with the true value in the intersection, otherwise deleting them, to obtain a denoised digital signal and a denoised analog signal; The denoised digital signal and the denoised analog signal are converted to spectra, and multi-factor comprehensive diagnostic image recognition of defects is performed on the spectrum image to obtain the equipment defect diagnosis results.

2. The method for diagnosing substation equipment defects based on sound signal heterogeneous topology according to claim 1, characterized in that: For the samples of the analog signal obtained by sampling, those below the average value are set to 0, and those above the average value are set to 1, thereby obtaining the converted digital signal.

3. The method for diagnosing substation equipment defects based on sound signal heterogeneous topology according to claim 1, characterized in that: For the denoised digital signal, after conversion into a spectrum diagram, image recognition is used to determine whether there is an abnormality in the sound of the equipment, and a first judgment result is obtained; for the denoised analog signal, after conversion into a spectrum diagram, image recognition is used to determine whether there is an abnormality in the sound of the equipment, and a second judgment result is obtained; when the first judgment result is consistent with the second judgment result, the equipment defect diagnosis result is obtained.

4. The method for diagnosing substation equipment defects based on sound signal heterogeneous topology according to claim 1, characterized in that: The image recognition uses a trained sound information spectrogram recognition model, and the training steps of the sound information spectrogram recognition model include: For the denoised digital signal or the denoised analog signal, a discrete Fourier expansion transform is performed to obtain a spectrum diagram; Based on the spectrum graph, the spectrum graph features are calculated, and the equipment defect diagnosis results are marked in the spectrum graph based on the spectrum graph features; The spectrum graph is input as a feature graph into the neural network for training to obtain a sound information spectrum graph recognition model.

5. The method for diagnosing substation equipment defects based on sound signal heterogeneous topology according to claim 4, characterized in that: The spectrum features include the maximum amplitude value, the minimum amplitude value, the median amplitude value, the average amplitude value, the peak-to-peak amplitude value, the main frequency, the vibration entropy, the 50Hz odd-even multiple frequency ratio, the main frequency proportion, the fundamental frequency proportion, the total harmonic distortion rate, the high-low frequency ratio and / or the high-frequency proportion.

6. The substation equipment defect diagnosis system based on sound signal heterogeneous topology is characterized by: include: A sound signal acquisition module, which is configured to: acquire digital signals and analog signals of the device to be diagnosed; A merge intersection processing module is configured to: design an average projection mapping interval of a digital-analog collaborative signal, perform digital-analog collaborative parallel denoising on digital signals and analog signals, wherein the steps of the digital-analog collaborative parallel denoising include: sampling the analog signal, taking the average of the maximum amplitude and the minimum amplitude of the sample, taking the average as the average projection mapping interval of the digital-analog collaborative signal, converting the analog signal into a digital signal, calculating the intersection of the acquired digital signal and the digital signal converted from the analog signal, retaining the acquired digital signal and the analog signal for the position with the true value in the intersection, otherwise deleting them, to obtain a denoised digital signal and a denoised analog signal; The defect diagnosis module is configured to: perform spectrum conversion and image recognition on the denoised digital signal and the denoised analog signal respectively, and obtain the equipment defect diagnosis result.

7. The substation equipment defect diagnosis system based on sound signal heterogeneous topology according to claim 6 is characterized in that: For the samples of the analog signal obtained by sampling, those below the average value are set to 0, and those above the average value are set to 1, thereby obtaining the converted digital signal.

8. The substation equipment defect diagnosis system based on sound signal heterogeneous topology according to claim 6 is characterized in that: For the denoised digital signal, after conversion into a spectrum diagram, image recognition is used to determine whether there is an abnormality in the sound of the equipment, and a first judgment result is obtained; for the denoised analog signal, after conversion into a spectrum diagram, image recognition is used to determine whether there is an abnormality in the sound of the equipment, and a second judgment result is obtained; when the first judgment result is consistent with the second judgment result, the equipment defect diagnosis result is obtained.

9. The substation equipment defect diagnosis system based on sound signal heterogeneous topology according to claim 6, characterized in that: The defect diagnosis module comprises: The FPGA is configured to: convert the de-noised digital signal into a frequency spectrum, and determine whether the device sound is abnormal through image recognition to obtain a first determination result; The central processing unit is configured to: convert the de-noised analog signal into a frequency spectrum, and determine whether the device sound is abnormal through image recognition to obtain a second determination result; The result merging and outputting module is configured to obtain the equipment defect diagnosis result when the first judgment result is consistent with the second judgment result.

10. The substation equipment defect diagnosis system based on sound signal heterogeneous topology according to claim 9, characterized in that: The spectrum features include the maximum amplitude value, the minimum amplitude value, the median amplitude value, the average amplitude value, the peak-to-peak amplitude value, the main frequency, the vibration entropy, the 50Hz odd-even multiple frequency ratio, the main frequency proportion, the fundamental frequency proportion, the total harmonic distortion rate, the high-low frequency ratio and / or the high-frequency proportion.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the substation equipment defect diagnosis method based on sound signal heterogeneous topology as described in any one of claims 1 to 5 are implemented.

12. A computer device comprising a computer-readable storage medium, a processor, and a computer program stored on the computer-readable storage medium and executable on the processor, characterized in that: When the processor executes the program, the steps in the substation equipment defect diagnosis method based on sound signal heterogeneous topology as described in any one of claims 1 to 5 are implemented.

Images