Hydro-generator ultrahigh frequency partial discharge characteristic signal rapid extraction method and device

Abstract

The embodiment of the invention provides a hydro-generator ultrahigh frequency partial discharge characteristic signal rapid extraction method and device, and relates to the technical field of signal processing. The method comprises the steps of dividing a working scene into a single interference scene and a complex interference scene according to the type or number of interference sources in the working scene where the hydraulic generator is located; determining a static amplitude threshold value corresponding to the single interference scene, or determining a dynamic amplitude threshold value corresponding to the complex interference scene; and scanning a partial discharge signal by adopting a time sliding window, recording a point location greater than the static amplitude threshold or the dynamic amplitude threshold in the time sliding window as a pulse position sample point, and taking a set formed by the pulse position sample points obtained in the scanning process as a characteristic signal of the partial discharge signal. According to the implementation mode provided by the invention, the ultrahigh-frequency partial discharge characteristic signal of the hydro-generator can be rapidly extracted under the complex interference background, and the computing power of an edge end is saved.

Description

Technical Field

[0001] The present application relates to the technical field of data compression, and in particular to a method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator, a device for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator, an electronic device and a corresponding storage medium. Background Art

[0002] As the core equipment of a hydropower station, the health of the insulation system of a hydroelectric generator is directly related to the safety and stability of the entire power system. The detection of partial discharge (PD) signals is an important means to reflect the insulation status of large-scale power equipment. However, traditional low-frequency partial discharge test technology, although the frequency range is 10kHz to 500kHz, is often seriously affected by noise and interference, especially in online test environments. Although ultra-high frequency technology performs well in anti-interference, it also brings new challenges: the number of samples collected per unit time increases dramatically. For example, with a sampling frequency of 1GHz, up to 1 billion sample points can be collected per second. Faced with such a huge amount of data, traditional data processing systems will be overwhelmed.

[0003] At present, the mainstream partial discharge test systems in the market mostly use low-frequency partial discharge test technology with a frequency range of 10kHz to 500kHz. In this frequency range, it can not only have good sensitivity to the stator winding near the partial discharge sensor, but also have good sensitivity to the partial discharge signal in the stator winding far away from the partial discharge sensor. In actual production, the site is complex and the interference sources are diverse. Although the UHF acquisition has better anti-interference ability and more accurate information acquisition, in the field of UHF partial discharge of hydro-turbine generators, the acquisition and processing of UHF signals is still a blank in China. At present, the UHF partial discharge signal monitoring technology is relatively mature, and the company with the highest market share only processes the UHF signal through simple high-pass filtering. The product has a high false alarm rate. Although it avoids the processing, storage and transmission of massive signals, this "violent" cleaning method causes many low-frequency partial discharge signals to be filtered out, greatly reducing the accuracy of the data, which is not conducive to actual production applications. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method and device for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator. In view of the complex multi-source interference in the actual production site, in order to realize the rapid extraction of ultra-high frequency partial discharge characteristic signals of the hydro-turbine generator and avoid missed alarms and false alarms, a method of adaptively selecting the amplitude threshold of the ultra-high frequency partial discharge signal is adopted for monitoring, so as to at least solve some of the problems in the background technology.

[0005] In order to achieve the above-mentioned purpose, the present application provides a method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator, the method comprising: dividing the working scene into a single interference scene and a complex interference scene according to the type or number of interference sources in the working scene where the hydro-turbine generator is located; determining a static amplitude threshold corresponding to the single interference scene, or determining a dynamic amplitude threshold corresponding to the complex interference scene; using a time sliding window to scan the partial discharge signal, recording a point in the time sliding window that is greater than the static amplitude threshold or the dynamic amplitude threshold as a pulse position sample point, and using a set of pulse position sample points obtained during the scanning process as a characteristic signal of the partial discharge signal.

[0006] Optionally, determining the static amplitude threshold corresponding to the single interference scenario includes: analyzing the signal data of the hydro-turbine generator in normal operation and when partial discharge is known to occur based on historically collected data, identifying the statistical characteristics and spectral characteristics of the partial discharge signal and the noise signal, and obtaining an initial value of the static amplitude threshold; verifying the validity of the initial value of the static amplitude threshold through retrospective analysis of historical data, adjusting and optimizing the initial value of the static amplitude threshold according to the verification result, and obtaining an optimized static amplitude threshold; and using the optimized static amplitude threshold as the final static amplitude threshold.

[0007] Optionally, determining the dynamic amplitude threshold corresponding to the complex interference scenario includes: obtaining a discrete sequence of signals in the complex interference scenario; determining an amplitude interval based on the amplitude distribution interval of the discrete sequence, and determining a number of alternative amplitude thresholds based on the amplitude interval; for each alternative amplitude threshold, dividing the discrete sequence into two subsequences, and calculating the inter-class variance between the two subsequences; and using the alternative amplitude threshold corresponding to the maximum inter-class variance as the dynamic amplitude threshold.

[0008] Optionally, for each candidate amplitude threshold, the discrete sequence is divided into two subsequences, and the inter-class variance between the two subsequences is calculated, including: setting the candidate amplitude threshold to k amplitude intervals; taking the number of discrete data in the discrete sequence falling into a certain amplitude interval as n m , the energy value of discrete data is m, then n m is the number of pixels with signal energy value m; the probability of the number of pixels with signal energy value m is P m =n m / N, N is the sum of the number of pixels of discrete data in the discrete sequence; respectively calculating the means of discrete data in the discrete sequence and the two subsequences; calculating the inter-class variance between the subsequences based on the means and the probability sum of the subsequences, wherein the inter-class variance is expressed as a function with k as a parameter.

[0009] Optionally, the inter-class variance between subsequences is calculated based on the mean and the probability of the subsequences, including: σ 2 =ω1(u1-u) 2 +ω2(u2-u) 2 ; Among them, σ 2 is the between-class variance, ω1 is the probability sum of the first subsequence in the two subsequences, ω2 is the probability sum of the second subsequence in the two subsequences, u, u1, u2 are the means of the discrete sequence and the discrete data in the two subsequences respectively.

[0010] Optionally, after taking the set of pulse position sample points obtained during the scanning process as the characteristic signal of the partial discharge signal, the method further includes: sampling the continuous partial discharge signal into a discrete partial discharge signal; converting the discrete partial discharge signal from a time domain signal into a frequency domain signal; and distinguishing between background interference signals and partial discharge pulse signals based on the power density on the frequency spectrum of the frequency domain signal.

[0011] Optionally, after taking the set of pulse position sample points obtained during the scanning process as the characteristic signal of the partial discharge signal, the method further includes: comparing the characteristic signal of the partial discharge signal with historical data, the historical data including historical data of normal operation of the generator and pathological data of historical typical partial discharge; obtaining analysis results through longitudinal comparative analysis, and judging whether a partial discharge event occurs based on the analysis results.

[0012] The present application also provides a device for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator, which includes: a scene division module, which is used to divide the working scene into a single interference scene and a complex interference scene according to the type or number of interference sources in the working scene where the hydro-turbine generator is located; a threshold determination module, which is used to determine the static amplitude threshold corresponding to the single interference scene, or determine the dynamic amplitude threshold corresponding to the complex interference scene; and a feature extraction module, which is used to scan the partial discharge signal using a time sliding window, record the point in the time sliding window that is greater than the static amplitude threshold or the dynamic amplitude threshold as a pulse position sample point, and use the set of pulse position sample points obtained during the scanning process as the characteristic signal of the partial discharge signal.

[0013] The present application also provides an electronic device, comprising: at least one processor; a memory connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the at least one processor implements the aforementioned method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator by executing the instructions stored in the memory.

[0014] The present application also provides a machine-readable storage medium on which instructions are stored. When the instructions are executed by a processor, the processor is configured to execute the aforementioned method for rapidly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator.

[0015] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the aforementioned method for rapidly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator.

[0016] The above technical solution has the following beneficial effects:

[0017] The implementation method of the present application distinguishes the actual on-site production and adopts different processing methods, ultimately achieving rapid extraction of ultra-high frequency partial discharge characteristic signals of hydro-turbine generators under complex interference backgrounds, saving computing power at the edge.

[0018] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present application, but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings:

[0020] Figure 1 A schematic diagram of the steps of a method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator according to an embodiment of the present application is schematically shown;

[0021] Figure 2 A schematic diagram showing the implementation steps of a method for rapidly extracting ultra-high frequency partial discharge characteristic signals of a hydro-generator according to an embodiment of the present application;

[0022] Figure 3 A schematic diagram of the structure of a device for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator according to an embodiment of the present application is shown;

[0023] Figure 4 The internal structure of an electronic device according to an embodiment of the present application is schematically shown. DETAILED DESCRIPTION

[0024] The specific implementation of the embodiment of the present application is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present application, and is not used to limit the embodiment of the present application.

[0025] Figure 1The schematic diagram of the steps of the method for quickly extracting the ultra-high frequency partial discharge characteristic signal of a hydro-turbine generator according to the embodiment of the present application is shown schematically. Figure 1 As shown, a method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-generator comprises:

[0026] S01. According to the types or quantities of interference sources in the working scene where the hydro-generator is located, the working scene is divided into a single interference scene and a complex interference scene;

[0027] S02. Determine a static amplitude threshold corresponding to the single interference scenario, or determine a dynamic amplitude threshold corresponding to the complex interference scenario;

[0028] S03. Scan the partial discharge signal using a time sliding window, record a point in the time sliding window that is greater than the static amplitude threshold or the dynamic amplitude threshold as a pulse position sample point, and use a set of pulse position sample points obtained during the scanning process as a characteristic signal of the partial discharge signal.

[0029] In this embodiment, different processing methods are used by distinguishing the actual production conditions on site, and ultimately the ultra-high frequency partial discharge characteristic signal of the hydro-generator under complex interference background is quickly extracted, saving computing power at the edge.

[0030] Since the partial discharge phenomenon is a low-probability event, electrical equipment is in normal working condition in most cases. Therefore, the amount of signal data collected by ultra-high frequency is huge, and most of them are signals in normal working condition. If the judgment method of adaptive amplitude threshold is always relied on, the computer needs to go through each test data point, which requires extremely high computing power of the computer, cannot meet the actual production conditions and needs on site, and is also not conducive to the rapid extraction of partial discharge signals. In some optional implementations of the present application, step S01 can be implemented in the following way. First, it is necessary to judge the working environment of the production site and confirm the type of interference source, such as corona discharge, radio interference, environmental electromagnetic interference, etc. The processing method of the subsequent ultra-high frequency acquisition signal is determined according to the number of interference source types.

[0031] In some optional implementations of the present application, the static amplitude threshold corresponding to the single interference scenario is determined, including: based on the signal data of the hydro-turbine generator in normal operation and when partial discharge is known to occur, the above data is analyzed to identify the statistical characteristics and spectral characteristics of the partial discharge signal and the noise signal, and the initial value of the static amplitude threshold is obtained; the validity of the initial value of the static amplitude threshold is verified by retrospective analysis of historical data, and the initial value of the static amplitude threshold is adjusted and optimized according to the verification result to obtain the optimized static amplitude threshold; the optimized static amplitude threshold is used as the final static amplitude threshold. The static amplitude threshold in this implementation and the dynamic amplitude threshold in the following text are collectively referred to as the amplitude threshold V m Specifically, if the test site environment is relatively stable and the interference source is single, the amplitude threshold V of the partial discharge signal can be set according to human experience. m , as follows: Based on the historically collected signal data of the turbine generator in normal operation and when partial discharge is known to occur, these data are analyzed to identify the statistical characteristics and spectral features of the partial discharge signal and the noise signal. The partial discharge signal usually appears as a signal with high amplitude, short rise time, and high frequency component, while the noise signal has a lower amplitude and a wider spectrum distribution. Based on these differences, a preliminary amplitude threshold V is first set m , this value needs to be high enough to exclude most of the noise, but not too high to avoid missing the real partial discharge signal. Subsequently, the effectiveness of the selected threshold is verified through retrospective analysis of historical data to ensure that it can effectively identify partial discharge events and minimize misjudgment. After verification, the threshold is adjusted and optimized according to the analysis results to improve the detection rate of partial discharge signals. The verified threshold is applied in real-time monitoring, and is reviewed and updated regularly to adapt to changes in equipment status and environmental conditions, so as to achieve rapid and accurate identification of partial discharge of hydro-turbine generators. Due to the huge amount of ultra-high frequency partial discharge data of large hydro-turbine generators, the method in this implementation can greatly improve the detection efficiency and save computing power when the background noise interference source is single.

[0032] In some optional implementations of the present application, determining the dynamic amplitude threshold corresponding to the complex interference scenario includes: obtaining a discrete sequence of signals in the complex interference scenario; determining an amplitude interval according to the amplitude distribution interval of the discrete sequence, and determining a number of candidate amplitude thresholds according to the amplitude interval; for each candidate amplitude threshold, dividing the discrete sequence into two subsequences, calculating the inter-class variance between the two subsequences; and using the candidate amplitude threshold corresponding to the maximum inter-class variance as the dynamic amplitude threshold. The traditional method for the PD pulse amplitude threshold V m The definition of is usually manually defined. This method, which relies on human subjective judgment, requires a lot of human experience and has a very limited scope of application. For example, it is only applicable to scenarios with a relatively single interference source.m It is not universal and has a large error with the actual test. Because the amplitude threshold V m Not too large nor too small: If V m If V is too small, a large number of interference pulses will be extracted as PD pulses, which will increase the amount of data by mistake, causing data redundancy and ultimately making the partial discharge diagnosis error inconsistent with the actual situation. On the contrary, if V m If it is too large, a large number of PD pulses will be missed, resulting in missing pathological data. If repair is not timely, single partial discharge will accumulate multiple times, leading to the gradual degradation of insulation materials and cumulative effects. Once an outbreak occurs, it will bring greater harm. Therefore, choose a suitable V m It is conducive to the rational extraction of massive data and improves the efficiency of partial discharge signal processing and the accuracy of diagnosis.

[0033] According to the amplitude threshold V m The data points of the PD signal can be divided into two categories according to the size of their ordinates, namely, PD pulse data points and non-PD pulse data points. Due to the different energy densities of the PD pulse signal and the noise signal, the ordinate range of the data points classified as PD pulse data points is large and scattered, while the non-PD pulse data points are concentrated near the value of 0. The variance of the random sequence can measure its degree of dispersion. If there is a certain amplitude threshold V m The inter-class variance of the PD pulse data points and the non-PD pulse data points is maximized, and the difference between the two types of data points is the largest. Therefore, the amplitude threshold V m is the optimal segmentation threshold. The larger its value is, the more obvious the difference between the two categories is, and the better the segmentation effect is. For partial discharge signals and interference signals, the maximum inter-class variance is defined as the amplitude threshold V m , to achieve the separation of the two signals and the extraction of the partial discharge signal.

[0034] The amplitude interval is determined according to the amplitude distribution interval of the discrete sequence, including: assuming that the discrete sequence x with a known length of N is i (i=1,2,3,…,N), the maximum and minimum values ​​of the discrete sequence are x max and x min , then there is an amplitude interval d x =(x max -x min ) / M, where M is the signal energy intensity level. Then the multiple candidate amplitude thresholds are Vm=k*d x , different values ​​of k correspond to an alternative amplitude threshold, and the value range of k is 1 to M.

[0035] In some optional implementations of the present application, for each candidate amplitude threshold, the discrete sequence is divided into two subsequences, and the inter-class variance between the two subsequences is calculated, including: setting the candidate amplitude threshold to k amplitude intervals; assuming that the candidate amplitude threshold V m =k*d x , the discrete data in the discrete sequence falls into a certain amplitude interval [(m-1)d x ,md x The number of ] is n m , the energy value of discrete data is m (m = 1, 2, 3, ..., M), then n m is the number of pixels of signal energy value m; the total number of pixels of the entire signal energy intensity level is equal to the sequence length N, which is also equal to the sum of the number of pixels of each signal energy intensity level, that is, N = n1 + n2 + ... + n m Therefore, the probability of the number of pixels with signal energy value m is P m =n m / N, N is the sum of the number of pixels of discrete data in the discrete sequence; the alternative amplitude threshold V m =k*d x , V m The collected signal discrete sequence is divided into two subsequences C1 and C2, where C1 represents the subsequence falling in the interval [0, k*d x ] can be regarded as the number of noise signals; C2 represents the number of discrete data falling in the interval [(k+1)*d x , M*d x ] can be regarded as the number of partial discharge signals. Then the probability ω and mean u of these two categories are:

[0036]

[0037]

[0038]

[0039]

[0040]

[0041] Wherein, the definitions of M and P are as described above, ω1 is the probability sum of the first subsequence C1 in the two subsequences, ω2 is the probability sum of the second subsequence C2 in the two subsequences, and u, u1, and u2 are the means of discrete sequences, C1, and C2, respectively.

[0042] The inter-class variance between subsequences is calculated based on the mean and the probability of the subsequences, including:

[0043]

[0044] From the above formula, we can see that σ 2 is a function of k, then the optimal candidate amplitude threshold V is obtained by determining the value of k m , so that the variance between the two classes is maximized.

[0045]

[0046] By determining the amplitude thresholds of the partial discharge signal and the interference signal through this implementation, the two signals are separated and the partial discharge signal is extracted.

[0047] In some embodiments of the present application, after taking the set of pulse position sample points obtained during the scanning process as the characteristic signal of the partial discharge signal, the method further includes: sampling the continuous partial discharge signal into a discrete partial discharge signal; converting the discrete partial discharge signal from a time domain signal into a frequency domain signal; and distinguishing the background interference signal from the partial discharge pulse signal based on the power density on the frequency spectrum of the frequency domain signal. For example, for the sampling of discrete signals, if a continuous partial discharge signal x(t) is sampled at a sampling interval T s For sampling, the discrete sampling signal is:

[0048] x(n)=x(nT s ) (n=0,1,2…) (8)

[0049] Where, sampling frequency fs = 1 / T s The sampling theorem needs to be satisfied, that is, fs ≥ 2B, where B is the maximum frequency component of the signal. Ultra-high frequency sampling, whose sampling frequency is much larger than the maximum frequency component of the partial discharge signal, can satisfy the sampling theorem. The discrete time domain signal is converted into a frequency domain signal through discrete Fourier transform (DFT), and the formula is as follows:

[0050]

[0051] Where x[n] is a discrete time domain signal, X[k] is the corresponding frequency domain signal, and N is the length of the PD signal. For the PD signal feature, in addition to the amplitude threshold, the power density can also be analyzed on the frequency spectrum. Since the energy density of the PD signal is much greater than that of the noise signal, the background interference signal and the PD pulse signal can be distinguished based on this.

[0052]

[0053] Wherein, P[k] represents the power of the kth frequency component, and X[k] represents the corresponding frequency domain signal.

[0054] In some embodiments of the present application, after the set of pulse position sample points obtained during the scanning process is used as the characteristic signal of the partial discharge signal, the method further includes: comparing the characteristic signal of the partial discharge signal with historical data, the historical data including historical data of normal operation of the generator and pathological data of historical typical partial discharge; obtaining analysis results through longitudinal comparative analysis, and judging whether a partial discharge event occurs according to the analysis results. The following is a description of the production site PD pulse signal extraction process based on a single interference source and a complex interference source.

[0055] The extraction process of PD pulse signals at production sites with a relatively single interference source. Continuously collect ultra-high frequency partial discharge signals within a specified time period, and obtain the static amplitude threshold V determined by the above step 2. m The time sliding window N is used to scan the partial discharge signal in sequence. When the first signal greater than V appears in the time window, m When the number of pulses is reached, the position of the point is recorded as a pulse position sample point P, and the time window is moved backward to repeatedly locate the PD pulse until the entire partial discharge signal is scanned to achieve the extraction of the entire pulse signal. The collected signal is tested for amplitude in real time, and the amplitude exceeding the threshold V is recorded. m If the amplitude of multiple sample points exceeds the standard (exceeds the set amplitude threshold) continuously within the set time period, it is preliminarily determined to be a partial discharge signal. Subsequently, the preliminarily determined partial discharge signal is compared with historical data, including historical data of normal operation of the generator and pathological data of historical typical partial discharges. The data is analyzed by longitudinal comparison to confirm whether a partial discharge event has occurred. This method can reduce misjudgments caused by complex background interference while ensuring high efficiency, and realize rapid and accurate extraction of partial discharge signals, thereby improving the accuracy and response speed of partial discharge monitoring.

[0056] The PD pulse signal extraction process of the production site under the complex interference background of multiple interference sources. Due to the complex on-site environment, the use of artificial experience to customize the static amplitude threshold has great limitations. Therefore, the appropriate adaptive dynamic amplitude threshold V is selected in step 2. m , use the time sliding window N to scan the partial discharge signal in sequence, when the first signal greater than V appears in the time window m When the number of pulses is reached, the position of the point is recorded as a pulse position sample point P, and the time window is moved backward to repeatedly locate the PD pulse until the entire partial discharge signal is scanned to achieve the extraction of the entire pulse signal. The collected signal is tested for amplitude in real time, and the amplitude exceeding the threshold V is recorded. mIf the amplitude of multiple sample points exceeds the standard (exceeds the set amplitude threshold) continuously within the set time period, it is preliminarily determined to be a partial discharge signal. Due to the complex on-site environment, in order to avoid false triggering by other phenomena, it is necessary to compare the preliminarily determined partial discharge signal with the historical typical partial discharge signal data, and to confirm whether a partial discharge event has occurred by longitudinally comparing and analyzing the data.

[0057] This implementation provides further comparison after extracting the preliminarily determined partial discharge signal, thereby improving the accuracy of the partial discharge signal.

[0058] Figure 2 The following is a schematic diagram showing the implementation steps of the method for quickly extracting the ultra-high frequency partial discharge characteristic signal of a hydro-turbine generator according to the embodiment of the present application. Figure 2 As shown, it includes a test site interference source type determination step, a static amplitude threshold or a dynamic amplitude threshold selection step, a partial discharge extraction step and a longitudinal comparison step.

[0059] Through the above implementation methods, different processing methods are adopted for different interference scenarios, and finally the ultra-high frequency partial discharge characteristic signal of the hydro-turbine generator under the complex interference background is quickly extracted, which has the advantage of saving the computing power of the edge end.

[0060] Based on the same inventive concept, the present application also provides a device for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator. Figure 3 The schematic diagram of the structure of the device for quickly extracting the ultra-high frequency partial discharge characteristic signal of a hydro-turbine generator according to the embodiment of the present application is shown schematically. Figure 3 As shown, the device includes: a scene division module, which is used to divide the working scene into a single interference scene and a complex interference scene according to the type or number of interference sources in the working scene where the hydro-turbine generator is located; a threshold determination module, which is used to determine the static amplitude threshold corresponding to the single interference scene, or determine the dynamic amplitude threshold corresponding to the complex interference scene; and a feature extraction module, which is used to scan the partial discharge signal using a time sliding window, record the point in the time sliding window that is greater than the static amplitude threshold or the dynamic amplitude threshold as a pulse position sample point, and use the set of pulse position sample points obtained during the scanning process as a feature signal for the partial discharge signal.

[0061] In some optional embodiments of the present application, determining the static amplitude threshold corresponding to the single interference scenario includes: analyzing the signal data of the hydro-turbine generator in normal operation and when partial discharge is known to occur based on historically collected data to identify the statistical characteristics and spectral characteristics of the partial discharge signal and the noise signal, and obtaining an initial value of the static amplitude threshold; verifying the validity of the initial value of the static amplitude threshold through retrospective analysis of historical data, adjusting and optimizing the initial value of the static amplitude threshold according to the verification result, and obtaining an optimized static amplitude threshold; and using the optimized static amplitude threshold as the final static amplitude threshold.

[0062] In some optional embodiments of the present application, determining the dynamic amplitude threshold corresponding to the complex interference scenario includes: obtaining a discrete sequence of signals in the complex interference scenario; determining an amplitude interval based on the amplitude distribution interval of the discrete sequence, and determining a number of alternative amplitude thresholds based on the amplitude interval; for each alternative amplitude threshold, dividing the discrete sequence into two subsequences, and calculating the inter-class variance between the two subsequences; and using the alternative amplitude threshold corresponding to the maximum inter-class variance as the dynamic amplitude threshold.

[0063] In some optional implementations of the present application, for each candidate amplitude threshold, the discrete sequence is divided into two subsequences, and the inter-class variance between the two subsequences is calculated, including: setting the candidate amplitude threshold to k amplitude intervals; taking the number of discrete data in the discrete sequence falling into a certain amplitude interval as n m , the energy value of discrete data is m, then n m is the number of pixels with signal energy value m; the probability of the number of pixels with signal energy value m is P m =n m / N, N is the sum of the number of pixels of discrete data in the discrete sequence; respectively calculating the means of discrete data in the discrete sequence and the two subsequences; calculating the inter-class variance between the subsequences based on the means and the probability sum of the subsequences, wherein the inter-class variance is expressed as a function with k as a parameter.

[0064] In some optional embodiments of the present application, the inter-class variance between subsequences is calculated based on the mean and the probability of the subsequences, including: σ 2 =ω1(u1-u) 2 +ω2(u2-u) 2 ; Among them, σ 2 is the between-class variance, ω1 is the probability sum of the first subsequence in the two subsequences, ω2 is the probability sum of the second subsequence in the two subsequences, u, u1, u2 are the means of the discrete sequence and the discrete data in the two subsequences respectively.

[0065] In some optional embodiments of the present application, after taking the set of pulse position sample points obtained during the scanning process as the characteristic signal of the partial discharge signal, the device also includes a frequency domain processing module, which is used to: sample the continuous partial discharge signal into a discrete partial discharge signal; convert the discrete partial discharge signal from a time domain signal into a frequency domain signal; and distinguish between background interference signals and partial discharge pulse signals based on the power density on the frequency spectrum of the frequency domain signal.

[0066] In some optional embodiments of the present application, after taking the set of pulse position sample points obtained during the scanning process as the characteristic signal of the partial discharge signal, the device also includes a longitudinal comparison module, which is used to: compare the characteristic signal of the partial discharge signal with historical data, wherein the historical data includes historical data of normal operation of the generator and pathological data of historical typical partial discharge; obtain analysis results through longitudinal comparison analysis, and determine whether a partial discharge event occurs based on the analysis results.

[0067] The specific definition of each functional module in the above-mentioned device for quickly extracting ultra-high frequency partial discharge characteristic signals of hydro-turbine generators can be found in the above-mentioned definition of the method for quickly extracting ultra-high frequency partial discharge characteristic signals of hydro-turbine generators, which will not be repeated here. Each module in the above-mentioned system can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. It also realizes the advantages of realizing the rapid extraction of ultra-high frequency partial discharge characteristic signals of hydro-turbine generators under complex interference backgrounds and saving computing power at the edge.

[0068] In some embodiments of the present application, an electronic device is also provided, comprising: at least one processor; a memory connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the at least one processor executes the aforementioned method for rapidly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator. The internal structure diagram thereof can be as follows Figure 4 shown. Figure 4The internal structure diagram of an electronic device according to an embodiment of the present application is schematically shown. The electronic device includes a processor A01, a network interface A02, a memory (not shown in the figure) and a database (not shown in the figure) connected via a system bus. Among them, the processor A01 of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes an internal memory A03 and a non-volatile storage medium A04. The non-volatile storage medium A04 stores an operating system B01, a computer program B02 and a database (not shown in the figure). The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A04. The network interface A02 of the electronic device is used to communicate with an external terminal through a network connection. When the computer program B02 is executed by the processor A01, a method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator is implemented.

[0069] Those skilled in the art will understand that Figure 4 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0070] In one embodiment provided in the present application, a machine-readable storage medium is provided, on which instructions are stored. When the instructions are executed by a processor, the processor is configured to execute the aforementioned method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator.

[0071] In one embodiment provided in the present application, a computer program product is provided, including a computer program, which, when executed by a processor, implements the aforementioned method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator.

[0072] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0073] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes 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 generate 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.

[0074] 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.

[0075] 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. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0076] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0077] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0078] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0079] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0080] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-generator, characterized in that: The method includes: According to the type or quantity of interference sources in the working scene where the hydro-generator is located, the working scene is divided into a single interference scene and a complex interference scene; Determining a static amplitude threshold corresponding to the single interference scenario, or determining a dynamic amplitude threshold corresponding to the complex interference scenario; A time sliding window is used to scan the partial discharge signal, and a point in the time sliding window that is greater than the static amplitude threshold or the dynamic amplitude threshold is recorded as a pulse position sample point. The set of pulse position sample points obtained during the scanning process is used as a characteristic signal of the partial discharge signal.

2. The method according to claim 1, characterized in that Determining a static amplitude threshold corresponding to the single interference scenario includes: Based on the historically collected signal data of the hydro-generator in normal operation and when partial discharge is known to occur, the above data are analyzed to identify the statistical characteristics and spectral features of the partial discharge signal and the noise signal, and the initial value of the static amplitude threshold is obtained; Verifying the validity of the initial value of the static amplitude threshold by back-analyzing historical data, and adjusting and optimizing the initial value of the static amplitude threshold according to the verification result to obtain an optimized static amplitude threshold; The optimized static amplitude threshold is used as the final static amplitude threshold.

3. The method according to claim 1, characterized in that Determining a dynamic amplitude threshold corresponding to the complex interference scenario includes: Acquire a discrete sequence of signals in the complex interference scenario; Determine an amplitude interval according to the amplitude distribution interval of the discrete sequence, and determine a plurality of candidate amplitude thresholds according to the amplitude interval; For each candidate amplitude threshold, the discrete sequence is divided into two subsequences, and the inter-class variance between the two subsequences is calculated; The candidate amplitude threshold corresponding to the maximum inter-class variance is used as the dynamic amplitude threshold.

4. The method according to claim 1, characterized in that For each candidate amplitude threshold, the discrete sequence is divided into two subsequences, and the inter-class variance between the two subsequences is calculated, including: Setting the candidate amplitude threshold to k amplitude intervals; The number of discrete data in the discrete sequence that falls into a certain amplitude interval is n m , the energy value of discrete data is m, then n m is the number of pixels with signal energy value m; The probability of the number of pixels with signal energy value m is P m =n m / N, N is the sum of the number of pixels of discrete data in the discrete sequence; Calculate the mean of discrete data in the discrete sequence and two subsequences respectively; The inter-class variance between sub-sequences is calculated based on the mean and the probability of the sub-sequences, and the inter-class variance is expressed as a function with k as a parameter.

5. The method according to claim 1, characterized in that The inter-class variance between subsequences is calculated based on the mean and the probability of the subsequences, including: s 2 =ω1(u1-u) 2 +ω2(u2-u) 2 Among them, σ 2 is the between-class variance, ω1 is the probability sum of the first subsequence in the two subsequences, ω2 is the probability sum of the second subsequence in the two subsequences, u, u1, u2 are the means of the discrete sequence and the discrete data in the two subsequences respectively.

6. The method according to claim 1, characterized in that After taking the set of pulse position sample points obtained during the scanning process as a characteristic signal of the partial discharge signal, the method further includes: Sampling a continuous partial discharge signal into a discrete partial discharge signal; Converting the discrete partial discharge signal from a time domain signal to a frequency domain signal; The background interference signal and the partial discharge pulse signal are distinguished based on the power density on the frequency spectrum of the frequency domain signal.

7. The method according to claim 1, characterized in that After taking the set of pulse position sample points obtained during the scanning process as a characteristic signal of the partial discharge signal, the method further includes: Comparing the characteristic signal of the partial discharge signal with historical data, wherein the historical data includes historical data of normal operation of the generator and historical pathological data of typical partial discharge; An analysis result is obtained through longitudinal comparative analysis, and whether a partial discharge event occurs is determined based on the analysis result.

8. A device for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-generator, characterized in that: The device includes: A scene division module, used for dividing the working scene where the hydro-generator is located into a single interference scene and a complex interference scene according to the type or number of interference sources in the working scene where the hydro-generator is located; a threshold determination module, configured to determine a static amplitude threshold corresponding to the single interference scenario, or determine a dynamic amplitude threshold corresponding to the complex interference scenario; and The feature extraction module is used to scan the partial discharge signal using a time sliding window, record the point in the time sliding window that is greater than the static amplitude threshold or the dynamic amplitude threshold as a pulse position sample point, and use the set of pulse position sample points obtained during the scanning process as a characteristic signal of the partial discharge signal.

9. An electronic device, characterized in that: include: at least one processor; a memory connected to the at least one processor; Wherein, the memory stores instructions that can be executed by the at least one processor, and the at least one processor implements the steps of the method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator as described in any one of claims 1 to 7 by executing the instructions stored in the memory.

10. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the steps of the method for quickly extracting ultra-high frequency partial discharge characteristic signals of a hydro-turbine generator as described in any one of claims 1 to 7 are implemented.

Images