Method, device and equipment for identifying transient characteristics of access points

Abstract

A method, device and equipment for identifying transient characteristics of access points include obtaining the recording time of each transient event recorded by each power quality monitoring device; fusing each transient event and transient events of the same transient event type based on the recording time; establishing a time-space frequency matrix corresponding to the fused transient events; analyzing the time-space frequency matrix to obtain monitoring points with strong correlation; obtaining an event feature set and event frequency of the transient events corresponding to each monitoring point with strong correlation; and judging whether the monitoring point with strong correlation can be constructed as a sensitive user access point based on the event feature set and event frequency, thereby realizing the screening of sensitive user access points.

Description

Technical Field

[0001] The present invention relates to the technical field of data monitoring, and in particular to a method, device and equipment for identifying transient characteristics of an access point. Background Art

[0002] With the rapid development of high-end manufacturing, power quality disturbances will cause even more severe economic losses to sensitive users. To improve the safe, reliable, and economical operation of the power grid, feasibility studies and designs must be conducted when constructing new sensitive users, taking into account user development, the current state of power grid operations, and planning. In particular, the access points for sensitive users must meet the requirements for safe power system operation and power supply reliability, and should not be connected to the same common connection point as interference sources that could impact power quality.

[0003] The increasing use of power electronics and integration of electrical equipment presents complex challenges, such as voltage sag propagation, that involve multiple uncertainties. Accurately modeling the generation, propagation, and impact of disturbances is becoming increasingly challenging. Traditional transient characteristic analysis relies on precise models, making node transient characteristic analysis challenging. This makes it difficult to uncover the complex temporal and spatial patterns of various events, hindering the screening of sensitive user access points. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, apparatus, and device for identifying transient characteristics of access points, so as to implement screening of sensitive user access points.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] A method for identifying transient characteristics of an access point, comprising:

[0007] Obtaining the recording time of each transient event recorded by each power quality monitoring device;

[0008] Filtering the transient events based on the recording time, and classifying the transient events whose recording time differences are within a preset threshold into an event record group of the same time period;

[0009] Obtain the monitoring point and transient event type of each transient event record in the same event record group;

[0010] Merge monitoring points and transient events of the same transient event type in the same event record group;

[0011] Establishing a spatiotemporal frequency matrix corresponding to the fused transient events, wherein the spatiotemporal frequency matrix records the number of transient events detected at different monitoring points in different time periods;

[0012] Calculate the correlation of each monitoring point in the spatiotemporal frequency matrix;

[0013] Determining whether the correlation is greater than a preset value;

[0014] When it is greater than the preset value, the corresponding two monitoring points are marked as having a strong correlation;

[0015] Obtain the event feature set and event frequency of transient events corresponding to each monitoring point with strong correlation;

[0016] Based on the event feature set and the event frequency, it is determined whether monitoring points with strong correlation can be constructed as sensitive user access points.

[0017] Optionally, in the above-mentioned access point transient feature identification method, the calculating the correlation of each monitoring point in the spatiotemporal frequency matrix includes:

[0018] Based on the formula Calculate the correlation ρ(X(i),X(j)) of each element in the spatiotemporal frequency matrix;

[0019] Where: X(i) and X(j) represent the frequency matrices of the i-th and j-th spatial nodes in each time period, respectively, and their elements are composed of X(i) = [x1(i), x2(i), …, xn(i)], X(j) = [x1(j), x2(j), …, xn(j)], where 1≤i, j≤m, and m is the number of elements in the spatiotemporal frequency matrix, and each element corresponds to a spatial node.

[0020] Optionally, in the access point transient feature identification method, judging whether a monitoring point with a strong correlation can be constructed as a sensitive user access point based on the event feature set and event frequency includes:

[0021] Determining that a preset feature value in the event feature set is greater than a preset value;

[0022] Determining whether the event frequency is greater than a preset frequency;

[0023] When the preset feature value in the event feature set is greater than a preset value and / or the event frequency is greater than a preset frequency, it indicates that the monitoring point cannot be constructed as a sensitive user access point.

[0024] Optionally, the above-mentioned access point transient feature identification method includes:

[0025] The preset characteristic value includes a residual voltage amplitude;

[0026] The preset frequency is the average frequency of the fused transient features corresponding to the monitoring points.

[0027] Optionally, in the above-mentioned access point transient feature identification method, the preset threshold is:

[0028] n1 is the total number of transient events before fusion, and n2 is the total number of transient events after fusion;

[0029] Establishing a relationship curve between the same time period screening threshold time_max and the λ;

[0030] The time period screening threshold time_max corresponding to the position with the maximum curvature in the relationship curve is used as the preset threshold.

[0031] An access point transient feature recognition device, comprising:

[0032] a screening unit, configured to obtain the recording time of each transient event recorded by each power quality monitoring device; screen the transient events based on the recording time, and classify the transient events whose recording time difference is within a preset threshold into an event recording group of the same time period;

[0033] A fusion unit is used to obtain the monitoring points and transient event types of each transient event record in the same event record group; and fuse transient events with the same monitoring points and transient event types in the same event record group;

[0034] A correlation analysis unit is used to establish a spatiotemporal frequency matrix corresponding to the fused transient events, wherein the spatiotemporal frequency matrix records the number of transient events detected at different monitoring points in different time periods; and calculate the correlation of each monitoring point in the spatiotemporal frequency matrix;

[0035] A strong correlation marking unit is used to determine whether the correlation is greater than a preset value; when it is greater than the preset value, the corresponding two monitoring points are marked as having a strong correlation relationship;

[0036] The access point analysis unit is used to obtain the event feature set and event frequency of transient events corresponding to each monitoring point with a strong correlation; based on the event feature set and event frequency, it is determined whether the monitoring point with a strong correlation can be constructed as a sensitive user access point.

[0037] Optionally, in the above-mentioned access point transient feature recognition device, the correlation analysis unit, when calculating the correlation of each monitoring point in the spatiotemporal frequency matrix, is specifically configured to:

[0038] Based on the formula Calculate the correlation ρ(X(i),X(j)) of each element in the spatiotemporal frequency matrix;

[0039] Where: X(i) and X(j) represent the frequency matrices of the i-th and j-th spatial nodes in each time period, respectively, and their elements are composed of X(i) = [x1(i), x2(i), …, xn(i)], X(j) = [x1(j), x2(j), …, xn(j)], where 1≤i, j≤m, and m is the number of elements in the spatiotemporal frequency matrix, and each element corresponds to a spatial node.

[0040] Optionally, in the above-mentioned access point transient feature recognition device, when the access point analysis unit determines whether a monitoring point with a strong correlation relationship can be constructed as a sensitive user access point based on the event feature set and the event frequency, it is specifically configured to:

[0041] Determining that a preset feature value in the event feature set is greater than a preset value;

[0042] Determining whether the event frequency is greater than a preset frequency;

[0043] When the preset feature value in the event feature set is greater than a preset value and / or the event frequency is greater than a preset frequency, it indicates that the monitoring point cannot be constructed as a sensitive user access point.

[0044] Optionally, in the above-mentioned access point transient feature recognition device, the preset threshold is:

[0045] n1 is the total number of transient events before fusion, and n2 is the total number of transient events after fusion;

[0046] Establishing a relationship curve between the same time period screening threshold time_max and the λ;

[0047] The time period screening threshold time_max corresponding to the position with the maximum curvature in the relationship curve is used as the preset threshold.

[0048] An access point transient feature recognition device, comprising: a memory and a processor;

[0049] The memory is used to store programs;

[0050] The processor is configured to execute the program to implement each step of any one of the above methods for identifying transient features of access points.

[0051] Based on the above technical solution, the above solution provided by the embodiment of the present invention obtains the recording time of each transient event recorded by each power quality monitoring device; fuses each transient event and transient events of the same transient event type based on the recording time; establishes a time-space frequency matrix corresponding to the fused transient events; analyzes the time-space frequency matrix to obtain monitoring points with strong correlation; obtains the event feature set and event frequency of the transient events corresponding to each monitoring point with strong correlation; and determines whether the monitoring point with strong correlation can be constructed as a sensitive user access point based on the event feature set and event frequency, thereby realizing the screening of sensitive user access points. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0053] Figure 1 A flowchart of a method for identifying transient characteristics of an access point disclosed in an embodiment of the present application;

[0054] Figure 2 Select a line chart for the optimal preset threshold value for transient event cleaning;

[0055] Figure 3 This is a schematic diagram of the strong correlation threshold value for transient event frequency;

[0056] Figure 4 A schematic diagram of the structure of the access point transient feature recognition device disclosed in an embodiment of the present application;

[0057] Figure 5 This is a schematic diagram of the structure of the access point transient feature recognition device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0058] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0059] To address the above issues, the present invention proposes a method for identifying transient features of access points based on event-time-space correlation analysis. When the number of data sets is large, the data types are diverse, and the time-space relationships are complex, the method can accurately and quickly extract association rules and disturbance patterns from different levels such as nodes and regions, so as to take necessary governance measures in advance or select more appropriate sensitive user access points.

[0060] See also Figure 1 , the present application discloses a method for identifying transient characteristics of an access point, which may include:

[0061] Step S101: obtaining the recording time of each transient event recorded by each power quality monitoring device.

[0062] In this solution, multiple different power quality monitoring devices are installed in the same substation. Different power quality monitoring devices are used to record each transient event detected at the point. When a transient event is detected, the location where the transient event is detected and the address identifier of the power quality monitoring device are recorded. The address identifier may refer to the geographical coordinate location of the power quality monitoring device.

[0063] Step S102: screening the transient events based on the recording time, and classifying the transient events whose recording time difference is within a preset threshold into an event recording group of the same time period.

[0064] In this solution, transient events occurring in the power grid will be recorded by multiple power quality monitoring devices in a very short time. In order to preliminarily distinguish different transient events, data screening can be performed based on the "trigger time" in the report. The "trigger time" is the time when the transient event is recorded. The recording time feature T is used as the only attribute to filter the event records X at the same time, and the preset threshold is set to time_max. The original multi-dimensional event data (recorded transient events) can be converted into data with time as the variable, recorded as X i ,T i , which can be understood as X i The recording time corresponding to the transient event is T i , the event filtering conditions are as follows:

[0065] if|T i -T j |≤time_max:X i ,X j ∈G x

[0066] else:X i ∈G x1 ,X j ∈G x2

[0067] Where: T i and T j Represents two transient events X i and X j The corresponding recording time; G x1 and G x2 Represents a group of event records in two different time periods, that is, if T i and T j If the absolute value of the time difference is less than or equal to time_max, it indicates that the two transient events can be attributed to the same event record group. i and T j If the absolute value of the time difference is greater than time_max, it indicates that the two transient events belong to different event record groups.

[0068] Step S103: Acquire the monitoring point and transient event type of each transient event record in the same event record group.

[0069] After preliminary processing in step S102, the original records of transient events can be grouped according to the detected time period to obtain several event record groups for the same period. Each event record group has multiple transient events. By extracting data from transient events, the monitoring points corresponding to each transient event and the transient event type can be obtained.

[0070] Step S104: merging monitoring points and transient events of the same transient event type in the same event record group.

[0071] At the same time, transient events caused by the same disturbance source will be repeatedly recorded by power quality monitoring devices in the same substation and in different substations. Therefore, it is necessary to merge the repeated records of these power quality monitoring devices into the same event, that is, merge the disturbance source characteristics of the same event record group. The merging conditions are as follows:

[0072] Condition 1: The spatial information corresponding to the transient event, i.e. the monitoring points, are consistent;

[0073] Condition 2: The transient event types corresponding to the transient events are consistent.

[0074] When merging, the monitoring points corresponding to each transient event in the event record group and the time type of the transient event are extracted. If two or more transient events in the event record group meet the above conditions one and two at the same time, these transient events are considered to be caused by the same disturbance source, and then these transient events are merged. The monitoring point where the transient event is recorded earliest is used as the spatial reference, and the event type recorded earliest is used as the event type reference.

[0075] Assuming that the number of original transient event records is n1, after grouping in the same period and fusing repeated records with the same disturbance source, the number of transient event records after cleaning is n2. The event repetition rate λ is calculated based on the deviation value of the number of event records before and after cleaning. The calculation formula is as follows:

[0076]

[0077] During transient event cleaning, the event repetition rate λ depends on the value of the same-segment filtering threshold, time_max. A larger time_max means that two records with relatively long time intervals are combined into the same event record group. This results in more records within the resulting group, leading to more events being merged into the same group. The smaller the number of cleaned event records, n2, the higher the calculated repetition rate. Therefore, time_max and λ are positively correlated.

[0078] Taking time_max as the independent variable and λ as the dependent variable, we can get Figure 2 The line graph shown is denoted as f(time_max)=λ. Taking into account the size and degree of change of the repetition rate value λ, the threshold value f(time_max) corresponding to the maximum curvature of the line in the curve formed by f(time_max)=λ is used as the optimal screening threshold, thereby realizing the optimal threshold selection for transient report data cleaning.

[0079] That is, the selection rule of the preset threshold is to establish a relationship curve between the time period screening threshold time_max and λ; and use the time period screening threshold time_max corresponding to the position with the maximum curvature in the relationship curve as the preset threshold.

[0080] Step S105: establishing a spatiotemporal frequency matrix corresponding to the fused transient events, wherein the spatiotemporal frequency matrix records the number of transient events detected in different controls in different time periods.

[0081] In this step, after the transient events are cleaned, a spatiotemporal frequency matrix of transient events is further established. The spatiotemporal frequency matrix is ​​used to analyze the correlation between the time characteristics and spatial characteristics of the disturbance events. Each element in the spatiotemporal frequency matrix is ​​used to represent the name or coordinate identification of the monitoring point and the frequency of the cleaned transient events detected in the corresponding period of time. For a certain city-level power grid area transient report, the transient events of the city can be spatially divided into m spaces based on the name of the monitoring point, that is, the number of monitoring points, and the number of transient events detected in each space (monitoring point) and each month in a year is counted based on the month as the basic time statistical unit. With m spatial dimensions as the row matrix and n time dimensions as the column matrix, a spatiotemporal frequency matrix X of the following form is formed, and the size of the matrix is ​​m×n.

[0082]

[0083] Where: X n represents the transient event frequency submatrix corresponding to each monitoring point in the nth time period; x i (j) represents the frequency element counted at the jth monitoring point in the i-th time period. It can be seen that the transient event spatiotemporal frequency matrix X is composed of the transient event frequency elements x in several spaces j (j = 1, 2, ..., m) at different times i (i = 1, 2, ..., n) i (j). The row vector X(j) of the matrix X represents the time frequency vector of the specific monitoring point j, and the column vector X i are the spatial frequency vectors at a specific time i.

[0084] Step S106: Calculate the correlation between the frequencies of each transient event in the spatiotemporal frequency matrix.

[0085] In this step, the frequency and correlation of long-term transient events in each space are comprehensively considered to perform spatiotemporal partitioning of transient events. Areas with the same disturbance frequency patterns are identified to facilitate the screening of sensitive user access areas and also contribute to the hierarchical and regional management of transient disturbances.

[0086] When calculating the correlation of transient event frequencies corresponding to each element of the spatiotemporal frequency matrix, the row vector X(·) of the transient event spatiotemporal frequency matrix X can be used as a sample and the Pearson correlation coefficient can be used to calculate the long-term transient event frequency correlation of each space, which is recorded as ρ(X(·), X(·)), or simply ρ. The calculation expression is as follows:

[0087]

[0088] Where: X(i) and X(j) represent the frequency matrices of the i-th and j-th monitoring points in each time period, respectively, and their elements are X(i) = [x1(i), x2(i), …, xn(i)], X(j) = [x1(j), x2(j), …, xn(j)], where 1 ≤ i, j ≤ m.

[0089] Step S107: determining whether the correlation is greater than a preset value;

[0090] Typically, if the calculated correlation ρ(X(i), X(j)) is greater than 0.6 (the preset value), the two frequency matrices are considered strongly correlated. If ρ(X(i), X(j)) falls within the range [0.8, 1] (a preset range), the correlation is extremely strong. In transient event analysis, the focus is on areas associated with a high incidence of transient events. Using a preset value of 0.6 may result in an excessive number of strongly correlated areas, making them less representative.

[0091] Therefore, the strong correlation determination threshold θ(%) (preset value) can be improved according to the actual calculation result of ρ(X(i), X(j)) and the frequency size to meet the screening requirements of high correlation and high frequency areas.

[0092] Assuming that n1 original records are cleaned in a year to obtain a total of n2 event records, the monthly average of the total number of event records for the whole year is used as the dividing line between low frequency and medium and high frequency, that is in, Indicates the rounding down operation, that is, taking The integer part of . If the frequency is high, then the area is the medium-high frequency area; otherwise, it is the low-frequency area.

[0093] Take the maximum value of the correlation coefficient ρ between the frequency matrix of each medium and high frequency region and other regions max，i , marked in sequence Figure 3 In the figure, all ρmax values ​​are distributed in The original strong correlation threshold θ=60 divides the region into two upper and lower regions: I and (II+III). In the upper region (II+III), if there is ρ max，i If it is not less than the original threshold value 60, then the ρ max，i As the new threshold θ′ candidate value, and sort this set of values ​​from small to large, take the minimum value in the θ′ candidate value set as the new threshold, that is, Figure 3 The high correlation threshold θ in the region is increased to θ′ (θ′>60), where θ′ is the preset value.

[0094] Step S108: When the correlation is greater than a preset value, the two are marked as having a strong correlation.

[0095] After calculating the correlation of the frequencies of each transient event, the correlation is compared with a preset value. When the correlation is greater than the preset value, the two are marked as a strong correlation. Here, the two refer to two monitoring points.

[0096] Step S109: obtaining event feature sets and event frequencies of transient events corresponding to each monitoring point having a strong correlation;

[0097] The event characteristics may include electrical characteristic information and non-electrical characteristic information of transient events detected by the monitoring point. Taking the electrical characteristic information and non-electrical characteristic information into comprehensive consideration, the non-electrical characteristics may include weather parameters, etc. A transient event association rule feature set is constructed based on the transient event report and external weather data. The rule feature set may include electrical characteristic information and non-electrical characteristic information. Each feature type information is converted into a data structure that is easy for a computer to process, and the symbolic identification is used as the input of a subsequent association rule algorithm.

[0098] Because the vast majority of fields in transient event reports are textual, categorical attributes such as substation name and voltage level, monitoring point name and voltage level, and day of the week are discretized using a "letter + category order" format, such as "Saturday" being labeled "F5." Quantitative attributes such as weather and node transient event characteristics require the following classification and discretization.

[0099] a) Discretization of weather characteristics

[0100] According to GB / T 35663-2017 "Basic Terminology of Weather Forecasting" and IEEE 346 standard, weather is divided into the following three categories:

[0101] ① Normal weather. Weather keywords include sunny, overcast, cloud, light rain, and showers;

[0102] ② Bad weather. Weather keywords include moderate rain and heavy rain;

[0103] ③ Extreme weather. Weather keywords include storm, thunder, typhoon, etc.

[0104] When weather information is discretized, the corresponding keywords are identified and assigned numbers G0, G1, and G2 to identify the weather type.

[0105] b) Discretization of node transient event characteristics

[0106] From the perspective of transient event types, voltage sag, voltage swell, and short-term interruption events can be distinguished. Combined with the residual voltage amplitude value interval characteristics of each event, transient event characteristics can be divided into the following three categories and 18 subcategories:

[0107] Table 1 Discretization identifiers of transient events

[0108]

[0109] In summary, taking into account non-electrical factors and electrical factors, the association rule pattern feature set formed is shown in Figure 02 below, thereby obtaining the electrical features and non-electrical features corresponding to each transient time.

[0110] Table 2 Association rule pattern feature set

[0111]

[0112] 2) Regional disturbance pattern extraction based on FP-growth algorithm

[0113] After determining the electrical and non-electrical characteristics of each transient event, frequent item sets of partitioned association rules based on the FP-growth algorithm in different dimensions are formed. Candidate association rules are generated through the frequent item sets, and then the association rules are filtered through a preset support threshold to finally obtain strong association rules, realizing the description of disturbance patterns and disturbance source characteristics.

[0114] First, the event characteristics of transient events corresponding to all monitoring nodes in the selected area are analyzed. The event characteristics include electrical characteristics and non-electrical characteristics. The electrical characteristics and non-electrical characteristics of transient events are comprehensively analyzed. Based on the electrical characteristics and non-electrical characteristics, the FP-growth algorithm is used to analyze and evaluate the electrical characteristics and non-electrical characteristics of the most serious transient events. These electrical characteristics and non-electrical characteristics are used as the event feature set of transient time.

[0115] Next, based on the divided transient event spatial regions, we focus on the event feature set selected based on the transient time corresponding to monitoring points with strong correlations with transient events, and obtain the event frequency corresponding to these strongly correlated monitoring points. The event frequency refers to the number of transient events detected at these monitoring points. Of course, the event feature set can also include only some preset electrical and non-electrical features, and does not need to be analyzed using the FP-growth algorithm.

[0116] Step S110: judging whether the positions corresponding to the elements with strong correlation can be used to construct user access points based on the event feature set and event frequency.

[0117] After obtaining the event characteristics and event frequencies of the monitoring points with strong correlation, the event characteristics and event frequencies are analyzed using preset analysis rules, and based on the analysis results, it is determined whether these monitoring points with strong correlation are suitable for constructing user access points.

[0118] Specifically, in this solution, let the event feature set of a transient event at a certain access point be E, and the total frequency of transient events corresponding to the monitoring point be x. If E and x of the access point simultaneously satisfy the following rules 1 and 2, then the monitoring point is determined to be unsuitable for sensitive user access.

[0119] Rule 1: Indicates that this point is a high-frequency transient event area;

[0120] Rule 2: There is a specific transient event feature in E. For example, |residual voltage amplitude interval -100%| ≥ 50%, indicating that the residual voltage amplitude deviation of the event at this point exceeds 50%.

[0121] When the event feature set E of a monitoring point satisfies Rule 2 and / or the total frequency of transient events x satisfies Rule 1, it indicates that the monitoring point is not suitable for sensitive users to access. Otherwise, it indicates that sensitive users can access the monitoring point.

[0122] As can be seen from the above scheme disclosed in the above embodiments, the present application proposes a method for identifying transient features of access points based on the spatiotemporal association of events, which can reasonably evaluate the adverse effects of each node as an access point for sensitive users; and the optimal preset threshold can be determined based on the rate of change of the number of records before and after fusion, and can mine multidimensional event samples with repeated records in time and space; it can comprehensively consider the frequency and correlation of long-term transient events in each space based on the spatiotemporal frequency matrix, and propose a strong correlation threshold selection method for medium and high frequency areas, so as to realize the spatiotemporal partitioning of transient events with the same disturbance frequency law, and can form a symbolic feature set of transient events detected by existing power quality monitoring devices and non-electrical information of the external environment, so as to accurately and quickly extract association rules and disturbance patterns.

[0123] The advantage of this analysis method is that it can comprehensively consider electrical and non-electrical factors when the number of voltage transient event samples recorded at the monitoring point is large, the data types are diverse, and the temporal and spatial relationships are complex, and the screening threshold is selected according to the actual data distribution to realize the disturbance characteristic analysis of each node, providing technical support for the design of existing access point user management solutions and the selection of sensitive user access point solutions.

[0124] This embodiment discloses an access point transient feature recognition device. For the specific working content of each unit in the device, please refer to the content of the above method embodiment.

[0125] The following describes an access point transient feature recognition device provided by an embodiment of the present invention. The access point transient feature recognition device described below and the access point transient feature recognition method described above can refer to each other.

[0126] See also Figure 4 , the access point transient feature recognition device disclosed in this application may include:

[0127] A screening unit A is configured to obtain the recording time of each transient event recorded by each power quality monitoring device; screen the transient events based on the recording time, and classify the transient events whose recording time difference is within a preset threshold into an event record group of the same time period;

[0128] Fusion unit B is used to obtain the monitoring points and transient event types of each transient event record in the same event record group; and fuse transient events with the same monitoring points and transient event types in the same event record group;

[0129] The correlation analysis unit C is used to establish a spatiotemporal frequency matrix corresponding to the fused transient events, wherein the spatiotemporal frequency matrix records the number of transient events detected at different monitoring points in different time periods; and calculate the correlation of each monitoring point in the spatiotemporal frequency matrix;

[0130] A strong correlation marking unit D is used to determine whether the correlation is greater than a preset value; when it is greater than the preset value, the corresponding two monitoring points are marked as having a strong correlation relationship;

[0131] The access point analysis unit E is used to obtain the event feature set and event frequency of transient events corresponding to each monitoring point with a strong correlation; based on the event feature set and event frequency, determine whether the monitoring point with a strong correlation can be constructed as a sensitive user access point.

[0132] Corresponding to the above method, when calculating the correlation of each monitoring point in the spatiotemporal frequency matrix, the correlation analysis unit is specifically used to:

[0133] Based on the formula Calculate the correlation ρ(X(i),X(j)) of each element in the spatiotemporal frequency matrix;

[0134] Where: X(i) and X(j) represent the frequency matrices of the i-th and j-th spatial nodes in each time period, respectively, and their elements are composed of X(i) = [x1(i), x2(i), …, xn(i)], X(j) = [x1(j), x2(j), …, xn(j)], where 1≤i, j≤m, and m is the number of elements in the spatiotemporal frequency matrix, and each element corresponds to a spatial node.

[0135] Corresponding to the above method, when the access point analysis unit determines whether a monitoring point with a strong correlation relationship can be constructed as a sensitive user access point based on the event feature set and the event frequency, it is specifically configured to:

[0136] Determining that a preset feature value in the event feature set is greater than a preset value;

[0137] Determining whether the event frequency is greater than a preset frequency;

[0138] When the preset feature value in the event feature set is greater than a preset value and / or the event frequency is greater than a preset frequency, it indicates that the monitoring point cannot be constructed as a sensitive user access point.

[0139] Figure 5For the hardware structure diagram of the server provided in the embodiment of the present invention, see Figure 5 As shown, it may include: at least one processor 100, at least one communication interface 200, at least one memory 300 and at least one communication bus 400;

[0140] In the embodiment of the present invention, the number of the processor 100, the communication interface 200, the memory 300, and the communication bus 400 is at least one, and the processor 100, the communication interface 200, and the memory 300 communicate with each other through the communication bus 400; obviously, Figure 5 The communication connections shown for the processor 100, communication interface 200, memory 300, and communication bus 400 are merely optional;

[0141] Optionally, the communication interface 200 may be an interface of a communication module, such as an interface of a GSM module;

[0142] The processor 100 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0143] The memory 300 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0144] The processor 100 is specifically configured to:

[0145] Obtaining the recording time of each transient event recorded by each power quality monitoring device;

[0146] Filtering the transient events based on the recording time, and classifying the transient events whose recording time differences are within a preset threshold into an event record group of the same time period;

[0147] Obtain the monitoring point and transient event type of each transient event record in the same event record group;

[0148] Merge monitoring points and transient events of the same transient event type in the same event record group;

[0149] Establishing a spatiotemporal frequency matrix corresponding to the fused transient events, wherein the spatiotemporal frequency matrix records the number of transient events detected at different monitoring points in different time periods;

[0150] Calculate the correlation of each monitoring point in the spatiotemporal frequency matrix;

[0151] Determining whether the correlation is greater than a preset value;

[0152] When it is greater than the preset value, the corresponding two monitoring points are marked as having a strong correlation;

[0153] Obtain the event feature set and event frequency of transient events corresponding to each monitoring point with strong correlation;

[0154] Based on the event feature set and the event frequency, it is determined whether monitoring points with strong correlation can be constructed as sensitive user access points.

[0155] For the convenience of description, the above system is described as being divided into various modules according to their functions. Of course, when implementing the present invention, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0156] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple. For relevant parts, refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0157] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0158] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0159] It should also be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0160] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying transient characteristics of an access point, characterized in that: include: Obtaining the recording time of each transient event recorded by each power quality monitoring device; Filtering the transient events based on the recording time, and classifying the transient events whose recording time differences are within a preset threshold into an event record group of the same time period; Obtain the monitoring point and transient event type of each transient event record in the same event record group; Merge monitoring points and transient events of the same transient event type in the same event record group; Establishing a spatiotemporal frequency matrix corresponding to the fused transient events, wherein the spatiotemporal frequency matrix records the number of transient events detected at different monitoring points in different time periods; Calculate the correlation of each monitoring point in the spatiotemporal frequency matrix; Determining whether the correlation is greater than a preset value; When it is greater than the preset value, the corresponding two monitoring points are marked as having a strong correlation; Obtain the event feature set and event frequency of transient events corresponding to each monitoring point with strong correlation; Determine whether a monitoring point with a strong correlation can be constructed as a sensitive user access point based on the event feature set and event frequency; The calculation process of the preset threshold is as follows: , n1 is the total number of transient events before fusion, and n2 is the total number of transient events after fusion; Establish the same period screening threshold time_max and event repetition rate The relationship curve between The time period screening threshold time_max corresponding to the position with the maximum curvature in the relationship curve is used as the preset threshold; The preset value is determined as follows: The monthly average of the total number of event records throughout the year is used as a boundary to distinguish between medium-high frequency areas and low frequency areas; Take the maximum value of the correlation coefficient between the frequency matrix of each medium and high frequency area and other areas as the candidate value; The minimum value among the alternative values ​​is used as the preset value.

2. The method for identifying transient characteristics of access points according to claim 1, wherein: The calculation of the correlation of each monitoring point in the spatiotemporal frequency matrix includes: Based on the formula , calculate the correlation of each element in the spatiotemporal frequency matrix ; Where: X(i) and X(j) represent the frequency matrices of the i-th and j-th spatial nodes in each time period, respectively, and their elements are X(i) = [x1(i), x2(i), …, xn(i)], X(j) = [x1(j), x2(j), …, xn(j)], where , where m is the number of elements in the space-time frequency matrix, and each element corresponds to a spatial node.

3. The method for identifying transient characteristics of access points according to claim 1, wherein: Judging whether a monitoring point with a strong correlation can be constructed as a sensitive user access point based on the event feature set and the event frequency includes: Determining that a preset feature value in the event feature set is greater than a preset value; Determining whether the event frequency is greater than a preset frequency; When the preset feature value in the event feature set is greater than a preset value and / or the event frequency is greater than a preset frequency, it indicates that the monitoring point cannot be constructed as a sensitive user access point.

4. The method for identifying transient characteristics of access points according to claim 3, wherein: include: The preset characteristic value includes a residual voltage amplitude; The preset frequency is the average frequency of the fused transient features corresponding to the monitoring points.

5. A device for identifying transient characteristics of an access point, characterized in that: include: a screening unit, configured to obtain a recording time of each transient event recorded by each power quality monitoring device; The transient events are screened based on the recording time, and the transient events whose recording time difference is within a preset threshold are classified into an event recording group of the same time period. The calculation process of the preset threshold is as follows: , n1 is the total number of transient events before fusion, and n2 is the total number of transient events after fusion; Establish the same period screening threshold time_max and event repetition rate The relationship curve between The time period screening threshold time_max corresponding to the position with the maximum curvature in the relationship curve is used as the preset threshold; A fusion unit is used to obtain the monitoring points and transient event types of each transient event record in the same event record group; and fuse transient events with the same monitoring points and transient event types in the same event record group; A correlation analysis unit is used to establish a spatiotemporal frequency matrix corresponding to the fused transient events, wherein the spatiotemporal frequency matrix records the number of transient events detected at different monitoring points in different time periods; and calculate the correlation of each monitoring point in the spatiotemporal frequency matrix; A strong correlation marking unit is used to determine whether the correlation is greater than a preset value; when it is greater than the preset value, the corresponding two monitoring points are marked as having a strong correlation relationship; The preset value is determined as follows: The monthly average of the total number of event records throughout the year is used as a boundary to distinguish between medium-high frequency areas and low frequency areas; Take the maximum value of the correlation coefficient between the frequency matrix of each medium and high frequency area and other areas as the candidate value; Taking the minimum value among the alternative values ​​as the preset value; The access point analysis unit is used to obtain the event feature set and event frequency of transient events corresponding to each monitoring point with a strong correlation; based on the event feature set and event frequency, it is determined whether the monitoring point with a strong correlation can be constructed as a sensitive user access point.

6. The access point transient feature recognition device according to claim 5, characterized in that: When calculating the correlation of each monitoring point in the spatiotemporal frequency matrix, the correlation analysis unit is specifically used to: Based on the formula , calculate the correlation of each element in the spatiotemporal frequency matrix ; Where: X(i) and X(j) represent the frequency matrices of the i-th and j-th spatial nodes in each time period, respectively, and their elements are X(i) = [x1(i), x2(i), …, xn(i)], X(j) = [x1(j), x2(j), …, xn(j)], where , where m is the number of elements in the space-time frequency matrix, and each element corresponds to a spatial node.

7. The access point transient feature recognition device according to claim 5, characterized in that: When the access point analysis unit determines whether a monitoring point with a strong correlation relationship can be constructed as a sensitive user access point based on the event feature set and the event frequency, it is specifically used to: Determining that a preset feature value in the event feature set is greater than a preset value; Determining whether the event frequency is greater than a preset frequency; When the preset feature value in the event feature set is greater than a preset value and / or the event frequency is greater than a preset frequency, it indicates that the monitoring point cannot be constructed as a sensitive user access point.

8. An access point transient feature recognition device, characterized in that: include: including memory and processor; The memory is used to store programs; The processor is configured to execute the program to implement each step of the method for identifying transient features of an access point according to any one of claims 1 to 5.

Images