Discharge detection system and detection method of high and low voltage switch cabinet

Abstract

The invention relates to the technical field of discharge detection, in particular to a discharge detection system and detection method for a high-low voltage switch cabinet, and the method comprises the steps: collecting the multi-dimensional data of the switch cabinet, including the transient ground voltage, and the temperature and humidity in the switch cabinet; calculating a similarity degree between any data and a plurality of adjacent data; calculating a smoothing coefficient of each piece of data, and determining a predicted value of each piece of data; wherein the order sum of the data i and the data, the similarity degree of which is smaller than a set threshold value, in the previous data adjacent to the data i is determined, and the smoothing coefficient of the data i is in positive correlation with the order sum; for one-dimensional data, the difference between the predicted value and the actual value of each data is calculated, and then the abnormal degree of the one-dimensional data is calculated; calculating the importance degree of the dimension data, wherein the importance degree is in positive correlation with the difference of adjacent data; further determining a weighted anomaly degree; and judging that the discharge is abnormal when the weighted abnormal degree is greater than an abnormal threshold value. According to the invention, the accuracy of discharge detection is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of discharge detection, and also to a discharge detection system and a detection method for high and low voltage switch cabinets. Background Art

[0002] High and low voltage switchgear is used to distribute high and low voltage electric energy from power plants, substations or transformers to different loads. Through the switches and circuit breakers inside it, the power system can be flexibly organized to achieve the distribution of electric energy among various users.

[0003] High and low voltage switchgear has perfect protection function, which can monitor and protect the electric energy from overvoltage, undervoltage, overcurrent, short circuit and other problems. When the power system fails, the high and low voltage switchgear can quickly cut off the fault circuit to avoid the expansion of the fault.

[0004] High and low voltage switchgear can also control electric energy and realize the switching, connection and switching of electric energy. Through its internal switch devices and control circuits, the power system can be easily controlled to realize the switching and control of various circuits and loads.

[0005] High and low voltage switchgear can be equipped with various power monitoring and measuring equipment, such as voltmeters, ammeters, power factor meters, etc. These devices can monitor the load conditions, voltage, current and other parameters of power in real time, and display and record them through data transmission.

[0006] In short, high and low voltage switchgear undertakes important electrical functions such as voltage control at all levels in power generation, distribution and transmission, and is widely used in various parts of the power system. Its operating conditions directly affect the safety of the power grid.

[0007] During the operation of high and low voltage switch cabinets, discharge is prone to occur due to the large number of devices in the cabinet, and discharge will accelerate the degradation of insulation performance. The patent application document with application publication number CN116298737A is named a switch cabinet discharge monitoring system, method and device. The detection method includes: realizing the discharge detection of the switch cabinet according to a set fixed threshold. However, using this discharge detection method may not detect certain abnormal discharge conditions, which leads to low accuracy in judging whether the discharge is abnormal, that is, low accuracy of discharge detection. Summary of the invention

[0008] In order to solve the above-mentioned technical problem of low accuracy in judging whether discharge is abnormal, the present invention proposes a discharge detection system and a detection method for a high and low voltage switch cabinet.

[0009] To this end, the present invention provides solutions in the following aspects.

[0010] In a first aspect of an embodiment of the present invention, a discharge detection method for a high and low voltage switch cabinet is provided, comprising: Collect multi-dimensional data of the switch cabinet, including transient ground voltage, temperature and humidity inside the switch cabinet; Calculate the similarity between any data and multiple adjacent data; Calculate the smoothing coefficient of each data, and then determine the predicted value of each data; the calculation formula of the smoothing coefficient is: ,in, is the smoothing coefficient of data i in the b-th dimension data, is the preset smoothing coefficient reference value; is the sum of the order of the data i and the previous set number H of data adjacent to it whose similarity is less than the set threshold, is the sum of the order of the previous set number H of data adjacent to the data i; For one-dimensional data: calculate the difference between the predicted value and the actual value of each data, and then calculate the abnormality of the dimensional data, which is positively correlated with the difference; calculate the importance of the dimensional data, which is positively correlated with the difference of adjacent data; The formula for weighted abnormality degree Y is: , n is the total number of dimensions, is the importance of the b-th dimension data, is the abnormality degree of the b-th dimension data; In response to the weighted abnormality degree Y being greater than the abnormality threshold, it is determined that the discharge is abnormal.

[0011] Considered The influence of the smoothing coefficient of data i of the b-th dimension data improves the accuracy of the prediction of data i. The reason is that multiple data with a small degree of similarity and close proximity have a more obvious trend, and the weight of these multiple data should be larger during prediction.

[0012] Optionally, the importance is calculated as: ; in, is the importance of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the previous data adjacent to the data i, is an absolute value.

[0013] The larger it is, the greater the fluctuation of the b-dimensional data, the greater the possibility of abnormal data in the b-dimensional data, and the greater the importance of the b-dimensional data.

[0014] Optionally, the importance is calculated as: ; in, is the importance of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the previous data adjacent to the data i.

[0015] The larger the value is, the greater the possibility that there are abnormal data in the b-th dimension data.

[0016] Optionally, the calculation formula for the degree of abnormality is: ; in, is the abnormality degree of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the predicted value of the data i.

[0017] The larger it is, the less the change of the b-th dimension data (actual value) conforms to the change trend, the greater the abnormality of the b-th dimension data, and the greater the possibility of abnormal discharge.

[0018] Optionally, the calculation formula for the degree of abnormality is: ; in, is the abnormality degree of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the predicted value of the data i.

[0019] Optionally, the similarity calculation formula is: ; in, is the similarity degree corresponding to the data i, is the positive correlation normalization function, For the data i, is the j-th data among the k data adjacent to the data i, and k is a preset value.

[0020] The bigger, the The closer the adjacent data are to each other (the smaller the difference is).

[0021] Optionally, the similarity calculation formula is: ; in, is the similarity degree corresponding to the data i, is the positive correlation normalization function, For the data i, is the j-th data among the k data adjacent to the data i, and k is a preset value.

[0022] In a second aspect of an embodiment of the present invention, a discharge detection system for a high and low voltage switchgear is provided, comprising: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a discharge detection method for a high and low voltage switchgear described in any one of the first aspects of the embodiment of the present invention is implemented.

[0023] Beneficial effects of the present invention: The larger the value, the greater the possibility that the multiple data with a smaller degree of similarity and neighbors are closer to the data i of the b-dimensional data. Since the multiple data with a smaller degree of similarity and neighbors have a more obvious trend, the weight of the multiple data should be larger when calculating the predicted value of data i to improve the accuracy of the prediction of data i. For any dimensional data: the greater the difference between the predicted value and the actual value of each data, the less consistent the change of the dimensional data (actual value) is with the change trend, and the greater the possibility of abnormal discharge; the greater the difference between adjacent data, the greater the fluctuation of the dimensional data, and the greater the possibility of abnormal discharge. The abnormality degree of the corresponding dimension is weighted by the importance of each dimensional data, so that the calculated weighted abnormality degree reflects the accuracy of whether there is abnormal discharge, that is, the accuracy of discharge detection is improved.

[0024] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 is a flow chart schematically showing a discharge detection method for a high and low voltage switchgear according to an embodiment of the present invention; Figure 2 The figure is a schematic structural block diagram schematically showing a discharge detection system for a high and low voltage switchgear according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0027] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0028] During the operation of high and low voltage switchgear, discharge is prone to occur due to the large number of devices in the cabinet, which will accelerate the degradation of insulation performance. When the insulation performance drops to a certain level, it will cause the insulation material to fail. It should be noted that insulation materials with lower insulation performance will lead to greater discharge intensity.

[0029] The smaller the performance of the insulation material, the lower the safety of the equipment operation. Therefore, the discharge should be detected. If abnormal discharge is detected, an early warning should be issued to remind relevant personnel to carry out maintenance.

[0030] Generally speaking, the present invention calculates the difference between the predicted value and the actual value of the data, and calculates the degree of fluctuation of the data, and determines whether the discharge is abnormal according to the difference between the predicted value and the actual value and the degree of fluctuation. The greater the difference between the predicted value and the actual value, the greater the possibility of abnormal discharge. The greater the degree of fluctuation of the data, the greater the possibility of abnormal discharge (under normal circumstances, the degree of fluctuation of the data is small).

[0031] Figure 1 FIG. 1 is a flow chart schematically showing a discharge detection method for a high and low voltage switch cabinet according to an embodiment of the present invention. Figure 1 The specific instructions are as follows: S1: Collect multi-dimensional data of the switch cabinet, including transient ground voltage, temperature and humidity inside the switch cabinet.

[0032] In one embodiment, a transient ground voltage sensor is arranged on the switch cabinet, and based on the transient ground voltage detection technology, the transient ground voltage (also referred to as discharge intensity in this article) of the switch cabinet is collected through the transient ground voltage sensor. The temperature inside the switch cabinet is collected through the temperature sensor, and the humidity inside the switch cabinet is collected through the humidity sensor.

[0033] Schematically: ((A1, B1, C1), (A2, B2, C2), ..., (Ai, Bi, Ci), ..., (Abm, Bbm, Cbm)), where (Ai, Bi, Ci) is the i-th data point, Ai is the i-th transient voltage data, Bi is the i-th temperature data, Ci is the i-th humidity data, and (Abm, Bbm, Cbm) can be considered as the data point corresponding to the current moment. The time of the data in the same data point (the time can be considered as the acquisition time) is the same, for example, the time corresponding to Ai, Bi, and Ci is the same. The intervals of the moments of adjacent data points are the same, for example, for a data sequence of any dimension, such as a transient voltage sequence: (A1, A2, ..., Abm), the time intervals of adjacent data are the same. Temperature sequence: (B1, B2, ..., Bbm). Humidity sequence: (C1, C2, ..., Cbm).

[0034] Those skilled in the art can set the time interval and the total number of data by themselves.

[0035] It should be noted that transient ground voltage detection technology is a technology for detecting insulation defects inside power equipment. It is widely used in the detection of internal insulation defects in power distribution equipment such as switch cabinets and ring network cabinets. When partial discharge occurs inside the switch cabinet, a small arc will form in the insulating medium of the equipment, which will cause a sharp redistribution of charge and generate transient electromagnetic waves. This electromagnetic wave propagates to the metal casing of the equipment and forms a transient ground voltage signal through ground capacitance coupling.

[0036] S2: Calculate the similarity between any data and multiple adjacent data.

[0037] It should be noted that there will be some normal discharge phenomena during the operation of the switch cabinet. This type of discharge is usually instantaneous (non-continuous discharge) and will not affect the normal operation of the equipment.

[0038] The intensity of discharge caused by aging, damage or manufacturing defects of equipment insulation materials is usually higher. This type of discharge is usually continuous discharge, and the intensity of this type of discharge usually increases over time. The reason is that discharge will accelerate the performance degradation of insulation materials, and materials with lower insulation performance will lead to greater discharge intensity. This type of discharge will cause damage to the equipment.

[0039] Therefore, the present invention first calculates the similarity between each data and multiple adjacent data, wherein the trend of multiple data with smaller similarity and close proximity is more obvious. In the process of data prediction, for a data P to be predicted, if the similarity of multiple data close to the data P is smaller, the weight of the multiple data should be greater.

[0040] In one embodiment, the calculation formula for the similarity is: ; in, is the degree of similarity corresponding to data i (i-th data) in the b-th dimension data, is the positive correlation normalization function, is the data i (actual value), is the j-th data (actual value) among the k data adjacent to the data i, and k is a preset value.

[0041] Those skilled in the art can set k by themselves. For example, k can be 4.

[0042] The positive correlation normalization function in this article can be the maximum and minimum normalization function of the positive correlation. Explanation: For one-dimensional data, the i-th data in the dimension Denoted as the difference A of the data, calculate the difference A of each data in the dimensional data, select the difference A of multiple abnormal data in the same type of historical data, where the similarity of the abnormal data is small, and for the abnormal data in the transient ground voltage data, it is the transient ground voltage data corresponding to the abnormal discharge moment; select the difference A of multiple non-abnormal data in the same type of historical data, the maximum value in the normalized function is the maximum value of all differences A, and the minimum value in the normalized function is the minimum value of all differences A. The other normalization processes in this article are similar and will not be repeated unless otherwise specified.

[0043] In one embodiment: By default There are multiple data before, and The adjacent 4 data are The first 2 data and the last 2 data; for the last data, the 4 data adjacent to the data are the 4 data before the data.

[0044] The bigger, the The closer the adjacent data are to each other (the smaller the difference is).

[0045] In another embodiment, the calculation formula of the similarity is: ; in, is the similarity degree corresponding to the data i, is the positive correlation normalization function, For data i, is the j-th data among the k data adjacent to data i, where k is a preset value.

[0046] So far, the similarity of each data has been obtained.

[0047] S3: Calculate the smoothing coefficient of each data, and then determine the predicted value of each data.

[0048] It should be noted that the data series collected based on transient ground voltage technology is a time series, in which each data represents the discharge intensity at a time point. During the discharge process, the change in discharge intensity usually has a certain trend. Therefore, the exponential smoothing method can be used to predict the data in the time series, so as to judge whether the discharge is abnormal based on the prediction results.

[0049] It should be further explained that, in the process of predicting data using the exponential smoothing method, for a piece of data L1 to be predicted, the larger the smoothing coefficient, the greater the weight of the data closer to the data L1.

[0050] In one embodiment, the calculation formula of the smoothing coefficient is: ; in, is the smoothing coefficient of data i in the b-th dimension data, is a preset smoothing coefficient reference value; for the data i, determine the first preset number H (may be 10) of data adjacent to the data i, It is the sum of the orders of the data whose similarity is less than the set threshold value among the preset number H data. For example, among the preset number H data, the order of the first data is 1, and the order of the last data is H. , It is the sum of the order of the set number H data.

[0051] In one embodiment, in order to achieve a better prediction effect, when the calculated smoothing coefficient of a certain data is less than the smoothing coefficient threshold (which may be 0.15), the smoothing coefficient of the data is replaced with the smoothing coefficient threshold.

[0052] The technology in this field can be set by yourself , set the threshold, smoothing coefficient threshold and preset number K. It can be 0.5, and the threshold can be set to 0.25.

[0053] The larger the value is, the greater the possibility that the multiple data with a smaller degree of similarity and a closer distance to the data i of the b-th dimension is. Since the multiple data with a smaller degree of similarity and a closer distance have a more obvious trend, the weight of the multiple data should be larger when calculating the predicted value of data i, that is, It should be larger to improve the accuracy of prediction of data i.

[0054] Further explanation is as follows: when the discharge of the switch cabinet is abnormal, as the degree of damage to the insulating material increases, the discharge intensity increases accordingly. For time i, when the insulation performance of the insulating material suddenly decreases, the discharge intensity will suddenly increase. The similarity of the discharge intensity data at the previous multiple moments adjacent to time i is small, and the discharge intensity data at the previous multiple moments adjacent to time i has an obvious upward trend. In order to improve the accuracy of the discharge intensity prediction at time i, so that the difference between the predicted value and the true value of the discharge intensity data at time i is small, the weight (degree of attention) of the discharge intensity data at the previous multiple moments adjacent to time i should be large, that is, the smoothing coefficient of the discharge intensity data at time i should be large. If the difference between the discharge intensity data (actual value) at time i and the predicted value of the data obtained later is still large, the abnormality of the discharge intensity data at time i is large (that is, the possibility of discharge abnormality is large).

[0055] When is small, there may be individual (possibly 1) outlier data in the first multiple data adjacent to data i of the b-th dimension data. The outlier data may be caused by factors such as voltage fluctuations in the switch cabinet (e.g., a sudden increase in voltage), and the outlier data may be close to the data i. The smoothing coefficient of the data i should be small to reduce the weight of the data close to the data i, and mainly rely on the data far from the data i to determine the predicted value of the data i, so as to improve the accuracy of the prediction of the data i. It should be noted that the degree of similarity corresponding to the outlier data is small (i.e., the difference between the outlier data and the multiple data adjacent to it is large), that is, the degree of similarity of the outlier data is less than the set threshold. When the distance between the outlier data and the data i is close, the smoothing coefficient of the data i should be small.

[0056] For example, Denoted as the adjustment factor corresponding to data i in the b-th dimension data, when the adjustment factor corresponding to the discharge intensity data at time i is small, there may be individual outlier data in the first multiple data adjacent to the discharge intensity data at time i, and the outlier data may be close to the discharge intensity data at time i, and the smoothing coefficient of the discharge intensity data at time i should be small. It should be noted that the insulation performance of the insulating material at the time corresponding to the outlier data is still high, and the discharge corresponding to the outlier data will not affect the normal operation of the equipment.

[0057] The discharge intensity is related to the temperature inside the switch cabinet. The greater the discharge intensity, the greater the possibility that the temperature inside the switch cabinet is higher. As the temperature rises, the relative humidity inside the switch cabinet will decrease because the air can accommodate more water vapor. The discharge intensity, temperature and humidity are correlated. The explanation of the calculation process of the smoothing coefficient of temperature data and humidity data is the same as the explanation of the calculation process of the smoothing coefficient of the discharge intensity mentioned above, so it will not be repeated. It should be noted that the humidity mentioned in this article is relative humidity.

[0058] Using the exponential smoothing method (existing technology), the predicted value of each data is obtained according to the smoothing coefficient of each data in each dimension. In one embodiment, the exponential smoothing method is a first-order exponential smoothing method. Schematically, ..., , ,……,in, is the predicted value of data i-1 in the b-th dimension data, is the actual value of data i-1 in the b-th dimension data, is the smoothing coefficient of data i in the b-th dimension data, is the predicted value of data i-2 in the b-th dimension data, is the actual value of the data i-2, is the smoothing coefficient of the data i-1.

[0059] S4: Calculate the abnormality degree corresponding to each dimension. Specifically, for one-dimensional data: calculate the difference between the predicted value and the actual value of each data, and then calculate the abnormality degree of the dimensional data, and the abnormality degree is positively correlated with the difference.

[0060] In one embodiment, the calculation formula of the abnormality degree is: ; in, is the abnormality degree of the b-th dimension data, bm is the total number of b-th dimension data, is the data i of the b-th dimension, is the predicted value of the data i. Indicates the difference between the predicted value and the actual value of the b-th dimension data.

[0061] The larger it is, the less the change of the b-th dimension data (actual value) conforms to the change trend, the greater the abnormality of the b-th dimension data, and the greater the possibility of abnormal discharge.

[0062] In another embodiment, the calculation formula of the abnormality degree is: ; in, is the abnormality degree of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the predicted value of the data i.

[0063] S5: Calculate the importance of each dimension. Specifically, for one-dimensional data: calculate the importance of the dimensional data, and the importance is positively correlated with the difference of adjacent data.

[0064] In one embodiment, the calculation formula of importance is: ; in, is the importance of the b-th dimension data, bm is the total number of b-th dimension data, is the data i of the b-th dimension, is the previous data adjacent to the data i (data i-1 of the b-th dimension data).

[0065] Indicates the difference between adjacent data. The larger the value is, the greater the fluctuation of the b-th dimension data is, and the greater the possibility of abnormal data in the b-th dimension data is, that is, the greater the possibility of abnormal discharge is, the greater the importance of the b-th dimension data is. For example, the greater the fluctuation of the transient voltage sequence is, the greater the possibility of abnormal transient voltage data in the transient voltage sequence is.

[0066] In another embodiment, the calculation formula of importance is: ; in, is the importance of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the previous data adjacent to the data i, is an absolute value.

[0067] S6: Calculate the weighted abnormality degree.

[0068] In one embodiment, the formula for weighted abnormality level Y is: , n is the total number of dimensions, which is 3, is the importance of the b-th dimension data, is the abnormality degree of the b-th dimension data.

[0069] right The explanation is as follows: For example, the calculation process of the normalized value of the importance of the transient ground voltage data sequence collected in this embodiment is: the importance (including greater importance and lesser importance) of multiple historical transient ground voltage data sequences (sequences with abnormal discharge data and sequences without abnormal discharge data) in the historical sequence are selected, and the maximum value and minimum value in the maximum and minimum normalized function are respectively the maximum value and minimum value of the importance selected above and the importance of the transient ground voltage data sequence collected in this embodiment. The explanation is the same and I will not repeat it here.

[0070] In the formula The role of Normalization.

[0071] Based on the above analysis, The larger it is, the greater the possibility of abnormal discharge.

[0072] The abnormality degree of the corresponding dimension is weighted by the importance of the data of each dimension, so that the calculated weighted abnormality degree can more accurately reflect whether there is abnormal discharge.

[0073] In one embodiment, the formula for weighted abnormality level Y is: , n is the total number of dimensions, which is 3, is the importance of the b-th dimension data, is the abnormality degree of the b-th dimension data.

[0074] S7: In response to the weighted abnormality degree Y being greater than the abnormality threshold, it is determined that the discharge is abnormal.

[0075] The greater the weighted abnormality degree Y is, the greater the possibility of abnormal discharge in the switch cabinet. In response to the weighted abnormality degree Y being greater than the abnormal threshold, it is judged that the discharge is abnormal, that is, abnormal discharge is detected, and an early warning should be issued to remind relevant personnel to perform maintenance.

[0076] Those skilled in the art can set the abnormal threshold by themselves, and the abnormal threshold can be taken as an empirical value of 0.75.

[0077] Figure 2 FIG. 2 is a schematic diagram schematically showing a structural block diagram of a discharge detection system for a high and low voltage switch cabinet according to an embodiment of the present invention. Figure 2 As shown, 01 is a communication bus. The present invention also provides a discharge detection system for high and low voltage switch cabinets, the system comprising a processor and a memory, the memory storing computer program instructions, and when the computer program instructions are executed by the processor, the discharge detection method for high and low voltage switch cabinets according to the present invention is implemented.

[0078] The system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface, whose configuration and functions are known in the art and thus will not be described in detail here.

[0079] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0080] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0081] Additionally, the articles "a" and "an" as used in this disclosure and the appended claims are generally construed to mean "one or more" unless specified otherwise or clear from context to be directed to a singular form.

[0082] Where the terms "includes," "having," "having," "having," or variations thereof are used in the detailed description or claims, such terms are intended to be inclusive in a manner similar to the term "comprising."

[0083] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as S1, S2, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that if descriptions such as "first", "second", "A1", "B1" appear in this article, their role is to distinguish different messages, devices, modules, etc., and do not represent the order of precedence, do not have any technical meaning, and do not limit "A1", "B1", "first" and "second" to different types. For example: the first feature of an image is the feature of the image. The role of adding "first" includes facilitating subsequent references to different objects, that is, facilitating indexing.

[0084] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A discharge detection method for high and low voltage switchgear, characterized in that: include: Collect multi-dimensional data of the switch cabinet, including transient ground voltage, temperature and humidity inside the switch cabinet; Calculate the similarity between any data and multiple adjacent data; Calculate the smoothing coefficient of each data, and then determine the predicted value of each data; the calculation formula of the smoothing coefficient is: ,in, is the smoothing coefficient of data i in the b-th dimension data, is the preset smoothing coefficient reference value; is the sum of the order of the data i and the previous set number H of data adjacent to it whose similarity is less than the set threshold, is the sum of the order of the previous set number H of data adjacent to the data i; For one-dimensional data: calculate the difference between the predicted value and the actual value of each data, and then calculate the abnormality of the dimensional data, which is positively correlated with the difference; calculate the importance of the dimensional data, which is positively correlated with the difference of adjacent data; the formula for weighted abnormality Y is: , n is the total number of dimensions, is the importance of the b-th dimension data, is the abnormality of the b-th dimension data, is the positive correlation normalization function; In response to the weighted abnormality degree Y being greater than the abnormality threshold, it is determined that the discharge is abnormal.

2. The discharge detection method for high and low voltage switch cabinet according to claim 1 is characterized in that: The calculation formula for importance is: ； in, is the importance of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the previous data adjacent to the data i.

3. The discharge detection method for high and low voltage switch cabinet according to claim 1 is characterized in that: The calculation formula for importance is: ； in, is the importance of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the previous data adjacent to the data i.

4. The discharge detection method for high and low voltage switch cabinet according to claim 1 is characterized in that: The calculation formula for the degree of abnormality is: ； in, is the abnormality degree of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the predicted value of the data i.

5. The discharge detection method for high and low voltage switch cabinet according to claim 1 is characterized in that: The calculation formula for the degree of abnormality is: ； in, is the abnormality degree of the b-th dimension data, bm is the total number of b-th dimension data, For the data i, is the predicted value of the data i.

6. The discharge detection method for high and low voltage switch cabinet according to claim 1 is characterized in that: The calculation formula for the degree of similarity is: ； in, is the similarity degree corresponding to the data i, For the data i, is the j-th data among the k data adjacent to the data i, and k is a preset value.

7. The discharge detection method for high and low voltage switch cabinet according to claim 1 is characterized in that: The calculation formula for the degree of similarity is: ； in, is the similarity degree corresponding to the data i, For the data i, is the j-th data among the k data adjacent to the data i, and k is a preset value.

8. The discharge detection system of high and low voltage switchgear is characterized by: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the discharge detection method for the high and low voltage switchgear according to any one of claims 1 to 7 is implemented.

Images