A switch cabinet partial discharge monitoring method and system

Abstract

The present application relates to the field of discharge monitoring technology, and specifically to a method and system for monitoring partial discharge of a switch cabinet, the method comprising: collecting current pulse signals; constructing current pulse discrimination coefficients of each time window according to the fluctuation signal characteristics, periodic signal characteristics and current pulse signal characteristics under the partial discharge state of the switch cabinet; obtaining the peak value of the current pulse signal; obtaining the current signal peak discrimination coefficient of each time window based on the change characteristics of the peak value of the current pulse signal when the noise interference is large, and then obtaining the partial discharge discrimination coefficient of each time window, and performing partial discharge monitoring on the switch cabinet. The present application improves the accuracy of partial discharge monitoring during the operation of the switch cabinet.

Description

Technical Field

[0001] The present application relates to the technical field of discharge monitoring, and in particular to a method and system for monitoring partial discharge of a switch cabinet. Background Art

[0002] Power switchgear is the most critical equipment for directly supplying power to distribution networks or general large users. It is widely used in power supply and distribution systems. Its failure will directly affect the quality of power supply services and lead to the most direct economic losses. Under the action of operating voltage, current, heat, force, etc., as well as during production, transportation, installation and maintenance, power switchgear will inevitably produce or leave various insulation defects. These defects gradually develop and cause the electrical strength of the equipment to decrease, leading to failures. Partial discharge monitoring is an important technical means to prevent sudden power outages caused by insulation failures in power switchgear. By accurately monitoring, locating and identifying insulation defects, the reliability of switchgear operation can be improved, and the number of power outages for maintenance and the occurrence of power outages can be reduced.

[0003] In the prior art, pulse current method, ultrasonic monitoring method, ultra-high frequency method, chemical monitoring method and other methods are usually used to monitor the partial discharge of switch cabinets, and the acquired signal data is analyzed to judge the partial discharge of the switch cabinet. Among them, the pulse current method is widely used in the signal acquisition of partial discharge monitoring of switch cabinets due to its advantages such as high monitoring sensitivity and good monitoring effect. However, in the process of monitoring by the pulse current method, since there are many electrical equipment in the monitoring environment, these electrical equipment may generate certain electromagnetic interference during operation, which in turn affects the acquisition of pulse current signals, and ultimately leads to more noise in the collected pulse current signals and poor signal quality, which affects the subsequent monitoring and judgment of partial discharge of switch cabinets. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a switch cabinet partial discharge monitoring method and system, and the technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a switch cabinet partial discharge monitoring method, the method comprising the following steps:

[0006] Collect the current pulse signal of the switch cabinet at each sampling moment in each time window;

[0007] According to the fluctuation signal characteristics, periodic signal characteristics and trend characteristics of the switch cabinet under the partial discharge state and the synchronization characteristics of the changes between local signals, the current pulse discrimination coefficient of each time window is constructed;

[0008] Obtaining the peak value of the current pulse signal; obtaining the current signal peak discrimination coefficient of each time window based on the variation characteristics of the peak value of the current pulse signal;

[0009] Based on the current pulse discrimination coefficient and the current signal peak discrimination coefficient, the partial discharge discrimination coefficient of each time window is obtained;

[0010] Partial discharge monitoring of switchgear is carried out based on the partial discharge discrimination coefficient.

[0011] Furthermore, the method for obtaining the current pulse discrimination coefficient is:

[0012] Based on the amplitude change and signal transition of the current pulse signal, the signal fluctuation coefficient of each time window is obtained;

[0013] Based on the periodicity and approximation of the current pulse signal change, the signal regularity coefficient of each time window is obtained;

[0014] The ratio of the signal regularity coefficient to the signal fluctuation coefficient is calculated as the current pulse discrimination coefficient of each time window.

[0015] Furthermore, the method for obtaining the signal fluctuation coefficient is:

[0016] Arrange all signal amplitude data in each time window in ascending order of their sampling time to obtain the signal amplitude sequence of each time window;

[0017] For each time window, the ratio between the variance of the amplitude of the current pulse signal in the time window and the maximum amplitude is calculated as the signal amplitude variation coefficient of each window; the time interval between the maximum and minimum amplitudes in the time window is calculated, and the trend test weight of the signal amplitude sequence in the time window is calculated, and the product of the time interval and the trend test weight is used as the signal variation coefficient of each time window;

[0018] According to the signal amplitude variation coefficient and the signal gradual variation coefficient, the signal fluctuation coefficient of each time window is obtained, and the signal fluctuation coefficient is positively correlated with the signal amplitude variation coefficient and negatively correlated with the signal gradual variation coefficient.

[0019] Furthermore, the method for obtaining the signal regularity coefficient is:

[0020] For each sampling moment, the signal amplitudes in all current pulse signals within a preset number of time windows before the sampling moment are arranged in time positive order to construct a first signal sequence, and the periodicity strength of the first signal sequence is calculated as the signal change periodicity coefficient at each sampling moment;

[0021] Calculate the PLV value of the signal amplitude sequence of all time windows;

[0022] For each time window, the sum of the PLV values ​​of the time window and all the previous time windows is calculated as the first sum, the difference between the signal amplitude sequences of the time window and other previous time windows is calculated, the mean of all the difference values ​​is calculated as the first mean, and the ratio of the first sum to the first mean is taken as the signal approximation degree of each time window;

[0023] Based on the signal change periodicity coefficient and the signal approximation degree, the signal regularity coefficient of each time window is obtained, and the signal regularity coefficient is positively correlated with the signal change periodicity coefficient and the signal approximation degree respectively.

[0024] Furthermore, the method for obtaining the current signal peak discrimination coefficient is:

[0025] Based on the peak value difference degree and peak value distribution density difference of the current pulse signal in each time window, the peak value consistency coefficient of each time window is obtained;

[0026] Based on the probability of each peak value appearing in all peak values ​​in each time window and the number of peak values, the probability of current peak value generation in each time window is obtained;

[0027] The product of the peak consistency coefficient and the probability of current peak generation is taken as the current signal peak discrimination coefficient of each time window.

[0028] Further, the peak consistency coefficient is obtained by:

[0029] For each time window, obtain the order in which the sampling moments corresponding to the peak values ​​in the time window appear in the time window; for any time window before the time window, obtain the absolute value of the difference between the amplitude and the peak value of the current pulse signal corresponding to the sampling moments of the same order; take the mean of all the absolute values ​​of the difference in any time window as the second mean of any time window; calculate the sum of the second means of all time windows before the time window as the peak difference degree;

[0030] For each time window, the absolute value of the difference between the sampling time corresponding to each peak in the time window and the first sampling time of the time window in which it is located is calculated as the first absolute value of the difference, and the average of all the first absolute values ​​of the difference is used as the peak distribution density of each time window; the average of the absolute values ​​of the difference between the peak distribution density of each time window and all previous time windows is calculated as the peak distribution density difference of each time window;

[0031] The peak consistency coefficient of each time window is obtained according to the peak difference degree and the peak distribution density difference. The peak consistency coefficient is positively correlated with the peak difference degree and negatively correlated with the peak distribution density difference.

[0032] Furthermore, the method for obtaining the probability of the current peak value being generated is:

[0033] For each time window, obtain the peak values ​​of all current pulse signals in each time window and a previously preset number of time windows, and construct a pair of peak values ​​of each current pulse signal in the order of the acquisition moments of the peak values ​​in the time window; for the pair of peak values ​​of each current pulse signal in each time window and a previously preset number of time windows, count the ratio of the total number of each identical pair to the total number of all pairs as the first ratio of each identical pair in the current pulse signal, and take the average of all first ratios of each time window as the probability of occurrence of the peak value of each time window;

[0034] For each time window, the absolute value of the difference between the number of times the peak value of the current pulse signal in the time window appears and the number of times the peak value of the current pulse signal in other time windows is calculated as the second absolute value of the difference in each time window, and the average of all the second absolute values ​​of the difference in each time window is taken as the average of the peak value occurrence differences in each time window;

[0035] The current peak generation probability of each time window is obtained according to the peak occurrence probability and the mean value of the peak occurrence difference. The current peak generation probability is positively correlated with the peak occurrence probability and negatively correlated with the mean value of the peak occurrence difference.

[0036] Furthermore, the formula of the partial discharge discrimination coefficient is: ; Where A is the partial discharge discrimination coefficient of each time window; B is the current pulse discrimination coefficient of each time window; and C is the current signal peak discrimination coefficient of each time window.

[0037] Further, the partial discharge monitoring of the switch cabinet according to the partial discharge discrimination coefficient includes:

[0038] Through the intelligent switchgear expert diagnosis system cloud platform, the segmentation threshold of the partial discharge discrimination coefficient of all time windows of a single switchgear is obtained; all current pulse signals are taken as input, and the denoising algorithm is used to take the segmentation threshold of the partial discharge discrimination coefficient as the segmentation threshold of denoising, and the output is the denoised current pulse signal;

[0039] The intelligent switchgear expert diagnosis system cloud platform uses the current pulse signal after denoising to monitor the operating status of the switchgear.

[0040] In a second aspect, an embodiment of the present application further provides a switch cabinet partial discharge monitoring system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0041] This application has at least the following beneficial effects:

[0042] The present application analyzes the current pulse signal during the operation of the switch cabinet, constructs a current pulse discrimination coefficient and a current signal peak discrimination coefficient, and then constructs a partial discharge discrimination coefficient, distinguishes the current pulse signal and the noise signal generated by the partial discharge in the current pulse signal, and then adjusts the denoising threshold of the wavelet denoising, thereby solving the problem that the effective information in the signal is easily smoothed when the current pulse signal is denoised, improving the quality of the current pulse signal after denoising, and further improving the accuracy of partial discharge monitoring during the operation of the switch cabinet. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 A flowchart of a method for monitoring partial discharge of a switch cabinet provided in one embodiment of the present application;

[0045] Figure 2 A partial discharge discrimination coefficient flow chart provided for one embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of a switch cabinet partial discharge monitoring method and system proposed in the present application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0047] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0048] The specific scheme of a switch cabinet partial discharge monitoring method and system provided by the present application is described in detail below with reference to the accompanying drawings.

[0049] See also Figure 1 , which shows a flow chart of the steps of a switch cabinet partial discharge monitoring method provided by an embodiment of the present application, the method comprising the following steps:

[0050] Step S1, obtaining a current pulse signal generated on the capacitors at both ends of the switch cabinet.

[0051] For each switch cabinet, a partial discharge pulse current monitoring device is set in the switch cabinet to continuously sample the current pulse signal generated in the switch cabinet to obtain the current pulse signal. Each interval time T is divided into a time window, and each time window has multiple signal amplitude data.

[0052] In this embodiment, the sampling frequency of the current pulse signal is 100 Hz, T=0.1s.

[0053] Step S2, constructing a current pulse discrimination coefficient of a single time window according to the noise signal characteristics and the current pulse signal characteristics under the partial discharge state of the switch cabinet; constructing a current signal peak discrimination coefficient of a single time window according to the change characteristics of the current pulse signal peak when the noise interference is large; constructing a partial discharge discrimination coefficient according to the current pulse discrimination coefficient and the current signal peak discrimination coefficient of each time window.

[0054] When the switch cabinet is operating normally and no partial discharge occurs, the current pulse signal is relatively stable, but it will continue to be affected by noise, which is manifested in the frequent appearance of burrs on the current pulse signal, low amplitude, no obvious regularity, and large amplitude differences between noises. When partial discharge occurs, the current pulse signal will continue to generate current pulse waveforms, which are mainly manifested in large mutation intensity, a certain decrease in the amplitude of each peak in a single waveform, and a short duration of a single waveform. As partial discharge continues, the current pulse waveform is relatively periodic as a whole.

[0055] The signal amplitude sequence of a single time window is constructed by arranging all the signal amplitude data in a single time window in ascending order according to their sampling time, and the constructed sequence is the signal amplitude sequence.

[0056] Furthermore, for each time window, the ratio between the variance of the amplitude of the current pulse signal in the time window and the maximum amplitude is calculated as the signal amplitude variation coefficient of each window; further, the time interval between the maximum and minimum amplitudes in the time window is calculated, and the trend test weight of the signal amplitude sequence in the time window is calculated, and the product of the time interval and the trend test weight is used as the signal variation coefficient of each time window.

[0057] It should be noted that the calculation method of the trend test weight of the signal amplitude sequence is as follows: the signal amplitude sequence is used as input, the Cox-Stuart trend test is used, the significance level parameter is set to 0.05, and the output is the trend test P value of the signal amplitude sequence. When the P value is less than the significance level parameter, the signal amplitude sequence is considered to be a decreasing sequence. At this time, the trend test weight of the signal amplitude sequence is set to On the contrary, when the P value is greater than or equal to the significance level parameter, it is considered that the signal amplitude sequence has no obvious gradual change pattern. At this time, the trend test weight of the signal replication sequence is set to ,in In this embodiment, set , The Cox-Stuart trend test is a well-known technique and will not be described in detail in this application.

[0058] Furthermore, the signal fluctuation coefficient of each time window is obtained according to the signal amplitude variation coefficient and the signal gradual variation coefficient, and the signal fluctuation coefficient is positively correlated with the signal amplitude variation coefficient and negatively correlated with the signal gradual variation coefficient.

[0059] It should be noted that the positive correlation relationship indicates that the dependent variable increases as the independent variable increases, and decreases as the independent variable decreases; the negative correlation relationship indicates that the dependent variable decreases as the independent variable increases, and increases as the independent variable decreases.

[0060] Specifically, the present application proposes an embodiment, in which, for each time window, the ratio of the signal amplitude variation coefficient to the signal gradient coefficient is used as the signal fluctuation coefficient of each time window.

[0061] Furthermore, the present application proposes another embodiment, in which, for each time window, a calculation result of an exponential function with a natural constant as the base and the negative value of the signal variation coefficient as the exponent is calculated as a first exponent, and the product of the signal amplitude variation coefficient and the first exponent is used as the signal fluctuation coefficient of each time window.

[0062] Furthermore, the signal regularity coefficient of each time window is obtained according to the periodicity and approximation of the current pulse signal change;

[0063] Specifically, for each sampling moment, The signal amplitudes of all current pulse signals within a time window are arranged in time positive order to construct a first signal sequence, and the periodicity strength of the first signal sequence is calculated as the signal change periodicity coefficient at each sampling moment. The periodicity strength can be calculated by the periodicity strength calculation formula in the STL decomposition algorithm. The STL decomposition algorithm is a well-known technology, and the present application will not repeat its process. In this embodiment, The value of is 20.

[0064] Calculate the PLV value (Phase Locking Value) of the signal amplitude sequence of all time windows; further, for each time window, calculate the sum of the PLV values ​​of the time window and all previous time windows as the first sum, calculate the difference between the signal amplitude sequences of the time window and other time windows before it, calculate the mean of all difference values ​​as the first mean, and take the ratio of the first sum to the first mean as the signal approximation degree of each time window. Among them, the calculation method of the PLV value is a well-known technology and will not be described in detail here. If it is not possible to obtain enough data, the mean filling method is used for data filling. Mean filling is a well-known technology and the present application will not describe its process in detail.

[0065] Furthermore, based on the signal change periodicity coefficient and the signal approximation degree, the signal regularity coefficient of each time window is obtained, and the signal regularity coefficient is positively correlated with the signal change periodicity coefficient and the signal approximation degree, respectively.

[0066] Specifically, the present application proposes an embodiment, which calculates, for each sampling moment, the product of the signal change periodicity coefficient and the signal approximation degree of the time window where the sampling moment is located as the signal regularity coefficient of each time window.

[0067] Furthermore, the present application proposes another embodiment, which calculates the sum of the signal change periodicity coefficient and the signal approximation degree as a second sum, and calculates the sum of an exponential function with a natural constant as the base and the second sum as the exponent as the signal regularity coefficient of each time window.

[0068] According to the noise signal characteristics and current pulse signal characteristics of the switch cabinet under partial discharge state, the current pulse discrimination coefficient of a single time window is constructed.

[0069] The ratio of the signal regularity coefficient to the signal fluctuation coefficient is calculated as the current pulse discrimination coefficient of each time window.

[0070] The larger the current pulse discrimination coefficient of a single time window is, the closer the signal change characteristics within the time window are to the current pulse signal characteristics when partial discharge occurs in the switch cabinet, and the higher the regularity, the more likely it is that the current pulse signal is generated by partial discharge, and the more adjustments should be made to the denoising effect within the time window.

[0071] When the signal fluctuation coefficient of a single time window is larger, it means that the signal amplitude change in the single time window is larger, but the signal amplitude is smaller, and there is no obvious gradual change in the signal amplitude in the time window. The closer the signal amplitude change characteristics in the time window are to the change characteristics of the noise signal, the more likely it is that the change is caused by noise, and the corresponding current pulse discrimination coefficient is smaller.

[0072] The larger the signal regularity coefficient at a single moment is, the greater the similarity between the signal amplitude data in the current time window and the signal amplitude data before the time window is, and the signal amplitude data before the current time window has a stronger periodicity. It is more likely that the current pulse is generated by the partial discharge of the switch cabinet, which leads to an increase in the similarity of the change of the current pulse signal, and the corresponding current pulse discrimination coefficient is larger.

[0073] At this point, the current pulse discrimination coefficient of each time window is obtained.

[0074] Furthermore, since the operating environments of switch cabinets are different and the electromagnetic interference they are subject to is also different, the degree of noise interference is also different. There may be a situation where the noise interference is large, and then the noise signal masks the current pulse signal. Therefore, when the current pulse is identified only by the above method, most of the characteristics of the current pulse signal are masked, and the current pulse identification coefficient is small, so the current pulse signal is identified as noise, which ultimately leads to the misjudgment of the actual situation that the local discharge of the switch cabinet generates the current pulse signal. Therefore, further analysis is needed.

[0075] Specifically, when the noise interference is large, the amplitude of the noise signal is close to the amplitude of the current pulse signal, so the overall signal will have more peaks, and the peak value of the noise signal varies more and irregularly, while the peak value generated by the current pulse signal is relatively constant, and the number of peak values ​​generated by each current pulse signal in a single time window and the peak value distribution are relatively similar. The probability of the current pulse signal peak value is higher, while the probability of each peak value of the noise signal is lower.

[0076] The FindPeaks peak-to-valley value monitoring algorithm is used to process the amplitude data of the current pulse signal at all sampling moments to obtain all peak values ​​of the current pulse signal and their corresponding sampling moments.

[0077] Among them, the FindPeaks peak-to-valley value monitoring algorithm is a well-known technology, and this application will not go into details about its process.

[0078] Furthermore, for each time window, the order in which the sampling moments corresponding to the peak values ​​in the time window appear in the time window is obtained; for each time window before the time window, the absolute value of the difference between the amplitude and the peak value of the current pulse signal corresponding to the sampling moments in the same order is obtained; the mean of all the absolute values ​​of the difference in the time window is taken as the second mean of each time window; and the sum of the second means of all time windows is calculated as the peak difference degree.

[0079] It should be noted that, if the first peak in a time window is the fifth signal amplitude data in the time window, the sampling time sequence of the peak in the time window is 5, and the peak sampling time sequences in other time windows are similar.

[0080] Furthermore, for each time window, the absolute value of the difference between the sampling moment corresponding to each peak in the time window and the first sampling moment of the time window is calculated as the first absolute value of the difference, and the average of all the first absolute values ​​of the difference is taken as the peak distribution density of each time window; the average of the absolute values ​​of the difference between the peak distribution density of each time window and all other previous time windows is calculated as the peak distribution density difference of each time window.

[0081] Furthermore, for each time window, the peak consistency coefficient of each time window is obtained according to the peak difference degree and the peak distribution density difference, and the peak consistency coefficient is positively correlated with the peak difference degree and negatively correlated with the peak distribution density difference.

[0082] Specifically, the present application proposes an embodiment, in which, for each time window, the ratio of the peak difference degree to the peak distribution density difference is used as the peak consistency coefficient of each time window.

[0083] Furthermore, the present application proposes another embodiment, in which, for each time window, a calculation result of an exponential function with a natural constant as the base and the negative value of the peak distribution density difference as the exponent is calculated as a second index, and the product of the peak difference degree and the second index is used as the peak consistency coefficient of each time window.

[0084] The larger the peak consistency coefficient of a single time window, the greater the similarity between the peak in the time window and the peaks in other time windows, and the distribution density is also similar. It is more likely to be the peak generated by the current pulse signal generated by the partial discharge of the switch cabinet, and the larger the corresponding current signal peak differentiation coefficient.

[0085] Furthermore, for each time window, obtain each time window and the previous The peak values ​​of all current pulse signals in a time window and the order of the peak values ​​in the time window are used to construct a binary group of the peak values ​​of each current pulse signal; for each time window and the previous The binary groups of the peak values ​​of each current pulse signal in each time window are counted, and the ratio of the total number of each identical binary group to the total number of all binary groups is counted as the first ratio of each identical binary group in the current pulse signal. The average of all the first ratios of each time window is taken as the probability of occurrence of the peak value in each time window.

[0086] Furthermore, for each time window, the absolute value of the difference between the number of occurrences of the peak value of the current pulse signal in the time window and the number of occurrences of the peak value of the current pulse signal in other time windows is calculated as the second absolute value of the difference of each time window, and the average of all the absolute values ​​of the second difference in each time window is taken as the mean value of the peak occurrence difference of each time window.

[0087] Furthermore, the current peak generation probability of each time window is obtained according to the peak occurrence probability and the mean of the peak occurrence difference, and the current peak generation probability is positively correlated with the peak occurrence probability and negatively correlated with the mean of the peak occurrence difference.

[0088] Specifically, the present application proposes an embodiment, in which, for each time window, the ratio of the peak occurrence probability to the peak occurrence difference mean is calculated as the current peak generation probability of each time window.

[0089] Furthermore, the present application proposes another embodiment, in which, for each time window, a calculation result of an exponential function with a natural constant as the base and the negative value of the mean of the peak occurrence difference as the exponent is calculated as a third exponent, and the product of the peak occurrence probability and the third exponent is used as the current peak generation probability of each time window.

[0090] For each time window, the product of the peak consistency coefficient and the current peak generation probability is calculated as the signal peak discrimination coefficient of each time window. The larger the signal peak discrimination coefficient of a single time window, the higher the consistency of the peak in the current time window with that in the other windows, and the greater the probability of occurrence, the more likely it is that the peak is generated by the current pulse signal caused by partial discharge of the switch cabinet.

[0091] The greater the probability of the current peak in the current time window, the higher the probability of the peak in the time window, and the smaller the difference between the peak data volume in the window and the peak data volume in other windows, the more likely the peak in the time window is a current pulse signal generated by partial discharge of the switch cabinet, and the greater the difference from the signal generated by noise, the greater the corresponding current signal peak discrimination coefficient.

[0092] At this point, the current signal peak discrimination coefficient of each time window can be obtained.

[0093] Furthermore, the partial discharge discrimination coefficient is constructed based on the current pulse discrimination coefficient and the current signal peak discrimination coefficient of each time window, and the formula is:

[0094]

[0095] In the formula, for each time window, A is the partial discharge discrimination coefficient; B is the current pulse discrimination coefficient, and C is the current signal peak discrimination coefficient. The partial discharge discrimination coefficient flow chart is as follows: Figure 2shown.

[0096] The larger the partial discharge discrimination coefficient of a single time window is, the more likely the current pulse signal in the time window is a current pulse signal generated by partial discharge during the operation of the switchgear.

[0097] At this point, the partial discharge discrimination coefficient of each time window is obtained.

[0098] Step S3, performing partial discharge monitoring on the switch cabinet according to the partial discharge discrimination coefficient of each time window.

[0099] Through the intelligent switch cabinet expert diagnosis system cloud platform, the partial discharge discrimination coefficient of all time windows of a single switch cabinet is used as input, and the segmentation threshold of the partial discharge discrimination coefficient is output by cross-validation. Cross-validation is a well-known technology and will not be described in detail. All current pulse signals are used as input, and the wavelet denoising algorithm is used to take the segmentation threshold of the partial discharge discrimination coefficient as the segmentation threshold of wavelet denoising, and the denoised current pulse signal is output.

[0100] The intelligent switch cabinet expert diagnosis system cloud platform uses the current pulse signal after denoising to monitor the operating status of the switch cabinet. When the switch cabinet has partial discharge, the intelligent switch cabinet expert diagnosis system cloud platform will send the location of the switch cabinet with partial discharge and the operating related parameters of the switch cabinet to the mobile device of the relevant staff. The relevant staff will inspect and maintain the switch cabinet according to the information sent by the platform.

[0101] Based on the same inventive concept as the above method, an embodiment of the present application also provides a switch cabinet partial discharge monitoring system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned switch cabinet partial discharge monitoring methods are implemented.

[0102] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0103] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0104] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A switch cabinet partial discharge monitoring method, characterized in that: The method comprises the following steps: Collect the current pulse signal of the switch cabinet at each sampling moment in each time window; According to the fluctuation signal characteristics, periodic signal characteristics and trend characteristics of the switch cabinet under the partial discharge state and the synchronization characteristics of the changes between local signals, the current pulse discrimination coefficient of each time window is constructed; Obtaining the peak value of the current pulse signal; obtaining the current signal peak discrimination coefficient of each time window based on the variation characteristics of the peak value of the current pulse signal; Based on the current pulse discrimination coefficient and the current signal peak discrimination coefficient, the partial discharge discrimination coefficient of each time window is obtained; Conduct partial discharge monitoring on switchgear according to partial discharge discrimination coefficient; The method for obtaining the current pulse discrimination coefficient is: Based on the amplitude change and signal transition of the current pulse signal, the signal fluctuation coefficient of each time window is obtained; Based on the periodicity and approximation of the current pulse signal change, the signal regularity coefficient of each time window is obtained; The ratio of the signal regularity coefficient to the signal fluctuation coefficient is calculated as the current pulse discrimination coefficient of each time window; The method for obtaining the current signal peak discrimination coefficient is: Based on the peak value difference degree and peak value distribution density difference of the current pulse signal in each time window, the peak value consistency coefficient of each time window is obtained; Based on the probability of each peak value appearing in all peak values ​​in each time window and the number of peak values, the probability of current peak value generation in each time window is obtained; The product of the peak consistency coefficient and the probability of the current peak generation is used as the current signal peak discrimination coefficient of each time window; The formula of the partial discharge discrimination coefficient is: A=B×C; wherein A is the partial discharge discrimination coefficient of each time window; B is the current pulse discrimination coefficient of each time window; and C is the current peak signal discrimination coefficient of each time window.

2. A switch cabinet partial discharge monitoring method according to claim 1, characterized in that: The method for obtaining the signal fluctuation coefficient is: Arrange all signal amplitude data in each time window in ascending order of their sampling time to obtain the signal amplitude sequence of each time window; For each time window, the ratio between the variance of the amplitude of the current pulse signal in the time window and the maximum amplitude is calculated as the signal amplitude variation coefficient of each window; the time interval between the maximum and minimum amplitudes in the time window is calculated, and the trend test weight of the signal amplitude sequence in the time window is calculated, and the product of the time interval and the trend test weight is used as the signal variation coefficient of each time window; According to the signal amplitude variation coefficient and the signal gradual variation coefficient, the signal fluctuation coefficient of each time window is obtained, and the signal fluctuation coefficient is positively correlated with the signal amplitude variation coefficient and negatively correlated with the signal gradual variation coefficient.

3. A switch cabinet partial discharge monitoring method as claimed in claim 2, characterized in that: The method for obtaining the signal regularity coefficient is: For each sampling moment, the signal amplitudes in all current pulse signals within a preset number of time windows before the sampling moment are arranged in time positive order to construct a first signal sequence, and the periodicity strength of the first signal sequence is calculated as the signal change periodicity coefficient at each sampling moment; Calculate the PLV value of the signal amplitude sequence of all time windows; For each time window, the sum of the PLV values ​​of the time window and all the previous time windows is calculated as the first sum, the difference between the signal amplitude sequences of the time window and other previous time windows is calculated, the mean of all the difference values ​​is calculated as the first mean, and the ratio of the first sum to the first mean is taken as the signal approximation degree of each time window; Based on the signal change periodicity coefficient and the signal approximation degree, the signal regularity coefficient of each time window is obtained, and the signal regularity coefficient is positively correlated with the signal change periodicity coefficient and the signal approximation degree respectively.

4. A switch cabinet partial discharge monitoring method according to claim 1, characterized in that: The method for obtaining the peak consistency coefficient is: For each time window, obtain the order in which the sampling moments corresponding to the peak values ​​in the time window appear in the time window; for any time window before the time window, obtain the absolute value of the difference between the amplitude and the peak value of the current pulse signal corresponding to the sampling moments of the same order; take the mean of all the absolute values ​​of the difference in any time window as the second mean of any time window; calculate the sum of the second means of all time windows before the time window as the peak difference degree; For each time window, the absolute value of the difference between the sampling time corresponding to each peak in the time window and the first sampling time of the time window in which it is located is calculated as the first absolute value of the difference, and the average of all the first absolute values ​​of the difference is used as the peak distribution density of each time window; the average of the absolute values ​​of the difference between the peak distribution density of each time window and all previous time windows is calculated as the peak distribution density difference of each time window; The peak consistency coefficient of each time window is obtained according to the peak difference degree and the peak distribution density difference. The peak consistency coefficient is positively correlated with the peak difference degree and negatively correlated with the peak distribution density difference.

5. A switch cabinet partial discharge monitoring method according to claim 1, characterized in that: The method for obtaining the probability of the current peak value generation is: For each time window, obtain the peak values ​​of all current pulse signals in each time window and a previously preset number of time windows, and construct a pair of peak values ​​of each current pulse signal in the order of the acquisition moments of the peak values ​​in the time window; for the pair of peak values ​​of each current pulse signal in each time window and a previously preset number of time windows, count the ratio of the total number of each identical pair to the total number of all pairs as the first ratio of each identical pair in the current pulse signal, and take the average of all first ratios of each time window as the probability of occurrence of the peak value of each time window; For each time window, the absolute value of the difference between the number of times the peak value of the current pulse signal in the time window appears and the number of times the peak value of the current pulse signal in other time windows is calculated as the second absolute value of the difference in each time window, and the average of all the second absolute values ​​of the difference in each time window is taken as the average of the peak value occurrence differences in each time window; The current peak generation probability of each time window is obtained according to the peak occurrence probability and the mean value of the peak occurrence difference. The current peak generation probability is positively correlated with the peak occurrence probability and negatively correlated with the mean value of the peak occurrence difference.

6. A switch cabinet partial discharge monitoring method according to claim 1, characterized in that: The partial discharge monitoring of the switch cabinet according to the partial discharge discrimination coefficient includes: Through the intelligent switchgear expert diagnosis system cloud platform, the segmentation threshold of the partial discharge discrimination coefficient of all time windows of a single switchgear is obtained; all current pulse signals are taken as input, and the denoising algorithm is used to take the segmentation threshold of the partial discharge discrimination coefficient as the segmentation threshold of denoising, and the output is the denoised current pulse signal; The intelligent switchgear expert diagnosis system cloud platform uses the current pulse signal after denoising to monitor the operating status of the switchgear.

7. A switch cabinet partial discharge monitoring system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the switch cabinet partial discharge monitoring method as described in any one of claims 1 to 6 are implemented.

Images