A monitoring method and system for electrical equipment

Abstract

The application relates to the technical field of data processing, in particular to a monitoring method and system for electrical equipment, which comprises real-time data of any monitoring index of the electrical equipment and historical data in a historical period to form a data sequence, obtains three abnormal characteristic values of each data in the data sequence according to the change law of each data under different conditions, obtains similar historical data similar to any abnormal characteristic value of the real-time data for any abnormal characteristic value, obtains the abnormal reaction degree of any abnormal characteristic value according to the correlation between the real-time data and each similar historical data under other abnormal characteristic values, obtains the abnormal degree of any monitoring index according to the abnormal reaction degree of each abnormal characteristic value and the three abnormal characteristic values of each data in the data sequence, and then performs abnormal early warning on the electrical equipment according to the abnormal degree of each monitoring index, thereby improving the accuracy of abnormal early warning on the electrical equipment.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a monitoring method and system for electrical equipment. Background Art

[0002] Electrical equipment is the core device in the power system, playing a key role in the control, transmission, and use of electrical energy. Monitoring and management of electrical equipment is a crucial step in its operation. Analyzing monitoring data from electrical equipment provides timely warnings of abnormalities and faults, thereby ensuring the safety of equipment and personnel, maintaining system stability, and improving power system operation and maintenance efficiency.

[0003] Existing electrical equipment monitoring methods often set fixed anomaly thresholds for each piece of monitoring data, triggering alarms when data exceeds these thresholds, or employ fixed anomaly recognition models to provide early warnings of electrical equipment status. However, numerous factors can influence electrical equipment monitoring data, such as sensor status and ambient temperature. Fixed anomaly thresholds or anomaly recognition models often struggle to adapt to varying data patterns under varying circumstances, resulting in low accuracy in warnings of electrical equipment anomalies.

[0004] Therefore, how to improve the accuracy of abnormal warning for electrical equipment has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, an embodiment of the present invention provides a monitoring method and system for electrical equipment to solve the problem of how to improve the accuracy of abnormality warning for electrical equipment.

[0006] In a first aspect, an embodiment of the present invention provides a method for monitoring an electrical device, the method comprising the following steps:

[0007] For any monitoring indicator of the electrical equipment, in the operation log of the electrical equipment, the real-time data of the monitoring indicator at the current moment and the historical data of the monitoring indicator in a preset historical period before the current moment are combined into a data sequence. Based on the local fluctuation degree and local change trend of each data in the data sequence, three abnormal characteristic values ​​of each data are obtained;

[0008] For any abnormal characteristic value, based on the similarity between the real-time data and each historical data, similar historical data similar to the abnormal characteristic value of the real-time data is obtained; based on the correlation between the real-time data and each similar historical data under other abnormal characteristic values, the degree of abnormal reaction of the abnormal characteristic value of the real-time data to the electrical equipment is obtained;

[0009] According to the abnormal response degree of each abnormal characteristic value of the real-time data and the three abnormal characteristic values ​​of each data in the data sequence, the abnormal degree of any monitoring indicator is obtained, the abnormal degree of each monitoring indicator of the electrical equipment is obtained, and according to the abnormal degree of all monitoring indicators, an abnormal warning is issued for the electrical equipment.

[0010] Preferably, the three abnormal characteristic values ​​of each data are obtained according to the local fluctuation degree and local change trend of each data in the data sequence, including:

[0011] According to the local change trend of each data in the data sequence, the target time interval of each data in the data sequence is obtained;

[0012] For any data in the data sequence, the time period of the data is obtained according to the target time interval, and the moment corresponding to the data is the last moment of the time period of the data;

[0013] If the time period between the first data in the data sequence and the any data is greater than or equal to the time period of the any data, then obtaining a subsequence of the any data in the data sequence according to the time period of the any data;

[0014] If the time period between the first data in the data sequence and the any data is shorter than the time period of the any data, then obtaining a subsequence of the any data in the operation log of the electrical device according to the time period of the any data;

[0015] Obtaining a first-order difference sequence of a subsequence of the any data, accumulating the absolute values ​​of the differences between each two adjacent data in the first-order difference sequence to obtain an accumulated value, and performing linear normalization on the accumulated value to obtain a fluctuation characteristic value of the any data;

[0016] Obtaining a change trend characteristic value of the any data according to a data change trend in a subsequence of the any data;

[0017] The any data, the fluctuation characteristic value of the any data, and the change trend characteristic value of the any data are used as three abnormal characteristic values ​​of the any data.

[0018] Preferably, obtaining the target time interval of each data in the data sequence according to the local change trend of each data in the data sequence includes:

[0019] For any data in a data sequence, if the number of data before the data in the data sequence is greater than or equal to a first preset number, and the number of data after the data is greater than or equal to the first preset number, the data is recorded as data to be analyzed, and a window of a preset length is constructed in the data sequence with the data to be analyzed as the center, where the first preset number is half of the difference between the preset length and a constant of 1;

[0020] Calculating a first difference between the data to be analyzed and the first data in the window, calculating a second difference between the last data in the window and the data to be analyzed, and using the absolute value of the difference between the first difference and the second difference as the degree of trend change of the data to be analyzed;

[0021] Obtain the degree of trend change of each of the to-be-analyzed data in the data sequence, sort all trend change degrees in descending order to obtain a descending sequence, record the to-be-analyzed data corresponding to the first second preset number of trend change degrees in the descending sequence as target data, sort all target data in ascending order according to the moment corresponding to each of the target data to obtain a target data sequence, calculate the time interval between each two adjacent target data in the target data sequence, select the minimum value among all time intervals and record it as the target time interval for each data in the data sequence.

[0022] Preferably, obtaining the change trend characteristic value of any data according to the data change trend in the subsequence of any data includes:

[0023] Calculating the absolute value of the difference between the any data and the first data in the subsequence of the any data to obtain the change amplitude of the any data;

[0024] Calculating a difference between the any data and the first data in a subsequence of the any data, calculating a sum of the change amplitude and a preset constant to obtain a sum result, and using a ratio between the difference and the sum result as a change trend value of the any data;

[0025] The product of the linear normalized value of the change amplitude and the change trend value is calculated to obtain the change trend characteristic value of any data.

[0026] Preferably, obtaining similar historical data similar to any abnormal feature value of the real-time data based on the similarity performance of any abnormal feature value between the real-time data and each historical data includes:

[0027] For any historical data, calculate the absolute value of the difference between the any abnormal feature value of the any historical data and real-time data, obtain a feature value difference index, linearly normalize the reciprocal of the sum between the feature value difference index and a preset constant, and obtain the similarity degree of the real-time data and the any historical data under the any abnormal feature value.

[0028] Obtain the similarity degree of the real-time data and each historical data under the any abnormal feature value, and in all similarity degrees, the historical data corresponding to the similarity degree greater than a preset similarity degree threshold is regarded as the similar historical data of the real-time data.

[0029] Preferably, the abnormal reaction degree of the any abnormal feature value of the real-time data to the electrical equipment is obtained according to the correlation performance of the real-time data and each similar historical data under other abnormal feature values, and the abnormal reaction degree of the any abnormal feature value of the real-time data to the electrical equipment is obtained according to the correlation performance of the real-time data and each similar historical data under other abnormal feature values.

[0030] For any similar historical data, in a data sequence, a third preset number of data closest to the time corresponding to the any similar historical data is obtained, and is recorded as reference data.

[0031] The time corresponding to the any similar historical data is recorded as a target historical time, and the time corresponding to each reference data is recorded as a reference time.

[0032] The absolute value of the difference between the data corresponding to each monitoring index at the target historical time and any reference time is calculated respectively, and all absolute values of the difference corresponding to the target historical time and the any reference time are accumulated to obtain the overall mutation degree between the target historical time and the any reference time in all monitoring indexes.

[0033] The overall mutation degrees corresponding to the target historical time and each reference time are accumulated to obtain a noise possibility index of the any similar historical data.

[0034] The abnormal reaction degree of the any abnormal feature value of the real-time data to the electrical equipment is obtained according to the correlation performance of the real-time data and each similar historical data under other abnormal features, and the noise possibility index of each similar historical data.

[0035] Preferably, the abnormal reaction degree of the any abnormal feature value of the real-time data to the electrical equipment is obtained according to the correlation performance of the real-time data and each similar historical data under other abnormal features, and the noise possibility index of each similar historical data.

[0036] For any other abnormal feature value, the degree of similarity between the real-time data and each similar historical data under the any other abnormal feature value is obtained based on the absolute value of the difference between each similar historical data and the real-time data, and recorded as the degree of association between the real-time data and each similar historical data under the any other abnormal feature value;

[0037] Calculating the difference between the constant 1 and the noise possibility index of each similar historical data respectively to obtain the weight of the correlation degree corresponding to each similar historical data;

[0038] According to the weight of the correlation degree corresponding to each of the similar historical data, a weighted sum is performed on the correlation degree corresponding to each of the similar historical data to obtain a correlation index between the any one abnormal feature value of the real-time data and any one of the other abnormal feature values;

[0039] The correlation index between any one abnormal characteristic value of the real-time data and each other abnormal characteristic value is obtained respectively, and the maximum value among all the correlation indices is selected as the abnormal reaction degree of any one abnormal characteristic value of the real-time data to the electrical equipment.

[0040] Preferably, the abnormal reaction degree of each abnormal characteristic value of the real-time data and the three abnormal characteristic values ​​of each data in the data sequence are used to obtain the abnormal degree of any monitoring indicator, including:

[0041] Mapping the three abnormal characteristic values ​​of each data in the data sequence to a three-dimensional coordinate system, wherein the three coordinate axes of the three-dimensional coordinate system respectively represent the three abnormal characteristic values ​​of each data in the data sequence;

[0042] In the three-dimensional coordinate system, respectively calculating the Euclidean distance between the real-time data and each historical data, and selecting the historical data corresponding to the fourth preset number of smallest Euclidean distances among all the Euclidean distances as the target historical data;

[0043] For any target historical data, respectively calculate the square of the distance between the real-time data and the any target historical data on each coordinate axis to obtain the square term of each coordinate axis, use the abnormal reaction degree of the abnormal characteristic value corresponding to each coordinate axis as the weight of the square term of each coordinate axis, perform weighted summation on the square terms of each coordinate axis to obtain a weighted summation result, and use the arithmetic square root of the weighted summation result as the weighted distance between the real-time data and the any target historical data;

[0044] The weighted distances between the real-time data and each of the target historical data are calculated respectively, and the average values ​​of all the weighted distances are linearly normalized to obtain the abnormality degree of any monitoring indicator.

[0045] Preferably, the abnormality warning of electrical equipment according to the abnormality degree of all monitoring indicators includes:

[0046] For the abnormality level of any monitoring indicator, if the abnormality level of any monitoring indicator is greater than or equal to the preset abnormality level threshold of any monitoring indicator, an abnormality warning is issued for any monitoring indicator.

[0047] In a second aspect, an embodiment of the present invention further provides a monitoring system for electrical equipment, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, it implements a monitoring method for electrical equipment as described in the first aspect.

[0048] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0049] The present invention is for any monitoring indicator of electrical equipment. In the operation log of the electrical equipment, the real-time data of any monitoring indicator at the current moment and the historical data of any monitoring indicator in a preset historical period before the current moment are combined into a data sequence. According to the local fluctuation degree and local change trend of each data in the data sequence, three abnormal characteristic values ​​of each data are obtained; for any abnormal characteristic value, according to the similarity performance of any abnormal characteristic value between the real-time data and each historical data, similar historical data similar to the any abnormal characteristic value of the real-time data is obtained; according to the correlation performance between the real-time data and each similar historical data under other abnormal characteristic values, the abnormal reaction degree of any abnormal characteristic value of the real-time data to the electrical equipment is obtained; according to the abnormal reaction degree of each abnormal characteristic value of the real-time data and the three abnormal characteristic values ​​of each data in the data sequence, the abnormal degree of any monitoring indicator is obtained, the abnormal degree of each monitoring indicator of the electrical equipment is obtained, and according to the abnormal degree of all monitoring indicators, an abnormal warning is issued to the electrical equipment. Among them, since the factors affecting the data of various monitoring indicators of electrical equipment are different, by analyzing the change rules of each data under each monitoring indicator under different situations, multiple abnormal characteristic values ​​of each data are obtained to reflect the abnormal performance of the data under different situations, thereby improving the accuracy of abnormal assessment of electrical equipment; according to the similarity between real-time data and historical data under each abnormal characteristic value, the abnormal response degree of each abnormal characteristic value to the electrical equipment is obtained to more accurately reflect the real abnormal state of the electrical equipment, and then combined with the multiple abnormal characteristic values ​​of each data under each monitoring indicator and the abnormal response degree of each abnormal characteristic value, abnormal early warning of the electrical equipment is performed, thereby improving the accuracy of abnormal early warning of the electrical equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 This is a flow chart of a method for monitoring electrical equipment provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0052] The following describes in detail embodiments of the present disclosure, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0053] It should be noted that the terms "first," "second," and the like in the specification of the present disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0054] In order to illustrate the technical solution of the present invention, specific embodiments are provided below.

[0055] See also Figure 1 , is a flow chart of a method for monitoring an electrical device provided by the first embodiment of the present invention, such as Figure 1 As shown, the method may include:

[0056] Step S101: For any monitoring indicator of the electrical equipment, in the operation log of the electrical equipment, the real-time data of any monitoring indicator at the current moment and the historical data of any monitoring indicator in a preset historical period before the current moment are combined into a data sequence, and three abnormal characteristic values ​​of each data are obtained according to the local fluctuation degree and local change trend of each data in the data sequence.

[0057] Electrical equipment is the core device in the power system and plays a key role in the control, transmission, and use of electric energy. Therefore, it is necessary to analyze the monitoring data of electrical equipment and provide timely warnings of abnormalities and faults in electrical equipment to ensure the safety of equipment and personnel, maintain the stability of system operation, and improve the operation and maintenance efficiency of the power system.

[0058] Existing electrical equipment monitoring methods often set fixed anomaly thresholds for each monitoring indicator and issue alarms when data exceeds these thresholds, or employ fixed anomaly recognition models to provide early warnings of electrical equipment status. However, numerous factors can influence electrical equipment monitoring data, such as sensor status and ambient temperature. Fixed anomaly thresholds or anomaly recognition models often struggle to adapt to varying data patterns under varying circumstances, resulting in low accuracy in warnings of electrical equipment anomalies.

[0059] Therefore, in an embodiment of the present invention, based on the changing rules of the monitoring data of each monitoring indicator of the electrical equipment, the abnormal characteristic values ​​of each monitoring indicator under different circumstances are analyzed to reflect the abnormal performance of the data under different circumstances, and then the abnormal degree of the electrical equipment at the current moment is obtained based on the abnormal characteristic values, and an abnormal warning is issued for the abnormal situation of the electrical equipment to improve the accuracy of the abnormal warning.

[0060] Considering that the primary monitoring indicators for electrical equipment are current and voltage, current transformers and voltage transformers are installed at key locations such as motors and transformers to monitor the current and voltage of the electrical equipment. These current transformers and voltage transformers acquire current and voltage data every second. This is not a limitation here; implementers can set monitoring indicators and data acquisition frequencies based on specific scenarios. The acquired current and voltage data are stored in the electrical equipment's operation log. Considering that adjustments to electrical equipment can affect the regularity of acquired data, in this embodiment of the present invention, only all monitoring data from the three days preceding the current moment is analyzed as historical data. This is not a limitation here; implementers can set this based on specific scenarios.

[0061] In an embodiment of the present invention, taking the mth monitoring indicator as an example, first, the monitoring value of the mth monitoring indicator in the operation log of the electrical equipment (including the monitoring value at the current moment) is linearly normalized to obtain the real-time data of the mth monitoring indicator at the current moment, as well as the historical data of the mth monitoring indicator before the current moment. The real-time data and the historical data of the mth monitoring indicator within the three days before the current moment are combined to form a data sequence for the mth monitoring indicator. That is, the real-time data of the mth monitoring indicator is the last data in the data sequence for the mth monitoring indicator. Linear normalization is a prior art and will not be described in detail here.

[0062] Since the abnormal situation of the monitoring index of the electrical equipment is usually reflected on the data value, the data fluctuation degree and the data change trend, after obtaining the data sequence of the mth monitoring index, the fluctuation characteristic value and the change trend characteristic value of each data in the data sequence of the mth monitoring index are obtained according to the data fluctuation degree and the data change trend in the data sequence, and then the data value, the fluctuation characteristic value and the change trend value of each data are taken as three kinds of abnormal characteristic values of each data.

[0063] Since the data fluctuation degree is embodied by the data change amplitude in the time period in which each data is located, and the data change trend is embodied by the overall rising or falling degree of the data in the time period in which each data point is located, the length of the time period corresponding to each data in the data sequence of the mth monitoring index needs to be set. If the length of the time period is set too short, the data change situation cannot be effectively reflected, and if the length of the time period is set too long, the data of multiple change rules may be included, which affects the calculation of the data fluctuation degree and the change trend. Therefore, in the embodiment of the present application, the data change trend of each data in the data sequence is obtained according to the data change trend of the data within 5s before and after each data in the data sequence, that is, the data change trend of the data within 5s before and after each data in the data sequence, which is not limited here, and the implementer can set it according to the specific scene to find at least two data whose data change rules change obviously, and then obtain the length of the time period corresponding to each data according to the time interval between the data whose data change rules change obviously.

[0064] Since the number of data before the first 5 data and the number of data after the last 5 data in the data sequence are less than 5, for any data in the data sequence, if the number of data before the data is greater than or equal to 5, and the number of data after the data is greater than or equal to 5, the data is recorded as an analysis data, and a window with a length of 11 is constructed in the data sequence with the analysis data as the center, that is, the window includes the data within 5s before and after the time point corresponding to the data, and the trend change degree of each analysis data in the data sequence is obtained according to the data change trend in the window of each analysis data, which is used to find the data whose data change rule changes obviously. For example, the trend change degree of the dth analysis data is obtained.

[0065] The first difference value between the dth analysis data and the first data in the window of the dth analysis data is calculated, the second difference value between the last data in the window of the dth analysis data and the dth analysis data is calculated, and the absolute value of the difference between the first difference value and the second difference value is taken as the trend change degree of the dth analysis data.

[0066] In an embodiment, the calculation formula of the trend change degree of the dth analysis data in the data sequence of the mth monitoring index is:

[0067] ;

[0068] in, Indicates the degree of trend change of the dth data sequence to be analyzed in the data sequence of the mth monitoring indicator, Indicates the dth data to be analyzed in the data sequence of the mth monitoring indicator, Represents the first data in the window of the dth data to be analyzed in the data sequence of the mth monitoring indicator, Represents the last data in the window of the dth data to be analyzed in the data sequence of the mth monitoring indicator, Indicates the absolute value symbol.

[0069] It should be noted that Indicates the overall change trend of the data before the dth data to be analyzed within the window of the dth data to be analyzed. Indicates the overall change trend of the data after the dth data to be analyzed in the window of the dth data to be analyzed. The larger the value is, the greater the difference in the trend of the data before and after the dth data to be analyzed is, and the The larger the value is, the more likely the dth data to be analyzed is the data with a significant change in the data change pattern.

[0070] Similarly, obtain the trend change degree of each data to be analyzed in the data sequence, sort all trend change degrees in descending order to obtain a descending sequence, and record the data to be analyzed corresponding to the first 5% of the data in the descending sequence as the target data. There is no restriction here, and the implementer can set it according to the specific scenario.

[0071] Furthermore, according to the moment corresponding to each target data, all target data are sorted in ascending order to obtain a target data sequence, the time interval between each two adjacent target data in the target data sequence is calculated, and the minimum value among all time intervals is selected as the target time interval for each data in the data sequence, that is, the length of the time period corresponding to each data in the data sequence of the mth monitoring indicator.

[0072] After obtaining the time period length corresponding to each data in the data sequence of the mth monitoring indicator, the time period of each data in the data sequence is obtained according to the time period length corresponding to each data, and then the subsequence of each data is obtained, which is used to obtain the fluctuation characteristic value and trend characteristic value of each data according to the data fluctuation degree and data change trend in the subsequence of each data.

[0073] Taking the i-th data in the data sequence as an example, the time period of the i-th data is obtained according to the target time interval, and the moment corresponding to the i-th data is the last moment of the time period of the i-th data.

[0074] Furthermore, according to the time period of the i-th data, a subsequence of the i-th data is obtained. Specifically:

[0075] If the time period between the first data and the i-th data in the data sequence is greater than or equal to the time period of the i-th data, then obtain the subsequence of the i-th data in the data sequence according to the time period of the i-th data;

[0076] If the time period between the first data and the i-th data in the data sequence is less than the time period of the i-th data, then a subsequence of the i-th data is obtained in the operation log of the electrical equipment according to the time period of the i-th data.

[0077] It should be noted that, since the moment corresponding to the i-th data in the data sequence of the m-th monitoring indicator is the last moment in the time period of the i-th data, the i-th data is the last data in the subsequence of the i-th data.

[0078] Similarly, a subsequence of each data in the data sequence is obtained, and further, the fluctuation characteristic value and the change trend characteristic value of each data in the data sequence are obtained.

[0079] Regarding the fluctuation characteristic value, since the degree of data fluctuation is reflected by the data variation range within the data subsequence, in the embodiment of the present invention, taking the i-th data in the data sequence of the m-th monitoring indicator as an example, the first-order difference sequence of the subsequence of the i-th data is used to express the data variation range within the subsequence of the i-th data, thereby obtaining the fluctuation characteristic value of the i-th data. Specifically:

[0080] Obtain a first-order difference sequence of a subsequence of the i-th data, accumulate the absolute values ​​of the differences between every two adjacent data in the first-order difference sequence to obtain an accumulated value, and linearly normalize the accumulated value to obtain a fluctuation characteristic value of the i-th data.

[0081] In one embodiment, the calculation formula for the fluctuation characteristic value of the i-th data in the data sequence of the m-th monitoring indicator is:

[0082] ;

[0083] in, Represents the fluctuation characteristic value of the i-th data in the data sequence of the m-th monitoring indicator, represents the number of all data in the subsequence of the i-th data (since the time period corresponding to each data in the data sequence of the m-th monitoring indicator is the same, the number of all data in the subsequence of each data in the data sequence is the same). It represents the absolute value of the difference between the j+1th data in the subsequence of the i-th data and its previous data (i.e., the j-th data), that is, the j-th data in the first-order difference sequence of the subsequence of the i-th data. It represents the absolute value of the difference between the jth data in the subsequence of the i-th data and its previous data (i.e., the j-1th data), that is, the j-1th data in the first-order difference sequence of the subsequence of the i-th data. Indicates the absolute value symbol, represents the linear normalization function.

[0084] It should be noted that It represents the change value between the j+1th data and its previous data (i.e., the jth data) in the subsequence of the i-th data, that is, the change value between the j-th data and its subsequent data (i.e., the j+1th data) in the subsequence of the i-th data. Indicates the change between the jth data and its previous data (i.e., the j-1th data) in the subsequence of the i-th data. The larger the value, the greater the difference between the before and after change values ​​of the jth data in the subsequence of the i-th data. The larger the value is, the more drastic the data fluctuation is during the time period of the i-th data.

[0085] Regarding the change trend characteristic value, since the data change trend is reflected by the degree of overall increase or decrease of the data in the subsequence of each data point, in the embodiment of the present invention, the degree of overall increase or decrease of the data in the subsequence of the i-th data is expressed based on the difference between the first data and the last data (i.e., the i-th data) in the subsequence of the i-th data in the data sequence of the m-th monitoring indicator, thereby obtaining the change trend characteristic value of the i-th data. Specifically:

[0086] Calculate the absolute value of the difference between the i-th data (i.e., the last data in the subsequence of the i-th data) and the first data in the subsequence of the i-th data to obtain the change amplitude of the i-th data;

[0087] Calculate the difference between the i-th data (i.e., the last data in the subsequence of the i-th data) and the first data in the subsequence of the i-th data, calculate the sum of the change amplitude and a preset constant, obtain a sum result, and use the ratio of the difference to the sum result as the change trend value of the i-th data;

[0088] The product of the linear normalized value of the change amplitude and the change trend value is calculated to obtain the change trend characteristic value of the i-th data.

[0089] In one embodiment, the calculation formula for the change trend characteristic value of the i-th data is:

[0090] ;

[0091] in, Indicates the change trend characteristic value of the i-th data in the data sequence of the m-th monitoring indicator, Represents the first data in the subsequence of the i-th data in the data sequence of the m-th monitoring indicator, Represents the i-th data in the data sequence of the m-th monitoring indicator, Represents a preset constant, used to prevent the denominator from being 0. In the embodiment of the present invention, set ,There is no restriction here, and the implementer can set it according to the specific scenario. Indicates the absolute value symbol, represents the linear normalization function.

[0092] It should be noted that That is the change range of the i-th data, The larger it is, the greater the overall change in the data within the subsequence of the i-th data; That is, the change trend value of the i-th data, which is used to distinguish whether the data in the i-th data subsequence is showing an overall growth trend or an overall decline trend. When, that is When it is a positive number, the data in the subsequence of the i-th data shows an overall growth trend. is a positive value; when When, that is When it is a negative number, the data in the subsequence of the i-th data shows an overall downward trend. is a negative value.

[0093] So far, the fluctuation characteristic value and change trend characteristic value of the i-th data in the data sequence of the m-th monitoring indicator are obtained, and then the i-th data, the fluctuation characteristic value of the i-th data and the change trend value of the i-th data are taken as the three abnormal characteristic values ​​of the i-th data. For the convenience of subsequent explanation, each abnormal characteristic value of the i-th data point is recorded as , represents the vth abnormal feature value of the i-th data point.

[0094] Similarly, three abnormal characteristic values ​​of each data in the data sequence of the mth monitoring indicator are obtained, which are used to subsequently obtain the abnormal degree of the electrical equipment at the current moment based on the abnormal characteristic values, and to issue abnormal warnings for the abnormal conditions of the electrical equipment, thereby improving the accuracy of abnormal warnings.

[0095] Step S102: for any abnormal characteristic value, based on the similarity performance of any abnormal characteristic value between the real-time data and each historical data, obtain similar historical data similar to any abnormal characteristic value of the real-time data; based on the correlation performance between the real-time data and each similar historical data under other abnormal characteristic values, obtain the abnormal reaction degree of any abnormal characteristic value of the real-time data to the electrical equipment.

[0096] Taking into account that when electrical equipment encounters abnormal conditions such as overload and short circuit, multiple abnormal characteristic values ​​of monitoring data under the same monitoring indicator often undergo correlated changes, therefore, by analyzing which abnormal characteristic values ​​have undergone correlated abnormal changes, it is possible to identify which abnormal characteristic value corresponds to an abnormal type that has a higher degree of response to the abnormal state of the electrical equipment at the current moment. For abnormal characteristic values ​​with a higher degree of response, they should be given a higher weight to improve the accuracy of abnormal warnings for electrical equipment.

[0097] Furthermore, the data collected by monitoring equipment (such as current transformers and voltage transformers used to monitor electrical equipment) is subject to interference from other complex factors, such as equipment stability, ambient temperature and humidity. This can lead to a high probability that the data will deviate from the true value, thus failing to accurately reflect the actual abnormal state of the electrical equipment at the moment. Due to the interference from these complex factors, abnormal data generated by these factors tends to have poor regularity, and their abnormal feature values ​​often lack strong correlation.

[0098] Therefore, in an embodiment of the present invention, taking the vth abnormal characteristic value as an example, according to the similarity of the vth abnormal characteristic value between each historical data and the real-time data in the data sequence of the mth monitoring indicator, similar historical data similar to the vth abnormal characteristic value of the real-time data is obtained, and then the similarity between the similar historical data and the real-time data under other abnormal characteristic values ​​is analyzed. If the similarity between the similar historical data and the real-time data under other abnormal characteristic values ​​is high, it means that the real-time data of the mth monitoring indicator is less likely to be interfered with by other complex factors such as the stability of the monitoring equipment, the ambient temperature and humidity, and the correlation between the vth abnormal characteristic value and the other abnormal characteristic values ​​is high, and then the vth abnormal characteristic value has a high degree of response to the abnormal state of the electrical equipment at the current moment. Otherwise, it means that the real-time data of the mth monitoring indicator is more likely to be interfered with by other complex factors such as the stability of the monitoring equipment, the ambient temperature and humidity, or the correlation between the vth abnormal characteristic value and the other abnormal characteristic values ​​is low, and then the vth abnormal characteristic value has a low degree of response to the abnormal state of the electrical equipment at the current moment.

[0099] According to the similarity of the vth abnormal characteristic value between each historical data and the real-time data in the data sequence of the mth monitoring indicator, the steps of obtaining similar historical data similar to the vth abnormal characteristic value of the real-time data are as follows:

[0100] (1) According to the similarity of the vth abnormal characteristic value between the real-time data and each historical data in the data sequence of the mth monitoring indicator, the similarity degree between the real-time data and each historical data under the vth abnormal characteristic value is obtained respectively.

[0101] Specifically, for any historical data, the absolute value of the difference between the vth abnormal eigenvalue of any historical data and the real-time data is calculated to obtain the eigenvalue difference index, and the inverse of the sum of the eigenvalue difference index and the preset constant is linearly normalized to obtain the degree of similarity between the real-time data and the any historical data under the vth abnormal eigenvalue.

[0102] In one embodiment, taking the kth historical data in the data sequence of the mth monitoring indicator as an example, the calculation formula for the similarity between the real-time data and the kth historical data under the vth abnormal characteristic value is:

[0103] ;

[0104] in, Indicates the similarity between real-time data and the k-th historical data under the v-th abnormal characteristic value, Represents the vth abnormal characteristic value of real-time data, represents the vth abnormal characteristic value of the kth historical data, Represents a preset constant, used to prevent the denominator from being 0. In the embodiment of the present invention, set ,There is no restriction here, and the implementer can set it according to the specific scenario. Indicates the absolute value symbol, represents the linear normalization function.

[0105] It should be noted that The smaller it is, the smaller the difference in the vth abnormal characteristic value between the real-time data and the kth historical data is. The larger , the more similar the vth abnormal feature value of the kth historical data is to the vth abnormal feature value of the real-time data.

[0106] Similarly, obtain the similarity between the real-time data in the data sequence of the mth monitoring indicator and each historical data under the vth abnormal characteristic value.

[0107] (2) According to the similarity between the real-time data in the data sequence of the mth monitoring indicator and each historical data under the vth abnormal characteristic value, the similar historical data of the real-time data is obtained.

[0108] Specifically, among all similarity levels, historical data corresponding to a similarity level greater than 0.8 is used as similar historical data to the real-time data. There is no restriction here, and the implementer can set it according to the specific scenario.

[0109] At this point, similar historical data similar to the vth abnormal characteristic value of the real-time data of the mth monitoring indicator has been obtained. Further, other abnormal characteristic values ​​(i.e., abnormal characteristic values ​​other than the vth abnormal characteristic value) are represented by r. According to the method of obtaining the similarity between the real-time data and each historical data under the vth abnormal characteristic value, the similarity between the real-time data and each similar historical data under the rth other abnormal characteristic value is obtained, which is recorded as the correlation degree between the real-time data and each similar historical data under the rth other abnormal characteristic value. The higher the correlation degree between the real-time data and each similar historical data under the rth other abnormal characteristic value, the less likely the real-time data is to be interfered with by other complex factors such as the stability of the monitoring equipment, the ambient temperature and humidity, and the stronger the correlation between the vth abnormal characteristic value and the rth abnormal characteristic value of the real-time data. That is, if the real-time data has an abnormality of the abnormal type corresponding to the vth abnormal characteristic value, the abnormal type corresponding to the vth abnormal characteristic value has a higher degree of response to the abnormal state of the electrical equipment. Conversely, the abnormal type corresponding to the vth abnormal characteristic value has a lower degree of response to the abnormal state of the electrical equipment.

[0110] Furthermore, according to the correlation degree between the real-time data in the data sequence of the mth monitoring indicator and each similar historical data under each other abnormal characteristic value except the vth abnormal characteristic value, the abnormal reaction degree of the vth abnormal characteristic value of the real-time data to the electrical equipment is obtained. The specific process is as follows:

[0111] (1) Taking the fth similar historical data in the data sequence of the mth monitoring indicator as an example, the noise possibility index of the fth similar historical data is obtained according to the data change of each monitoring indicator at the corresponding moment of the fth similar historical data.

[0112] Taking into account that the monitoring equipment may be interfered by electromagnetic and other factors during data collection, there may be some noise data in the historical data. These noise data do not conform to the actual status of the electrical equipment at the corresponding monitoring moment and will affect the correlation evaluation results between different types of abnormal characteristic values. Therefore, it is necessary to reduce the impact of noise data on the abnormal response degree of the vth type of abnormal characteristic value.

[0113] Since each abnormal characteristic value of each data in the data sequence of the mth monitoring indicator is obtained through the data of the mth monitoring indicator, and electromagnetic interference will affect multiple monitoring devices at the same time, that is, when the data of one monitoring indicator at a certain moment is interfered by noise, the data of other monitoring indicators at the same moment may also be interfered by noise. Therefore, taking the fth similar historical data in the data sequence of the mth monitoring indicator as an example, according to the data of each monitoring indicator at the corresponding moment of the fth similar historical data in the data sequence of the mth monitoring indicator, the noise possibility index of the fth similar historical data is obtained. Specifically:

[0114] Because noise data usually manifests as sudden changes in data values, in the data sequence, the 10 data closest to the time corresponding to the f-th similar historical data are obtained and recorded as reference data. There is no restriction here, and the implementer can set it according to the specific scenario;

[0115] Record the time corresponding to the f-th similar historical data as the target historical time, and record the time corresponding to each reference data as the reference time;

[0116] Calculate the absolute value of the difference between the data corresponding to the target historical moment and any reference moment for each monitoring indicator, and add up all the absolute values ​​of the difference between the target historical moment and any reference moment to obtain the overall mutation degree in all monitoring indicators between the target historical moment and any reference moment;

[0117] The overall mutation degree corresponding to the target historical moment and each reference moment is accumulated to obtain the noise possibility index of the f-th similar historical data.

[0118] In one embodiment, the calculation formula for the noise likelihood index of the f-th similar historical data is:

[0119] ;

[0120] in, represents the noise possibility indicator of the f-th similar historical data, represents the number of all reference data, G represents the number of all monitoring indicators of the electrical equipment (in the embodiment of the present invention, the monitoring indicators are current and voltage, so G=2), Indicates the time (target time) corresponding to the fth similar historical data. The data of monitoring indicators, represents the fth similar historical data The time (reference time) corresponding to the reference data The data of monitoring indicators, Indicates the absolute value symbol, represents the linear normalization function.

[0121] It should be noted that is the overall mutation degree, The larger the value, the greater the mutation degree of the f-th similar historical data. The larger , the more likely the fth similar historical data is noise.

[0122] Similarly, the noise possibility index of each similar historical data in the data sequence of the mth monitoring indicator that is similar to the vth abnormal characteristic value of the real-time data is obtained.

[0123] (2) Based on the degree of correlation between the real-time data in the data sequence of the mth monitoring indicator and each similar historical data under each other abnormal characteristic value except the vth abnormal characteristic value, and the noise possibility index of each similar historical data, the abnormal reaction degree of the vth abnormal characteristic value of the real-time data to the electrical equipment is obtained.

[0124] Specifically, for the rth other abnormal feature value, the similarity between the real-time data and each similar historical data under any other abnormal feature value is obtained according to the absolute value of the difference between each similar historical data and the real-time data, that is, according to the method for obtaining the similarity between the real-time data and each historical data under the vth abnormal feature value, the similarity between the real-time data and each similar historical data under the rth other abnormal feature value is obtained, and recorded as the correlation degree between the real-time data and each similar historical data under the rth other abnormal feature value;

[0125] Calculating the difference between the constant 1 and the noise possibility index of each similar historical data respectively to obtain the weight of the correlation degree corresponding to each similar historical data;

[0126] According to the weight of the correlation degree corresponding to each of the similar historical data, the correlation degree corresponding to each of the similar historical data is weighted and summed to obtain the correlation index between the vth abnormal feature value of the real-time data and the rth other abnormal feature value;

[0127] The correlation index between the vth abnormal characteristic value of the real-time data and each other abnormal characteristic value is obtained respectively, and the maximum value among all the correlation indices is selected as the abnormal reaction degree of the vth abnormal characteristic value of the real-time data to the electrical equipment.

[0128] In one embodiment, the calculation formula for the correlation index between the vth abnormal characteristic value of the real-time data of the mth monitoring indicator and the rth other abnormal characteristic value is:

[0129] ;

[0130] in, Represents the correlation index between the vth abnormal characteristic value of the real-time data of the mth monitoring indicator and the rth other abnormal characteristic value, It represents the number of similar historical data in the data sequence of the mth monitoring indicator that are similar to the vth abnormal characteristic value of the real-time data. It represents the degree of association between the fth similar historical data with the vth abnormal characteristic value of the real-time data under the rth other abnormal characteristic value in the data sequence of the mth monitoring indicator, It represents the noise possibility index of similar historical data with similar v-th abnormal characteristic value to real-time data in the data sequence of the m-th monitoring indicator.

[0131] It should be noted that The larger the value, the more similar the vth abnormal characteristic value of the real-time data of the mth monitoring indicator is to the rth other abnormal characteristic value. The larger the value is, the greater the correlation index between the vth abnormal characteristic value of the real-time data of the mth monitoring indicator and the rth other abnormal characteristic value is, and the higher the degree of reaction of the abnormal type corresponding to the vth abnormal characteristic value to the abnormal state of the electrical equipment is; The smaller it is, the less likely it is that the similar historical data similar to the vth abnormal characteristic value of the real-time data of the mth monitoring indicator is noise. The higher the credibility, the The larger the value is, the greater the correlation index between the vth abnormal characteristic value of the real-time data of the mth monitoring indicator and the rth other abnormal characteristic value is, and the higher the degree of reaction of the abnormal type corresponding to the vth abnormal characteristic value to the abnormal state of the electrical equipment is.

[0132] Similarly, obtain the correlation index between the vth abnormal characteristic value of the real-time data of the mth monitoring indicator and each other abnormal characteristic value, and select the maximum value among all the correlation indices as the abnormal reaction degree of the vth abnormal characteristic value of the real-time data of the mth monitoring indicator to the electrical equipment, which is recorded as .

[0133] At this point, the abnormal reaction degree of the vth abnormal characteristic value of the real-time data of the mth monitoring indicator to the electrical equipment is obtained.

[0134] Step S103, based on the abnormal response degree of each abnormal characteristic value of the real-time data and the three abnormal characteristic values ​​of each data in the data sequence, obtain the abnormal degree of any monitoring indicator, obtain the abnormal degree of each monitoring indicator of the electrical equipment, and issue an abnormal warning to the electrical equipment based on the abnormal degree of all monitoring indicators.

[0135] According to step S102, the degree of abnormal reaction of each abnormal characteristic value of the real-time data of the m-th monitoring indicator to the electrical equipment is obtained. Furthermore, by combining the three abnormal characteristic values ​​of each data in the data sequence of the m-th monitoring indicator (i.e., the data value, the fluctuation characteristic value, and the change trend characteristic value of each data), and the degree of abnormal reaction of each abnormal characteristic value of the real-time data of the m-th monitoring indicator to the electrical equipment, the abnormal degree of the electrical equipment at the current moment of the m-th monitoring indicator is obtained. Specifically:

[0136] Mapping the three abnormal characteristic values ​​of each data in the data sequence to a three-dimensional coordinate system, wherein the three coordinate axes of the three-dimensional coordinate system respectively represent the three abnormal characteristic values ​​of each data in the data sequence;

[0137] In the three-dimensional coordinate system, the Euclidean distance between the real-time data and each historical data is calculated respectively, and the historical data corresponding to the 20 smallest Euclidean distances among all the Euclidean distances are selected as the target historical data. There is no restriction here, and the implementer can set the number of target historical data according to the specific scenario;

[0138] Since abnormal data is significantly different from other data, in a three-dimensional coordinate system, the greater the data density at the location of the real-time data, the smaller the degree of abnormality of the real-time data. Therefore, for any target historical data, the square of the distance between the real-time data and the any target historical data on each coordinate axis is calculated to obtain the square term of each coordinate axis. The abnormal reaction degree of the abnormal characteristic value corresponding to each coordinate axis is used as the weight of the square term of each coordinate axis. The square terms of each coordinate axis are weighted and summed to obtain a weighted summation result. The arithmetic square root of the weighted summation result is used as the weighted distance between the real-time data and the any target historical data.

[0139] The weighted distances between the real-time data and each of the target historical data are calculated respectively, and the average values ​​of all weighted distances are linearly normalized to obtain the abnormality degree of the mth monitoring indicator.

[0140] In one embodiment, the calculation formula for the abnormality degree of the mth monitoring indicator is:

[0141] ;

[0142] in, represents the abnormality degree of the mth monitoring indicator, T represents the number of target historical data, and Y represents the number of all abnormal characteristic values ​​of the mth monitoring indicator (in the embodiment of the present invention, Y=3). represents the yth abnormal characteristic value of the tth target historical data, Indicates the yth abnormal characteristic value of real-time data, Indicates the degree of abnormal response of the yth abnormal characteristic value of real-time data to the abnormality of electrical equipment. represents the linear normalization function.

[0143] It should be noted that The larger the value is, the farther the distance between the real-time data and the t-th target historical data is in the three-dimensional coordinate system, and the smaller the data density at the location of the real-time data is, that is, the greater the abnormality of the real-time data is. The larger the value is, the greater the possibility that the mth monitoring indicator of the electrical equipment will be abnormal.

[0144] Similarly, the abnormality level of each monitoring indicator of the electrical equipment is obtained. In the embodiment of the present invention, the monitoring indicators of the electrical equipment are current and voltage. Based on experimental statistics, the preset abnormality level threshold of current is set to 0.6, and the preset abnormality level threshold of voltage is set to 0.6. There is no limitation here. The implementer can set it according to the specific scenario. If the abnormality level of current is greater than or equal to 0.6, the electrical equipment will be issued a current abnormality alarm; if the abnormality level of voltage is greater than or equal to 0.6, the electrical equipment will be issued a voltage abnormality alarm.

[0145] At the same time, since a single current or voltage anomaly may indicate a need for local adjustment or inaccurate monitoring equipment data, a lower urgency level is set for the single current or voltage anomaly alarm. If both the current and voltage anomaly levels exceed their corresponding preset anomaly thresholds, indicating a more serious anomaly, a higher urgency level is required, requiring immediate investigation and repair of the anomaly location. Setting the urgency level is conventional technology and will not be detailed here.

[0146] In summary, the present invention, for any monitoring indicator of electrical equipment, in the operation log of the electrical equipment, forms a data sequence with the real-time data of any monitoring indicator at the current moment and the historical data of any monitoring indicator in a preset historical period before the current moment, and obtains three abnormal characteristic values ​​of each data according to the local fluctuation degree and local change trend of each data in the data sequence; for any abnormal characteristic value, according to the similarity performance of any abnormal characteristic value between the real-time data and each historical data, obtains similar historical data similar to any abnormal characteristic value of the real-time data, and according to the correlation performance between the real-time data and each similar historical data under other abnormal characteristic values, obtains the abnormal reaction degree of any abnormal characteristic value of the real-time data to the electrical equipment; according to the abnormal reaction degree of each abnormal characteristic value of the real-time data and the three abnormal characteristic values ​​of each data in the data sequence, obtains the abnormal degree of any monitoring indicator, obtains the abnormal degree of each monitoring indicator of the electrical equipment, and performs abnormal warning on the electrical equipment according to the abnormal degree of all monitoring indicators. Among them, since the factors affecting the data of various monitoring indicators of electrical equipment are different, by analyzing the change rules of each data under each monitoring indicator under different situations, multiple abnormal characteristic values ​​of each data are obtained to reflect the abnormal performance of the data under different situations, thereby improving the accuracy of abnormal assessment of electrical equipment; according to the similarity between real-time data and historical data under each abnormal characteristic value, the abnormal response degree of each abnormal characteristic value to the electrical equipment is obtained to more accurately reflect the real abnormal state of the electrical equipment, and then combined with the multiple abnormal characteristic values ​​of each data under each monitoring indicator and the abnormal response degree of each abnormal characteristic value, abnormal early warning of the electrical equipment is performed, thereby improving the accuracy of abnormal early warning of the electrical equipment.

[0147] Based on the same inventive concept as the above method, an embodiment of the present invention also provides a monitoring system for electrical equipment, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-mentioned methods for monitoring electrical equipment.

[0148] The above examples are only used to illustrate the technical solutions of the present application, but not limit the present application; although the present application has been described in detail with reference to the foregoing examples, those ordinarily skilled in the art should understand: the technical solutions recorded in the foregoing examples can still be modified, or some technical features can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A monitoring method for electrical equipment, characterized in that: The monitoring method for electrical equipment comprises: For any monitoring indicator of the electrical equipment, in the operation log of the electrical equipment, the real-time data of the monitoring indicator at the current moment and the historical data of the monitoring indicator in a preset historical period before the current moment are combined into a data sequence. Based on the local fluctuation degree and local change trend of each data in the data sequence, three abnormal characteristic values ​​of each data are obtained; For any abnormal characteristic value, based on the similarity between the real-time data and each historical data, similar historical data similar to the abnormal characteristic value of the real-time data is obtained; based on the correlation between the real-time data and each similar historical data under other abnormal characteristic values, the degree of abnormal reaction of the abnormal characteristic value of the real-time data to the electrical equipment is obtained; According to the abnormal response degree of each abnormal characteristic value of the real-time data and the three abnormal characteristic values ​​of each data in the data sequence, the abnormal degree of any monitoring indicator is obtained, the abnormal degree of each monitoring indicator of the electrical equipment is obtained, and according to the abnormal degree of all monitoring indicators, an abnormal warning is issued for the electrical equipment; According to the local fluctuation degree and local change trend of each data in the data sequence, three abnormal characteristic values ​​of each data are obtained, including: According to the local change trend of each data in the data sequence, the target time interval of each data in the data sequence is obtained; For any data in the data sequence, the time period of the data is obtained according to the target time interval, and the moment corresponding to the data is the last moment of the time period of the data; If the time period between the first data in the data sequence and the any data is greater than or equal to the time period of the any data, then obtaining a subsequence of the any data in the data sequence according to the time period of the any data; If the time period between the first data in the data sequence and the any data is shorter than the time period of the any data, obtaining a subsequence of the any data in the operation log of the electrical equipment according to the time period of the any data; Obtaining a first-order difference sequence of a subsequence of the any data, accumulating the absolute values ​​of the differences between each two adjacent data in the first-order difference sequence to obtain an accumulated value, and performing linear normalization on the accumulated value to obtain a fluctuation characteristic value of the any data; Obtaining a change trend characteristic value of the any data according to a data change trend in a subsequence of the any data; The any data, the fluctuation characteristic value of the any data, and the change trend characteristic value of the any data are used as three abnormal characteristic values ​​of the any data; The obtaining, based on the similarity performance of any one abnormal feature value between the real-time data and each historical data, similar historical data that is similar to any one abnormal feature value of the real-time data includes: For any historical data, calculate the absolute value of the difference between any one of the abnormal characteristic values ​​of the historical data and the real-time data to obtain a characteristic value difference index, perform linear normalization on the inverse of the sum of the characteristic value difference index and a preset constant, and obtain the degree of similarity between the real-time data and the any one of the historical data under any one of the abnormal characteristic values; Obtaining a similarity between the real-time data and each of the historical data under any of the abnormal feature values, and taking, among all similarities, historical data corresponding to a similarity greater than a preset similarity threshold as similar historical data to the real-time data; The step of obtaining the degree of abnormal response of any abnormal characteristic value of the real-time data to the electrical equipment based on the correlation between the real-time data and each of the similar historical data under other abnormal characteristic values ​​includes: For any similar historical data, in the data sequence, a third preset number of data closest to the time corresponding to the any similar historical data is obtained and recorded as reference data; Recording the time corresponding to any similar historical data as the target historical time, and recording the time corresponding to each reference data as the reference time; Calculate the absolute value of the difference between the data corresponding to the target historical moment and any reference moment for each monitoring indicator, and add up all the absolute values ​​of the difference between the target historical moment and any reference moment to obtain the overall mutation degree in all monitoring indicators between the target historical moment and any reference moment; Accumulating the overall mutation degree corresponding to the target historical moment and each reference moment to obtain a noise possibility index for any similar historical data; According to the correlation performance between the real-time data and each of the similar historical data under other abnormal characteristics, and the noise possibility index of each of the similar historical data, the abnormal reaction degree of any abnormal characteristic value of the real-time data to the electrical equipment is obtained.

2. A monitoring method for electrical equipment according to claim 1, characterized in that: The step of obtaining a target time interval for each data in the data sequence according to a local change trend of each data in the data sequence includes: For any data in a data sequence, if the number of data before the data in the data sequence is greater than or equal to a first preset number, and the number of data after the data is greater than or equal to the first preset number, the data is recorded as data to be analyzed, and a window of a preset length is constructed in the data sequence with the data to be analyzed as the center, where the first preset number is half of the difference between the preset length and a constant of 1; Calculating a first difference between the data to be analyzed and the first data in the window, calculating a second difference between the last data in the window and the data to be analyzed, and using the absolute value of the difference between the first difference and the second difference as the degree of trend change of the data to be analyzed; Obtain the degree of trend change of each of the to-be-analyzed data in the data sequence, sort all trend change degrees in descending order to obtain a descending sequence, record the to-be-analyzed data corresponding to the first second preset number of trend change degrees in the descending sequence as target data, sort all target data in ascending order according to the moment corresponding to each of the target data to obtain a target data sequence, calculate the time interval between each two adjacent target data in the target data sequence, select the minimum value among all time intervals and record it as the target time interval for each data in the data sequence.

3. The method for monitoring electrical equipment according to claim 1, wherein: Obtaining a change trend characteristic value of any data according to a data change trend in a subsequence of any data includes: Calculating the absolute value of the difference between the any data and the first data in the subsequence of the any data to obtain the change amplitude of the any data; Calculating a difference between the any data and the first data in a subsequence of the any data, calculating a sum of the change amplitude and a preset constant to obtain a sum result, and using a ratio between the difference and the sum result as a change trend value of the any data; The product of the linear normalized value of the variation amplitude and the variation trend value is calculated to obtain the variation trend characteristic value of any data.

4. A monitoring method for electrical equipment according to claim 1, characterized in that: The step of obtaining the degree of abnormal response of any abnormal characteristic value of the real-time data to the electrical equipment based on the correlation between the real-time data and each of the similar historical data under other abnormal characteristics, and the noise possibility index of each of the similar historical data, includes: For any other abnormal feature value, the degree of similarity between the real-time data and each similar historical data under the any other abnormal feature value is obtained based on the absolute value of the difference between each similar historical data and the real-time data, and recorded as the degree of association between the real-time data and each similar historical data under the any other abnormal feature value; Calculating the difference between the constant 1 and the noise possibility index of each similar historical data respectively to obtain the weight of the correlation degree corresponding to each similar historical data; According to the weight of the correlation degree corresponding to each of the similar historical data, a weighted sum is performed on the correlation degree corresponding to each of the similar historical data to obtain a correlation index between the any one abnormal feature value of the real-time data and any one of the other abnormal feature values; The correlation index between any one abnormal characteristic value of the real-time data and each other abnormal characteristic value is obtained respectively, and the maximum value among all the correlation indices is selected as the abnormal reaction degree of any one abnormal characteristic value of the real-time data to the electrical equipment.

5. The method for monitoring electrical equipment according to claim 1, wherein: The abnormality degree of any monitoring indicator is obtained based on the abnormal response degree of each abnormal characteristic value of the real-time data and the three abnormal characteristic values ​​of each data in the data sequence, including: Mapping the three abnormal characteristic values ​​of each data in the data sequence to a three-dimensional coordinate system, wherein the three coordinate axes of the three-dimensional coordinate system respectively represent the three abnormal characteristic values ​​of each data in the data sequence; In the three-dimensional coordinate system, respectively calculating the Euclidean distance between the real-time data and each historical data, and selecting the historical data corresponding to the fourth preset number of smallest Euclidean distances among all the Euclidean distances as the target historical data; For any target historical data, respectively calculate the square of the distance between the real-time data and the any target historical data on each coordinate axis to obtain the square term of each coordinate axis, use the abnormal reaction degree of the abnormal characteristic value corresponding to each coordinate axis as the weight of the square term of each coordinate axis, perform weighted summation on the square terms of each coordinate axis to obtain a weighted summation result, and use the arithmetic square root of the weighted summation result as the weighted distance between the real-time data and the any target historical data; The weighted distances between the real-time data and each of the target historical data are calculated respectively, and the average values ​​of all the weighted distances are linearly normalized to obtain the abnormality degree of any monitoring indicator.

6. The method for monitoring electrical equipment according to claim 1, characterized in that: The abnormality warning of electrical equipment is issued based on the abnormality degree of all monitoring indicators, including: For the abnormality level of any monitoring indicator, if the abnormality level of any monitoring indicator is greater than or equal to the preset abnormality level threshold of any monitoring indicator, an abnormality warning is issued for any monitoring indicator.

7. A monitoring system for electrical equipment, 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 method for monitoring electrical equipment according to any one of claims 1 to 6 are implemented.

Images