Ultrasonic-based partial discharge analysis system and method for oil-immersed transformers

By combining four piezoelectric ultrasonic sensors and a signal processing module, the problems of signal interference and insufficient timeliness in the ultrasonic method for transformer partial discharge analysis are solved, achieving efficient partial discharge identification and location, and improving the operational reliability and safety of transformers.

CN118501638BActive Publication Date: 2024-11-19XIAMEN UNIV OF TECH
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202410953989.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-19
Estimated Expiration
2044-07-17

AI Technical Summary

Technical Problem

Existing ultrasonic methods are susceptible to low-frequency noise interference in transformer partial discharge analysis, resulting in insufficient signal accuracy and effectiveness, poor detection timeliness, and inability to achieve real-time monitoring.

Method used

Four piezoelectric ultrasonic sensors are used to acquire high-sensitivity partial discharge signals. The signals are then filtered and amplified by a signal processing module. The monitoring platform module analyzes the signals using a preset algorithm, including wavelet noise reduction, fault identification, and location, thereby improving the effectiveness and real-time performance of the signals.

Benefits of technology

It improves the accuracy and real-time performance of partial discharge analysis, enabling timely identification of discharge types and location of discharge sources, thereby enhancing the operational reliability and safety of transformers.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118501638B_ABST
    Figure CN118501638B_ABST
Patent Text Reader

Abstract

The present invention provides an ultrasonic-based partial discharge analysis system and method for an oil-immersed transformer. By utilizing the above technical solution, four piezoelectric ultrasonic sensors can be used to obtain a highly sensitive partial discharge ultrasonic signal within a preset frequency band, thereby improving the effectiveness of the partial discharge ultrasonic signal. The partial discharge ultrasonic signal is then filtered and amplified by a signal processing module to obtain a target signal, thereby further improving the effectiveness of the signal. The target signal is analyzed by a monitoring platform module equipped with various functional modules, and the discharge status of the current target transformer is determined by a preset algorithm, thereby improving the effectiveness of the signal, thereby improving the effectiveness and real-time performance of the partial discharge analysis.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of electrical equipment detection, and in particular to an ultrasonic-based partial discharge analysis system and method for an oil-immersed transformer. Background Art

[0002] During the operation of the transformer, the operating status of its insulation system needs to be monitored in real time to ensure the effectiveness and reliability of the transformer. At present, the main method for monitoring the operating status of the transformer insulation system is partial discharge analysis, which can determine the fault type and discharge location in real time when an abnormality occurs. The ultrasonic method is a non-invasive detection method in partial discharge analysis, which has the characteristics of simple structure, economy and strong flexibility in application scenarios.

[0003] However, for the technical solution of partial discharge analysis using ultrasonic method, ultrasonic signals are easily affected by low-frequency noise, such as Gaussian white noise, narrowband noise and other types of noise, resulting in poor accuracy and effectiveness of ultrasonic signals. At the same time, the current ultrasonic method usually requires offline monitoring when the transformer is out of system operation, so there are certain deficiencies in timeliness. In summary, there are problems with the effectiveness and real-time performance of measurement and analysis in the prior art, so a partial discharge analysis system and corresponding method based on ultrasonic waves are needed to improve the effectiveness and real-time performance of partial discharge analysis. Summary of the invention

[0004] The embodiments of the present invention provide an ultrasonic-based oil-immersed transformer partial discharge analysis system and method to solve the problem that the accuracy and effectiveness of ultrasonic signals are poor, thereby resulting in insufficient effectiveness and real-time performance of partial discharge analysis.

[0005] In order to achieve the above objectives, in a first aspect, the present application provides an ultrasonic-based partial discharge analysis system for oil-immersed transformers, the system comprising:

[0006] Signal acquisition module, signal processing module and monitoring platform module;

[0007] The signal acquisition module includes four piezoelectric ultrasonic sensors installed at preset positions of the target transformer, the signal processing module includes a filter amplifier circuit and a digital-to-analog conversion device, the filter amplifier circuit includes a filter subcircuit, an amplifier subcircuit and a zero adjustment subcircuit, the monitoring platform module includes a monitoring device, and the monitoring device is equipped with a monitoring software platform, the signal acquisition module is connected to the signal processing module, and the signal processing module is connected to the monitoring platform module;

[0008] Among them, three piezoelectric ultrasonic sensors are arranged on the front of the target transformer box in a magnetic way, and the fourth piezoelectric ultrasonic sensor is arranged on the back of the target transformer box in a magnetic way, and the four piezoelectric ultrasonic sensors form a regular tetrahedron;

[0009] The signal acquisition module is used to obtain the partial discharge ultrasonic signal of the target transformer, the signal processing module is used to process the partial discharge ultrasonic signal to obtain the amplified signal in the target frequency band as the target signal, and the monitoring platform module is used to determine the partial discharge condition of the target transformer according to the target signal.

[0010] As an optional implementation, the monitoring software platform includes a data reading submodule, a data display submodule and a main function module;

[0011] Wherein, the data reading submodule is used to collect the target signal according to the preset collection frequency, the data display submodule is used to display the filtered target signal in real time on the monitoring software platform, or transmit the filtered target signal to an external oscilloscope, and display the filtered target signal in real time on the external oscilloscope, and the main function module includes a partial discharge alarm submodule, and the partial discharge alarm submodule is used to carry the energy mean criterion algorithm, analyze the target signal, determine the partial discharge situation and issue an alarm;

[0012] In addition, the monitoring software platform also includes a digital filtering submodule or a data storage submodule, which is used to perform corresponding functions according to the operation status of the monitoring software platform. The digital filtering submodule is used to perform high-pass and low-pass filtering on the target signal, and the data storage submodule is used to save the target signal.

[0013] As an optional implementation, the monitoring software platform further includes a function jump submodule and a data processing submodule;

[0014] Among them, the function jump submodule is used to select the main function module, the main function module also includes the data processing submodule, the data processing submodule includes multiple function submodules, each function submodule is used to carry a preset algorithm, and when each function submodule becomes the current main function module, the preset algorithm is imported into the monitoring software platform, the target signal is processed and analyzed, and the partial discharge situation is determined.

[0015] As an optional implementation, the data processing submodule includes a wavelet denoising submodule, and the wavelet denoising submodule is used to determine a preset wavelet threshold function according to a generalized cross-validation method, and perform a wavelet transform on the target signal to obtain a target denoised signal;

[0016] The generalized cross-validation method is determined by a first formula, and the wavelet threshold function is determined by a second formula. The first formula includes:

[0017]

[0018] in, is the high-frequency wavelet coefficient of the noisy signal, and N is the number of high-frequency wavelet coefficients; is the number of wavelet coefficients with the threshold set to 0; j is the number of decomposition levels; is the kth wavelet coefficient at the jth scale of the decomposition, that is, the wavelet threshold function;

[0019] And, the second formula includes:

[0020]

[0021] in, is the adjustment factor; N is the number of high-frequency wavelet coefficients; is the number of wavelet coefficients with the threshold set to 0; j is the number of decomposition levels; is the kth wavelet coefficient at the jth scale of decomposition; for estimated wavelet coefficients, i.e., the wavelet threshold function; is the threshold value; is a symbolic function.

[0022] As an optional implementation, the data processing submodule further includes a fault identification submodule, which is used to determine a target eigenvalue after decomposing the target noise reduction signal according to a set empirical mode decomposition method, and determine a mixed eigenvector according to the target eigenvalue, and determine a discharge type of the target transformer by a least squares support vector machine, wherein the discharge type includes tip discharge, suspension discharge, surface discharge or bubble discharge;

[0023] Furthermore, the data processing submodule further includes a fault location submodule, which is used to process the target noise reduction signal, determine the local discharge delay according to a preset energy accumulation function through an arrival time difference method, and further determine the local discharge source coordinate position.

[0024] In a second aspect, the present application provides an ultrasonic-based partial discharge analysis method for an oil-immersed transformer, the method comprising:

[0025] Acquiring a partial discharge ultrasonic signal of the target transformer;

[0026] Processing the partial discharge ultrasonic signal to obtain an amplified signal in a target frequency band as a target signal;

[0027] A partial discharge condition of the target transformer is determined according to the target signal.

[0028] In a third aspect, the present application provides an ultrasonic-based oil-immersed transformer partial discharge analysis device, the device comprising:

[0029] An acquisition module, used for acquiring a partial discharge ultrasonic signal of the target transformer;

[0030] A processing module, used for processing the partial discharge ultrasonic signal to obtain an amplified signal in a target frequency band as a target signal;

[0031] The processing module is further used to determine the partial discharge condition of the target transformer according to the target signal.

[0032] In a fourth aspect, the present application provides an electronic device, characterized in that the electronic device includes a memory and a processor, the memory stores at least one program, and the at least one program is executed by the processor to implement the method described in the second aspect.

[0033] In a fifth aspect, the present application provides a computer-readable storage medium, wherein at least one program is stored in the storage medium, and the at least one program is executed by a processor to implement the method described in the second aspect.

[0034] In a sixth aspect, the present application provides a computer program product, comprising a computer program, characterized in that when the computer program is executed by a processor, the steps of the method described in the second aspect are implemented.

[0035] The ultrasonic-based oil-immersed transformer partial discharge analysis system and method provided in the present application obtains highly sensitive partial discharge ultrasonic signals within a preset frequency band through four piezoelectric ultrasonic sensors, which can improve the effectiveness of the partial discharge ultrasonic signals, and then filters and amplifies the partial discharge ultrasonic signals through a signal processing module to obtain a target signal, further improving the effectiveness of the signal, and analyzes the target signal through a monitoring platform module equipped with various functional modules, and determines the discharge status of the current target transformer through a preset algorithm, thereby improving the effectiveness of the signal, thereby improving the effectiveness and real-time performance of the partial discharge analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram of the structure of an ultrasonic-based partial discharge analysis system for oil-immersed transformers provided in an embodiment of the present application;

[0037] Figure 2A schematic diagram of the structure of another ultrasonic-based partial discharge analysis system for oil-immersed transformers provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of the structure of another ultrasonic-based partial discharge analysis system for oil-immersed transformers provided in an embodiment of the present application;

[0039] Figure 4 A schematic flow chart of a partial discharge analysis method for oil-immersed transformers based on ultrasound provided in an embodiment of the present application;

[0040] Figure 5 A schematic flow chart of another ultrasonic-based partial discharge analysis method for oil-immersed transformers provided in an embodiment of the present application;

[0041] Figure 6 A schematic flow chart of another ultrasonic-based partial discharge analysis method for oil-immersed transformers provided in an embodiment of the present application;

[0042] Figure 7 A schematic flow chart of another ultrasonic-based partial discharge analysis method for oil-immersed transformers provided in an embodiment of the present application;

[0043] Figure 8 A schematic flow chart of another ultrasonic-based partial discharge analysis method for oil-immersed transformers provided in an embodiment of the present application;

[0044] Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The transformer is one of the most core equipment in the power system and traction power supply system. It bears the heavy responsibility of voltage level conversion, and its operating status directly affects the safety and stability of the entire system. Therefore, ensuring the reliability of the transformer is of great significance to the safe operation of the power system and traction power supply system. The reliability of the transformer mainly depends on the state of its insulation system. The deterioration of the insulation will affect the overall health of the transformer. If it is not handled, these deteriorations will directly affect the service life of the transformer and even cause serious accidents such as major economic losses or casualties. Partial discharge analysis is an important technology for monitoring the insulation status of transformers. Most catastrophic insulation breakdowns are developed from the cumulative effect of long-term partial discharge. Therefore, it is necessary to detect partial discharge in time and determine its fault type and discharge location to facilitate subsequent fault repair.

[0046] At present, the commonly used partial discharge detection technologies include current pulse method, ultra-high frequency method and ultrasonic method. Among them, the current pulse method is easily affected by high-frequency current noise, and the ultra-high frequency method is easily affected by strong electromagnetic interference in field applications, which reduces the online application performance of the ultra-high frequency positioning method. The ultrasonic method has high economic and portability because of its small sensor size, low cost, no need for pre-opening, and easy to move. However, ultrasonic signals are not only affected by Gaussian white noise, but also by narrowband noise, most of which are distributed in the low-frequency domain.

[0047] First, the internal structure of the transformer is complex. When ultrasonic waves propagate in different media inside the transformer, they will produce refraction, reflection and attenuation, and will be mixed with a large amount of Gaussian white noise interference and narrowband interference, which may cause the information carried by the ultrasonic signal received by the sensor to be blurred or have too large errors. Ultrasonic identification of the type of partial discharge of the transformer requires the acquisition of multiple partial discharge characteristic parameters. Large noise interference will drown the originally weak partial discharge signal, resulting in the extracted characteristic parameters not being able to fully reflect the characteristics of the partial discharge signal. In addition, the ultrasonic positioning method must obtain the time delay of each ultrasonic signal collected. If the noise is too large, it will be difficult to determine the arrival time of the ultrasonic wave, resulting in inaccurate delay estimation. Therefore, selecting a suitable filtering and denoising method is the key to accurately identifying the type of partial discharge and locating the coordinate position of the partial discharge source.

[0048] Secondly, the time difference equation cannot be solved by simple numerical calculation, so it needs to be solved in combination with an optimization algorithm. The effects of different optimization algorithms vary greatly, which has a great impact on the number of iterations and positioning accuracy. Some iterative methods require a good initial value guess. If the initial value is not appropriate, it may not converge and lead to positioning failure. Therefore, a positioning method that meets the positioning needs is urgently needed.

[0049] Finally, the commonly used transformer offline detection technology currently requires that when the transformer is out of system operation, relevant equipment is used to detect some typical physical quantities that affect the insulation of the distribution transformer, such as insulation resistance, transformer ratio, etc., and the insulation status is judged in combination with relevant regulations and historical data. This method has poor timeliness and cannot detect partial discharge in time.

[0050] The above technical problems will lead to a decrease in the accuracy of ultrasonic detection of partial discharge generation, identification of partial discharge type and positioning of partial discharge source coordinates.

[0051] Based on this, it is necessary to provide an ultrasonic-based partial discharge analysis system and method for oil-immersed transformers for the above-mentioned technical problems.

[0052] The technical concept of the present application is to obtain a highly sensitive local discharge ultrasonic signal within a preset frequency band through four piezoelectric ultrasonic sensors, thereby improving the effectiveness of the local discharge ultrasonic signal, and then filter and amplify the local discharge ultrasonic signal through a signal processing module to obtain a target signal, thereby further improving the effectiveness of the signal, and analyze the target signal through a monitoring platform module equipped with various functional modules, and determine the discharge status of the current target transformer through a preset algorithm, thereby improving the effectiveness of the signal, thereby improving the effectiveness and real-time nature of the local discharge analysis.

[0053] To further illustrate the various embodiments, the present invention provides drawings. These drawings are part of the disclosure of the present invention, which are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, a person of ordinary skill in the art should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0054] The present invention will now be further described with reference to the accompanying drawings and specific implementation methods.

[0055] Embodiment 1:

[0056] See also Figure 1 , Figure 1 A schematic diagram of the structure of an ultrasonic-based partial discharge analysis system for oil-immersed transformers provided in an embodiment of the present application. The system includes:

[0057] Signal acquisition module, signal processing module and monitoring platform module;

[0058] The signal acquisition module includes four piezoelectric ultrasonic sensors installed at preset positions of the target transformer, the signal processing module includes a filter amplifier circuit and a digital-to-analog conversion device, the filter amplifier circuit includes a filter subcircuit, an amplifier subcircuit and a zero adjustment subcircuit, the monitoring platform module includes a monitoring device, and the monitoring device is equipped with a monitoring software platform, the signal acquisition module is connected to the signal processing module, and the signal processing module is connected to the monitoring platform module;

[0059] Among them, three piezoelectric ultrasonic sensors are arranged on the front of the target transformer box in a magnetic way, and the fourth piezoelectric ultrasonic sensor is arranged on the back of the target transformer box in a magnetic way, and the four piezoelectric ultrasonic sensors form a regular tetrahedron;

[0060] The signal acquisition module is used to obtain the partial discharge ultrasonic signal of the target transformer, the signal processing module is used to process the partial discharge ultrasonic signal to obtain the amplified signal in the target frequency band as the target signal, and the monitoring platform module is used to determine the partial discharge condition of the target transformer according to the target signal.

[0061] See also Figure 2 , Figure 2 This is a schematic diagram of another ultrasonic-based partial discharge analysis system for oil-immersed transformers provided in an embodiment of the present application. Three piezoelectric ultrasonic sensors are magnetically arranged on the front of the target transformer box. Figure 3 , Figure 3 A schematic structural diagram of another ultrasonic-based oil-immersed transformer partial discharge analysis system provided in an embodiment of the present application, wherein a fourth piezoelectric ultrasonic sensor is magnetically disposed on the back of the target transformer box, and the four piezoelectric ultrasonic sensors form a regular tetrahedron.

[0062] Specifically, as a non-destructive testing method, ultrasonic testing has no electrical connection with high-voltage equipment and is not affected by electromagnetic interference. It can realize partial discharge fault identification and accurately locate the coordinates of the partial discharge source.

[0063] Since there are many types of partial discharge inside the transformer, their frequency spectrum is widely distributed, but most of them are concentrated in the range of 20kHz to 150kHz. Therefore, Pengxiang Technology PXR04 sensor is selected, which has a resonant frequency of 40kHz and a frequency bandwidth of 30kHz to 140kHz, which can basically cover the partial discharge ultrasonic signal frequency band and has high sensitivity. When the signal source strength remains unchanged, the higher the sensitivity of the sensor, the greater the output voltage signal amplitude, which is conducive to subsequent signal processing.

[0064] Four piezoelectric ultrasonic sensors are magnetically arranged on the front and back sides of the transformer box to receive ultrasonic signals generated by partial discharge of the oil-immersed transformer. Ultrasonic sensors 1, 2, and 3 are magnetically placed on the front side of the transformer box shell and are distributed in an equilateral triangle. Ultrasonic sensor 4 is magnetically placed at the center point of the back side shell of the transformer box. The four ultrasonic sensors form a regular tetrahedron. This arrangement method has good directivity for ultrasonic reception.

[0065] See also Figure 4 , Figure 4A flow chart of a method for analyzing partial discharge of an oil-immersed transformer based on ultrasound is provided in an embodiment of the present application, which is used to illustrate the signal flow of a signal processing module. The signal processing module designed in the present application is composed of a filter amplifier circuit and a digital-to-analog conversion device, wherein the filter amplifier circuit is used to receive the original partial discharge ultrasonic signal, the filter amplifier circuit is connected to the partial discharge ultrasonic sensor, and is used for noise reduction and amplification of the original partial discharge ultrasonic signal, and the filter amplifier circuit is connected to the digital-to-analog conversion device to convert the analog signal into a digital signal that can be read by the host computer software.

[0066] The partial discharge ultrasonic signal is input into the filter amplifier circuit through the IPX RF antenna interface. The filter amplifier circuit is composed of a filter subcircuit, an amplifier subcircuit and a zeroing subcircuit. The filter subcircuit is used as the signal receiving front end, from which the original partial discharge ultrasonic signal is input. The filter subcircuit is used to perform bandpass filtering on the partial discharge ultrasonic signal to obtain a target frequency band signal. Among them, the target frequency band is the frequency band corresponding to the partial discharge ultrasonic signal. The input end of the amplifier subcircuit is connected to the output end of the filter subcircuit. The amplifier subcircuit is used to amplify the signal after narrowband filtering to obtain an amplified signal. The amplifier subcircuit can amplify weak partial discharge signals and highlight partial discharge signals. The zeroing subcircuit is used to control the bias voltage of the amplifier chip to play a zeroing role.

[0067] The filter subcircuit consists of a high-pass RC filter and a low-pass RC filter. As an optional implementation, the high-pass cutoff frequency is 20kHz and the low-pass cutoff frequency is 265kHz. The amplifier chip uses AD8421, which has a high gain volume and is easy to control. Under the condition of 100 amplification gain, no attenuation occurs in each frequency band. The zero adjustment subcircuit uses OPA177GS chip and sliding rheostat to achieve circuit zero adjustment. Use IPX interface to input and output ultrasonic signals.

[0068] This application uses a linear voltage-stabilized switching power supply with an output voltage of positive and negative 12V to power the filter amplifier circuit. The power input circuit adopts multi-capacitor parallel voltage stabilization, and the lightning protection diode prevents the power supply from being reversely connected and burning the chip, which can effectively protect the circuit.

[0069] The digital-to-analog conversion device can be realized based on a synchronous acquisition module device with a resolution of 12-bit, 8 analog input channels, 4 analog output channels, 8 digital input channels, and 8 digital output channels. The sampling rate of each channel is 500KSa / s, and the maximum acquisition frequency is 250kHz, which has covered most of the partial discharge ultrasonic signal range. The digital-to-analog conversion device is connected to the monitoring platform of the host computer, and the collected and processed data is uploaded to the monitoring platform. The data processing module in the monitoring platform determines whether partial discharge occurs, as well as the type of partial discharge and the coordinate position of the partial discharge source.

[0070] As an optional implementation, the monitoring software platform includes a data reading submodule, a data display submodule and a main function module;

[0071] Wherein, the data reading submodule is used to collect the target signal according to the preset collection frequency, the data display submodule is used to display the filtered target signal in real time on the monitoring software platform, or transmit the filtered target signal to an external oscilloscope, and display the filtered target signal in real time on the external oscilloscope, and the main function module includes a partial discharge alarm submodule, and the partial discharge alarm submodule is used to carry the energy mean criterion algorithm, analyze the target signal, determine the partial discharge situation and issue an alarm;

[0072] In addition, the monitoring software platform also includes a digital filtering submodule or a data storage submodule, which is used to perform corresponding functions according to the operation status of the monitoring software platform. The digital filtering submodule is used to perform high-pass and low-pass filtering on the target signal, and the data storage submodule is used to save the target signal.

[0073] The target signal of preset frequency can be obtained through the data reading submodule installed on the monitoring software platform. The raw data can be visualized on a virtual interface or an external oscilloscope through the data display submodule. The target signal is analyzed by the energy mean algorithm through the partial discharge submodule, and the partial discharge situation is determined thereby, which improves the validity of the target signal and improves the validity and real-time performance of the partial discharge analysis.

[0074] As an optional implementation, the monitoring software platform further includes a function jump submodule and a data processing submodule;

[0075] Among them, the function jump submodule is used to select the main function module, the main function module also includes the data processing submodule, the data processing submodule includes multiple function submodules, each function submodule is used to carry a preset algorithm, and when each function submodule becomes the current main function module, the preset algorithm is imported into the monitoring software platform, the target signal is processed and analyzed, and the partial discharge situation is determined.

[0076] The currently displayed signal and the corresponding function to be executed are determined by the function jump submodule, and the current problem of low data loading efficiency is improved by loading the algorithm onto the signal data. In addition, the data processing submodule includes an algorithm designed based on the target signal characteristics and required functions, thereby improving the adaptability of the target signal, application scenario and functional algorithm, and improving the efficiency of data processing.

[0077] Specifically, as an example, the ultrasonic monitoring platform for partial discharge of oil-immersed transformers can be written using Labview software, including a data reading submodule, a digital filtering submodule, a data display submodule, a data storage submodule, a partial discharge alarm submodule and a function jump submodule, and can respectively select whether to enable the data storage and filtering functions.

[0078] like Figure 5 As shown, Figure 5 It is another flow chart of the partial discharge analysis method of oil-immersed transformer based on ultrasonic wave provided by the embodiment of the present application, which is used to illustrate the application process of the data reading submodule. The front end of the data reading submodule is connected to the digital-to-analog conversion device, and the communication between the digital-to-analog conversion device and the host computer software is completed by calling the dynamic link library of the corresponding acquisition card. The number of sampling channels and the range of each channel can be set in this module. There are two acquisition modes for analog input, continuous acquisition and limited number of acquisition, which can realize long-term continuous acquisition and timed acquisition. The sampling frequency of the digital-to-analog conversion device can be set, and the maximum sampling frequency is 500kHz. After setting the analog input trigger source, analog input sampling clock, analog input pre-trigger source, analog input pre-trigger point number, and analog input software of the digital-to-analog conversion device, data collection begins. Before starting the acquisition, a data cache is cleaned up to ensure that there is no interference from the cached data of the last sampling at the beginning of this sampling. When the sampling is completed and the sampling cycle is exited, the cached data is cleared and the acquisition card is closed.

[0079] The digital filter submodule is used to perform high-pass and low-pass filtering on the read data to remove power supply noise interference. This module processes the collected signal using a 4th-order Butterworth filter to eliminate low-frequency and high-frequency noise, which can effectively remove power supply noise interference.

[0080] The data display submodule consists of eight oscilloscopes, four of which are used to display the read four-channel partial discharge ultrasonic signals, and the remaining four oscilloscopes are used to display the corresponding spectrum of the four-channel partial discharge ultrasonic signals. Two tab controls are used to distinguish the eight oscilloscopes, each of which contains four tabs, and each tab contains an oscilloscope. Therefore, different oscilloscopes are displayed by selecting different tabs, realizing the visualization of partial discharge ultrasonic signals. Users can freely select the oscilloscope to be displayed, and can more clearly understand the current operating status of the oil-immersed transformer.

[0081] It should be noted that in the system provided in the present application, the oscilloscope may be a virtual oscilloscope device or an actual oscilloscope device connected to some devices in the system, which should be broadly understood by those skilled in the art.

[0082] The data storage submodule is used to save data. Before opening the data storage function, you need to select the file save location and name the file. The data is saved in the current system file in the TDMS format. The saved data includes the sampling point time, data amplitude, sampling data interval, data length, and channel name. A single file can save partial discharge ultrasonic data within 2 seconds at most. After this time, the module will automatically generate the next file. The file name will add the current time and suffix number after the named file name to distinguish files at different times.

[0083] The partial discharge alarm submodule uses the energy mean criterion to identify the occurrence of partial discharge. The energy mean criterion is shown in Formula 1. The partial discharge ultrasonic threshold voltage value is calculated by this criterion. Signals exceeding the threshold voltage value are regarded as partial discharge signals, and signals less than the threshold voltage value are regarded as background noise. The length of a signal is at least 2ms. If a signal greater than the threshold voltage value appears again within 2ms, the two signals will be regarded as the same signal. In one case, when the system is running, the current threshold voltage value will be calculated at all times. When a signal higher than the threshold value appears, the system will alarm with a red light and record the time. At the same time, the system will also display how many partial discharges have occurred during the operation. The staff can manually check whether the data at that moment is abnormal.

[0084]

[0085] In the formula, is the partial discharge ultrasonic signal, k is the signal length, ave represents the mean, and abs represents the absolute value.

[0086] The function jump submodule consists of multiple buttons, which can jump between the wavelet noise reduction submodule, fault identification submodule and fault location submodule, jump from the monitoring platform to any of the three submodules, jump from any submodule to the monitoring platform, and jump between submodules. It is convenient for users to quickly switch to the required function interface.

[0087] In actual use, users can choose whether to turn on the digital filter submodule. The system defaults to a high-pass cutoff frequency of 1kHz and a low-pass cutoff frequency of 230kHz. If you choose to turn it on, you need to set the appropriate high-pass cutoff frequency and low-pass cutoff frequency to meet the requirements. You can also choose to turn on the sampling channel and set the sampling frequency and reading speed. A maximum of 4 sampling channels can be turned on, and the maximum acquisition frequency is 500kHz. The recommended reading speed under four channels is 200,000. At this speed, the waveform displayed by the oscilloscope is relatively smooth. The platform can also display the current number of cache points and the number of acquisition cycles. The current number of cache points represents the number of data cached in the system cache at this time. This value is related to the reading speed and the smoothness of the system operation. If the reading speed is low or the system is stuck, the number of cache points will surge. In order to ensure that the displayed partial discharge ultrasonic waveform has a high timeliness, the number of cache points should be kept at a low level as much as possible. At the same time, you can choose whether to turn on the data storage submodule. The storage file name needs to be set before turning on the module. If you want to view the partial discharge ultrasonic data waveforms of different channels, you can freely switch the currently displayed oscilloscope through the tab.

[0088] As an optional implementation, the data processing submodule includes a wavelet denoising submodule, which is used to determine a preset wavelet threshold function according to a generalized cross-validation method, and perform a wavelet transform on the target signal to obtain a target denoised signal.

[0089] Through the wavelet transform method, according to the generalized cross-validation method, a preset wavelet threshold function is determined, and the target signal is subjected to wavelet transform to obtain a target noise reduction signal. The target noise reduction signal contains a higher proportion of effective information, thereby improving the effectiveness of signal processing, thereby improving the effectiveness and real-time performance of partial discharge analysis.

[0090] As an optional implementation, the data processing submodule further includes a fault identification submodule, which is used to determine a target eigenvalue after decomposing the target noise reduction signal according to a set empirical mode decomposition method, and determine a mixed eigenvector according to the target eigenvalue, and determine a discharge type of the target transformer by a least squares support vector machine, wherein the discharge type includes tip discharge, suspension discharge, surface discharge or bubble discharge;

[0091] Furthermore, the data processing submodule further includes a fault location submodule, which is used to process the target noise reduction signal, determine the local discharge delay according to a preset energy accumulation function through an arrival time difference method, and further determine the local discharge source coordinate position.

[0092] The target noise reduction signal is processed by the fault identification submodule and the fault location submodule to obtain the corresponding discharge type and discharge coordinates. In addition, visualization processing can be realized, which improves the practicality of the solution and improves the effectiveness and real-time performance of partial discharge analysis.

[0093] Specifically, the data processing module includes a wavelet denoising submodule, a fault identification submodule, and a fault location submodule. These three submodules realize rapid data processing by loading the algorithm into Labview. The traditional data processing method is to import the saved data into Matlab for calculation, but the data waveform cannot be viewed during data import, which is very intuitionistic. Often, all data can only be imported at once, which will result in a very long import time and troublesome subsequent waveform viewing. The advantage of the method proposed in this design is that it can process data faster, and the partial discharge ultrasonic signal that needs to be processed can be intercepted through the oscilloscope tool, which saves the process of importing data into Matlab and saves time.

[0094] like Figure 6 As shown, Figure 6 This is another flow chart of an ultrasonic-based partial discharge analysis method for oil-immersed transformers provided in an embodiment of the present application, which is used to illustrate the corresponding functions and data processing flow of the wavelet denoising submodule. The wavelet denoising submodule includes two side-by-side oscilloscopes, a data reading submodule, a data storage submodule, and an improved wavelet threshold algorithm, wherein the left oscilloscope displays the undenoised partial discharge ultrasonic signal, and the right oscilloscope displays the denoised partial discharge ultrasonic signal.

[0095] Since Butterworth filter and RC filter are bandpass filters, they can only eliminate low-frequency and high-frequency noise, and the noise signal mixed in the partial discharge signal also includes Gaussian white noise, which is distributed in the full frequency band and cannot be removed by the above method. Therefore, an improved wavelet threshold algorithm is proposed for noise reduction. This method can effectively remove the Gaussian white noise mixed in the partial discharge signal.

[0096] The core idea of ​​the wavelet denoising algorithm is to use wavelet transform to decompose the signal into wavelet coefficients of different scales and frequencies, and then remove the noise in the signal by processing these coefficients while retaining the useful information of the signal as much as possible. The steps of wavelet threshold denoising are:

[0097] 1) Select an appropriate mother wavelet and decomposition layer number to perform wavelet transform and generate transformed wavelet coefficients;

[0098] 2) Set the threshold of each layer and process the wavelet coefficients of each layer using the threshold function;

[0099] 3) Use inverse wavelet transform to restore the denoised signal.

[0100] It can be seen that the key point of wavelet denoising algorithm is to obtain an accurate threshold and a wavelet threshold function with excellent performance. Therefore, this method proposes an improved wavelet threshold algorithm.

[0101] Among them, the improved wavelet threshold algorithm proposes a wavelet threshold determination method based on generalized cross validation (GCV), the formula is shown in the first formula. This method avoids dependence on the characteristics of the noise signal and can effectively restore the pure signal without the evaluation of the characteristics of the noise signal. A new threshold function between the soft threshold function and the hard threshold function is proposed. The function formula is shown in the second formula. This threshold function overcomes the shortcomings of low reconstruction accuracy of the signal processed by the soft threshold function and the oscillation phenomenon of the signal processed by the hard threshold function.

[0102] The generalized cross-validation method is determined by a first formula, and the wavelet threshold function is determined by a second formula. The first formula includes:

[0103]

[0104] in, is the high-frequency wavelet coefficient of the noisy signal, and N is the number of high-frequency wavelet coefficients; is the number of wavelet coefficients with the threshold set to 0; j is the number of decomposition levels; is the kth wavelet coefficient at the jth scale of the decomposition, that is, the wavelet threshold function;

[0105] And, the second formula includes:

[0106]

[0107] in, is the adjustment factor; N is the number of high-frequency wavelet coefficients; is the number of wavelet coefficients with the threshold set to 0; j is the number of decomposition levels; is the kth wavelet coefficient at the jth scale of decomposition; for estimated wavelet coefficients, i.e., the wavelet threshold function; is the threshold value; is a symbolic function.

[0108] In one scheme, the user opens a data file in the wavelet denoising submodule, uses the oscilloscope tool to intercept the required undenoised partial discharge ultrasonic waveform in the data, clicks the start denoising button, and the system executes the improved wavelet threshold denoising algorithm, selects the appropriate wavelet basis function according to the nature and spectral characteristics of the signal, determines the number of decomposition layers n according to the frequency information and noise level contained in the signal, uses the wavelet decomposition algorithm to decompose the signal and generate a series of wavelet coefficients including 1 low-frequency wavelet coefficient and n-1 high-frequency wavelet coefficients, and then uses the particle swarm optimization algorithm to calculate the different thresholds. minimum value, by using the minimum Estimation of function value and determination of optimal threshold , using the threshold And the optimized wavelet threshold function, process all high-frequency wavelet coefficients to obtain the high-frequency wavelet coefficients after noise reduction, and finally use the high-frequency wavelet coefficients and low-frequency wavelet coefficients after noise reduction to perform wavelet inverse transformation reconstruction to obtain the denoised partial discharge ultrasonic signal. The denoised partial discharge ultrasonic signal is displayed on the oscilloscope on the right, and the threshold of this noise reduction can also be displayed.

[0109] like Figure 7 As shown, Figure 7 This is another flow chart of an ultrasonic-based partial discharge analysis method for oil-immersed transformers provided in an embodiment of the present application, which is used to illustrate the data flow and corresponding functions of the fault identification submodule. The fault identification submodule includes an oscilloscope, an identification result display window, a data reading submodule, a function jump submodule, and an improved wavelet threshold algorithm and an EEMD-LSSVM fault identification algorithm.

[0110] The fault identification submodule constructs an identification model according to a set of features, and determines the target discharge type of the oil-immersed transformer equipment through the identification model, wherein the target discharge type includes one of the following: tip discharge, suspension discharge, surface discharge, and bubble discharge.

[0111] The fault identification submodule analyzes the data processed by the wavelet denoising submodule, and can determine whether a partial discharge fault occurs in the oil-immersed transformer equipment according to the target data set. When a partial discharge fault is determined to occur, the target discharge type of the oil-immersed transformer equipment can also be determined. Therefore, the purpose of timely determining whether a partial discharge fault occurs in the oil-immersed transformer equipment and determining the target discharge type of the oil-immersed transformer equipment can be achieved, that is, the purpose of timely discovering the insulation problems that may exist in the oil-immersed transformer equipment and the purpose of accurately diagnosing the insulation problems that occur can be achieved, which can avoid the expansion of insulation defects and thus the problem of insulation breakdown, and achieve the effect of improving the efficiency of partial discharge detection of oil-immersed transformer equipment.

[0112] The fault identification submodule uses ensemble empirical mode decomposition (EEMD) to decompose the partial discharge ultrasonic signal after wavelet denoising into multiple layers of intrinsic mode components (IMF). This method is a time-frequency analysis method and an adaptive time-frequency localization analysis method. The decomposed IMF is related to the sampling frequency and will change based on the data itself, getting rid of the limitations of Fourier transform and having unique advantages in preprocessing nonlinear and non-stationary signals such as partial discharge ultrasonic signals. Compared with empirical mode decomposition (EMD), this method uses the statistical characteristics of the uniform distribution of Gaussian white noise spectrum to add different white noises to the original signal, so that the signal has continuity at different scales, and then performs EMD decomposition on the signal to eliminate modal aliasing.

[0113] For the IMF components obtained by decomposition, they are reduced to eliminate the adverse effects caused by inconsistent signal sizes due to different discharge degrees. The maximum amplitude of each IMF after reduction is calculated and the layer number where the maximum amplitude is found, the maximum energy of each IMF after reduction is calculated and the layer number where the maximum energy is found, the maximum energy entropy of each IMF after reduction is calculated and the layer number where the maximum energy entropy is found, the minimum envelope entropy of each IMF after reduction is calculated and the layer number where the minimum envelope entropy is found, the minimum entropy of each IMF after reduction is calculated and the layer number where the minimum original entropy is found.

[0114] Eleven characteristic parameters, including the number of decomposition layers, IMF amplitude, maximum value of reduced amplitude, the number of layers where the maximum value of reduced amplitude is located, maximum value of reduced energy, the number of layers where the maximum value of reduced energy is located, maximum value of energy entropy, the number of layers where the maximum value of energy entropy is located, minimum value of envelope entropy, the number of layers where the minimum value of envelope entropy is located, minimum value of sample entropy, and the number of layers where the minimum value of sample entropy is located, are selected to form a new hybrid characteristic vector (ME), and the hybrid characteristic vector is used as a training sample to train the least squares support vector machine (LSSVM). The fault identification submodule will identify the type of partial discharge according to the hybrid characteristic vector of the partial discharge ultrasonic signal that needs to be identified.

[0115] In one scheme, the user opens a data file in the fault identification submodule. If the non-noise-reduced file is selected, the oscilloscope tool is used to intercept the required non-noise-reduced partial discharge ultrasonic waveform in the data, click the identification button and check to start the noise reduction function. The system will first execute the improved wavelet threshold noise reduction algorithm and then extract the characteristic value of the noise-reduced data and classify it according to the fault identification model to obtain the partial discharge type; if the noise-reduced file is selected, the identification button is directly clicked, and the system will extract the characteristic value of the data and classify it according to the fault identification model to obtain the partial discharge type. Then, it is known when and what type of fault the transformer has, which plays a very critical role in the safe operation of the transformer.

[0116] like Figure 8 As shown, Figure 8 This is another flow chart of an ultrasonic-based partial discharge analysis method for oil-immersed transformers provided in an embodiment of the present application, which is used to illustrate the data flow and corresponding functions of the fault location submodule. The fault location submodule includes four oscilloscopes, a positioning model display window, a data reading submodule, a function jump submodule, and an improved wavelet threshold algorithm and an E-CSBO positioning algorithm. The four oscilloscopes are used to display four-channel partial discharge ultrasonic signals, and the positioning model display window is used to display the three-dimensional model of the transformer and the location of the discharge source.

[0117] This module calculates the coordinate position of the local discharge source based on the time difference of arrival method (TDOA). The TDOA technology is mainly divided into two parts. The first is to obtain the time difference between the arrival of the local discharge ultrasonic signal at each sensor, and then solve the hyperbolic time difference equation group according to the time difference and the sensor coordinates, and finally obtain the coordinate position of the local discharge source.

[0118] Among them, the energy accumulation method is used to obtain the time difference. The traditional energy accumulation method is to calculate the energy accumulation curves of the local discharge ultrasonic signals after noise reduction of the four channels respectively, calculate the inflection point of the energy accumulation curve, and regard the inflection point as the starting time of the local discharge signal. Finally, the time difference is obtained by subtracting the starting time.

[0119] If the total number of signal sampling points is N, then The cumulative energy function at the moment is defined as:

[0120]

[0121] Where: Indicated in The signal amplitude at time is the scaling factor, which scales the cumulative energy curve proportionally without affecting the performance of the method; is the average energy of the ultrasonic signal, defined as:

[0122]

[0123] However, since the main energy of the signal is concentrated after the arrival of the ultrasonic wave, the power of the ultrasonic wave at the moment of arrival is less than the average power of the entire signal, resulting in the minimum value of the cumulative energy function appearing slightly later than the actual and accurate signal arrival time. If the amplitude of the signal waveform increases significantly at the moment of arrival, the calculation delay of the arrival time is relatively small; but if the signal increases slowly at the moment of arrival, the calculation delay of the arrival time may be large.

[0124] Therefore, an improved energy accumulation method is proposed. This method first normalizes the obtained energy accumulation curve to obtain an energy accumulation curve between [-1, 1], which can eliminate the influence of different discharge intensities. Then, the second-order derivative of the normalized energy accumulation curve is solved to obtain the concavity of the normalized energy accumulation curve. When the second-order derivative is greater than 0.004, it is regarded as the starting time of local discharge. This method has higher sensitivity to the degree of change of the energy accumulation curve, can quickly discover the turning point of the energy accumulation curve, and is more accurate in determining the starting time of local discharge.

[0125] TDOA locates the local discharge source based on the arrival time difference of ultrasonic signals propagating to different sensors. The coordinates of the ultrasonic sensors are , the ultrasonic signal reaches the The time for each sensor is , the local discharge source coordinates to be determined are , for sensor 1 and sensor 2, the hyperbolic surface with the coordinates of the two sensors as the focus and the distance difference between the sensors as the major axis is as follows:

[0126]

[0127] Where: It is the arrival time difference between the signals received by sensor 1 and sensor 2.

[0128] The TDOA values ​​measured by multiple ultrasonic sensors can be used to construct a hyperbolic time difference equation system. The solution of this nonlinear equation system is the local discharge source coordinates:

[0129]

[0130] The above equations are transformed into the following constrained optimization problem:

[0131]

[0132] Where: is the objective function; are the equivalent transformer length, width and height respectively; i=2, 3, 4; The unit is The location of the partial discharge source is obtained by solving the spherical equations, and the optimization algorithm is used to solve the approximate solution. This method uses the circulatory system based optimization algorithm (CSBO) to solve the above problem. This optimization algorithm is a meta-heuristic optimization algorithm designed to imitate the circulation process of blood through blood vessels in the human body. This algorithm draws on the functions of the two main circulations in the blood circulation system - pulmonary circulation and systemic circulation, to guide the optimization process in the algorithm.

[0133] The basic calculation process of the CSBO algorithm is:

[0134] (1) Initialization: Randomly generate the initial population. Each individual (blood drop) in the population represents a potential solution to the optimization problem, and its position represents a point in the search space. (2) Calculate fitness: Calculate the fitness (objective function value) of each individual in the population, which is equivalent to the flow of blood in the body. (3) Pulmonary circulation: Send a part of individuals with poor fitness (equivalent to deoxygenated blood) into the "pulmonary circulation" and update their positions through a random process in the hope of obtaining a better solution. (4) Systemic circulation: Send individuals with better fitness (equivalent to oxygen-rich blood) into the "systemic circulation" and update their positions by simulating the flow of blood in the body to further optimize the solution. (5) Iterative update: In each iteration, the positions of individuals in the population are updated according to a specific mathematical model. This process involves calculating new positions and fitness, and may accept or reject these new solutions depending on whether they improve the current optimal solution. (6) Termination condition: The algorithm stops when the preset number of iterations is reached or other termination conditions are met. (7) Output optimal solution: Return the optimal solution or optimal fitness value found in the iterative process.

[0135] Compared with particle swarm optimization algorithm, whale swarm optimization algorithm and ant colony optimization algorithm, CSBO has excellent search accuracy and speed, and is more accurate in solving the coordinate position of local discharge source.

[0136] In order to realize the visualization of fault location, this scheme designs a three-dimensional model display software to display the specific location of the local discharge source. The software is programmed with Labview. By loading the oil-immersed transformer model and the calculated local discharge source coordinate position, the position relationship between the local discharge source and the transformer is displayed, which is convenient for maintenance personnel to find the transformer fault location more conveniently and quickly, and speed up the troubleshooting progress.

[0137] In one scheme, the user opens a data file in the fault location submodule. If the non-noise reduction file is selected, the oscilloscope tool is used to intercept the required non-noise reduction partial discharge ultrasonic waveform in the data. Click the positioning button and check the start noise reduction function. The system will first perform the improved wavelet threshold noise reduction algorithm on one channel, calculate the energy accumulation curve of the noise-reduced data and perform normalization processing, and then solve the second-order derivative to obtain the moment when the second-order derivative is greater than 0.004, and determine this moment as the starting moment of partial discharge. The same method is used to process the data of the remaining three channels to obtain the time difference of the starting moment of partial discharge between the four channels. The CSBO algorithm is used to solve the coordinate position of the partial discharge source based on the time difference and the sensor coordinates, and the transformer model and the partial discharge source are displayed in the positioning model display window. This method can enable operators to more clearly understand the location of partial discharge, which is of great help for subsequent maintenance.

[0138] This embodiment uses four piezoelectric ultrasonic sensors to obtain high-sensitivity local discharge ultrasonic signals within a preset frequency band, which can improve the effectiveness of the local discharge ultrasonic signals. The local discharge ultrasonic signals are then filtered and amplified by the signal processing module to obtain the target signal, which further improves the effectiveness of the signal. The target signal is analyzed by the monitoring platform module equipped with various functional modules, and the discharge status of the current target transformer is determined by a preset algorithm, thereby improving the effectiveness of the signal, thereby improving the effectiveness and real-time performance of the local discharge analysis.

[0139] Embodiment 2:

[0140] The present application also provides an ultrasonic-based partial discharge analysis method for oil-immersed transformers, which is applied to the above-mentioned system, and the method comprises:

[0141] Acquiring a partial discharge ultrasonic signal of the target transformer;

[0142] Processing the partial discharge ultrasonic signal to obtain an amplified signal in a target frequency band as a target signal;

[0143] A partial discharge condition of the target transformer is determined according to the target signal.

[0144] It should be noted that the methods implemented by the functional modules in the corresponding systems described above and their corresponding implementations should also fall within the protection scope of the methods provided in this application.

[0145] In addition, the present application also provides an ultrasonic-based virtual device for partial discharge analysis of oil-immersed transformers to implement the method through corresponding program design.

[0146] The device comprises:

[0147] An acquisition module, used for acquiring a partial discharge ultrasonic signal of the target transformer;

[0148] A processing module, used for processing the partial discharge ultrasonic signal to obtain an amplified signal in a target frequency band as a target signal;

[0149] The processing module is further used to determine the partial discharge condition of the target transformer according to the target signal.

[0150] Through four piezoelectric ultrasonic sensors, high-sensitivity local discharge ultrasonic signals within a preset frequency band are obtained, which can improve the effectiveness of the local discharge ultrasonic signals. The local discharge ultrasonic signals are then filtered and amplified by the signal processing module to obtain the target signal, which further improves the effectiveness of the signal. The target signal is analyzed by the monitoring platform module equipped with various functional modules, and the discharge status of the current target transformer is determined by a preset algorithm, thereby improving the effectiveness of the signal, thereby improving the effectiveness and real-time performance of the local discharge analysis.

[0151] Embodiment three:

[0152] The present invention also provides an electronic device, such as Fig. 9 As shown, the electronic device includes a processor 901, a memory 902, a bus 903, and a computer program stored in the memory 902 and executable on the processor 901. The processor 901 includes one or more processing cores. The memory 902 is connected to the processor 901 via the bus 903. The memory 902 is used to store program instructions. When the processor executes the computer program, the steps in the above method embodiment of the first embodiment of the present invention are implemented.

[0153] Further, as an executable solution, the electronic device may be a computer unit, and the computer unit may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The computer unit may include, but is not limited to, a processor and a memory. Those skilled in the art may understand that the composition structure of the above-mentioned computer unit is only an example of a computer unit and does not constitute a limitation on the computer unit, and may include more or less components than the above-mentioned, or a combination of certain components, or different components. For example, the computer unit may also include input and output devices, network access devices, buses, etc., which are not limited in the embodiments of the present invention.

[0154] Further, as an executable solution, the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the computer unit, and various interfaces and lines are used to connect various parts of the entire computer unit.

[0155] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the computer unit by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required for a function; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0156] Embodiment 4:

[0157] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method in the embodiment of the present invention are implemented.

[0158] If the module / unit integrated in the computer unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory) and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0159] Embodiment five:

[0160] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the method described above are implemented.

[0161] Although the present invention has been specifically shown and described in conjunction with the preferred embodiments, it should be understood by those skilled in the art that various changes may be made to the present invention in form and details without departing from the spirit and scope of the present invention as defined by the appended claims, all of which are within the scope of protection of the present invention.

Claims

1. An ultrasonic-based partial discharge analysis system for oil-immersed transformers, characterized in that: The system comprises: Signal acquisition module, signal processing module and monitoring platform module; The signal acquisition module includes four piezoelectric ultrasonic sensors installed at preset positions of the target transformer, the signal processing module includes a filter amplifier circuit and a digital-to-analog conversion device, the filter amplifier circuit includes a filter subcircuit, an amplifier subcircuit and a zero adjustment subcircuit, the monitoring platform module includes a monitoring device, and the monitoring device is equipped with a monitoring software platform, the signal acquisition module is connected to the signal processing module, and the signal processing module is connected to the monitoring platform module; Among them, three piezoelectric ultrasonic sensors are arranged on the front of the target transformer box in a magnetic way, and the fourth piezoelectric ultrasonic sensor is arranged on the back of the target transformer box in a magnetic way, and the four piezoelectric ultrasonic sensors form a regular tetrahedron; The signal acquisition module is used to obtain the partial discharge ultrasonic signal of the target transformer, the signal processing module is used to process the partial discharge ultrasonic signal to obtain an amplified signal in a target frequency band as a target signal, and the monitoring platform module is used to determine the partial discharge condition of the target transformer according to the target signal; Wherein, the monitoring software platform includes a data reading submodule, a data display submodule and a main function module; Wherein, the data reading submodule is used to collect the target signal according to the preset collection frequency, the data display submodule is used to display the filtered target signal in real time on the monitoring software platform, or transmit the filtered target signal to an external oscilloscope, and display the filtered target signal in real time on the external oscilloscope, and the main function module includes a partial discharge alarm submodule, and the partial discharge alarm submodule is used to carry the energy mean criterion algorithm, analyze the target signal, determine the partial discharge situation and issue an alarm; Furthermore, the monitoring software platform further includes a digital filtering submodule or a data storage submodule, wherein the digital filtering submodule or the data storage submodule is used to perform corresponding functions according to the operation status of the monitoring software platform, the digital filtering submodule is used to perform high-pass and low-pass filtering on the target signal, and the data storage submodule is used to save the target signal; Wherein, the monitoring software platform also includes a function jump submodule and a data processing submodule; The function jump submodule is used to select the main function module, the main function module also includes the data processing submodule, the data processing submodule includes multiple function submodules, each function submodule is used to carry a preset algorithm, and when each function submodule becomes the current main function module, the preset algorithm is imported into the monitoring software platform, the target signal is processed and analyzed, and the partial discharge situation is determined; The data processing submodule includes a wavelet denoising submodule, which is used to determine a preset wavelet threshold function according to a generalized cross-validation method, and perform a wavelet transform on the target signal to obtain a target denoised signal; The generalized cross-validation method is determined by a first formula, and the wavelet threshold function is determined by a second formula. The first formula includes: in, is the high-frequency wavelet coefficient of the noisy signal, and N is the number of high-frequency wavelet coefficients; is the number of wavelet coefficients with the threshold set to 0; j is the number of decomposition levels; is the kth wavelet coefficient at the jth decomposition level, that is, the wavelet threshold function; And, the second formula includes: in, is the adjustment factor; N is the number of high-frequency wavelet coefficients; is the number of wavelet coefficients with the threshold set to 0; j is the number of decomposition levels; is the kth wavelet coefficient at the jth decomposition level; for estimated wavelet coefficients, i.e., the wavelet threshold function; is the threshold value; is a symbolic function; Among them, the data processing submodule also includes a fault identification submodule, which is used to decompose the target noise reduction signal into multiple layers of intrinsic mode components according to the collective empirical mode decomposition method, and perform reduction processing on the multiple layers of intrinsic mode components, and respectively determine the maximum amplitude of each intrinsic mode component after reduction and the layer number where the maximum amplitude is located, the maximum energy of each intrinsic mode component after reduction and the layer number where the maximum energy is located, the maximum energy entropy of each intrinsic mode component after reduction and the layer number where the maximum energy entropy is located, the minimum envelope entropy of each intrinsic mode component after reduction and the layer number where the minimum envelope entropy is located. The number of layers, the minimum sample entropy of each intrinsic mode component after reduction, and the number of layers where the original entropy minimum is found, and the number of decomposition layers, the amplitude of the intrinsic mode component, the maximum value of the reduced amplitude, the number of layers where the maximum value of the reduced amplitude is located, the maximum value of the reduced energy, the number of layers where the maximum value of the reduced energy is located, the maximum value of energy entropy, the number of layers where the maximum value of energy entropy is located, the minimum value of envelope entropy, the number of layers where the minimum value of envelope entropy is located, the minimum value of sample entropy, and the number of layers where the minimum value of sample entropy is located are selected to form a mixed feature vector, and the discharge type of the target transformer is determined by a least squares support vector machine, and the discharge type includes tip discharge, suspension discharge, surface discharge or bubble discharge; The data processing submodule further includes a fault location submodule, which is used to process the target noise reduction signal, determine the local discharge delay according to a preset energy accumulation function through an arrival time difference method, and then determine the local discharge source coordinate position; wherein, determining the local discharge delay according to the preset energy accumulation function includes: normalizing the energy accumulation function to obtain an energy accumulation curve between [-1, 1], then solving the second-order derivative of the normalized energy accumulation curve, and determining the local discharge initial time when the second-order derivative is greater than 0.004; The function jump submodule is used to implement jumps among the wavelet denoising submodule, the fault identification submodule and the fault location submodule.

2. An ultrasonic-based oil-immersed transformer partial discharge analysis method using the ultrasonic-based oil-immersed transformer partial discharge analysis system according to claim 1, characterized in that: The method comprises: Acquiring a partial discharge ultrasonic signal of the target transformer; Processing the partial discharge ultrasonic signal to obtain an amplified signal in a target frequency band as a target signal; A partial discharge condition of the target transformer is determined according to the target signal.

3. An ultrasonic-based oil-immersed transformer partial discharge analysis device using the ultrasonic-based oil-immersed transformer partial discharge analysis system according to claim 1, characterized in that: The device comprises: An acquisition module, used for acquiring a partial discharge ultrasonic signal of a target transformer; A processing module, used for processing the partial discharge ultrasonic signal to obtain an amplified signal in a target frequency band as a target signal; The processing module is further used to determine the partial discharge condition of the target transformer according to the target signal.

4. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory stores at least one program, and the at least one program is executed by the processor to implement the method according to claim 2.

5. A computer-readable storage medium, characterized in that: The storage medium stores at least one program, and the at least one program is executed by a processor to implement the method according to claim 2.

6. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to claim 2 are implemented.

Citation Information

Patent Citations

  • Partial discharge signal denoising method based on wavelet adaptive threshold

    CN103576060A

  • Transformer partial discharge denoising method based on synchronous compression wavelet transform domain

    CN110244202A

  • Electrified detection system for nuclear power station oil-immersed transformer

    CN111175620A

  • GIS equipment mechanical fault diagnosis system and method

    CN111638047A

  • Transformer operation state on-line monitoring system and multi-parameter intelligent sensor thereof

    CN112880740A