Low-voltage power grid data processing method, device, equipment, medium and product

Abstract

The invention provides a low-voltage power grid data processing method and device, equipment, a medium and a product, and the method comprises the steps: collecting historical data and real-time data of a low-voltage power grid when a low-voltage power failure event is detected; constructing a historical data rule base according to the historical data of the low-voltage power grid, wherein the historical data rule base comprises at least one rule for determining a low-voltage power failure event based on the historical data; constructing a real-time feature vector according to the real-time data; obtaining a target rule matched with the real-time feature vector from a historical data rule base; and determining the power failure reason of the low-voltage power grid according to the target rule. According to the technical scheme, the real-time performance and accuracy of power grid data processing are improved.

Description

Technical Field

[0001] The present application relates to the field of smart grid technology, and in particular to a method, device, equipment, medium and product for processing low-voltage grid data. Background Art

[0002] With the rapid development of new power systems, the large-scale integration of distributed power sources (such as photovoltaics and energy storage), electric vehicle charging stations, and diverse loads in low-voltage distribution networks has led to a high degree of complexity and multi-sourced causes of power outages. Therefore, it is urgent to build a low-voltage power outage attribution technology system that adapts to highly complex scenarios to improve power system reliability.

[0003] In the existing technology, an overload judgment mechanism based on a fixed threshold is usually adopted to identify abnormal equipment through a preset current upper limit value. In terms of feature extraction, the equipment operating parameters are considered, and the rule matching process generally uses hard conditions to determine the cause of the current low-voltage power outage event.

[0004] However, the above methods all use fixed data for attribution analysis, which reduces the real-time performance and accuracy of data processing. Summary of the Invention

[0005] The present application provides a method, device, equipment, medium and product for processing low-voltage power grid data, so as to solve the technical problems of low real-time performance and accuracy of existing data processing.

[0006] In a first aspect, the present application provides a method for processing low-voltage power grid data, comprising:

[0007] When a low-voltage power outage is detected, historical and real-time data of the low-voltage power grid are collected;

[0008] Building a historical data rule base according to the historical data of the low-voltage power grid, wherein the historical data rule base includes at least one rule for determining the low-voltage power outage event based on the historical data;

[0009] constructing a real-time feature vector according to the real-time data;

[0010] Obtaining a target rule that matches the real-time feature vector from the historical data rule library;

[0011] According to the target rule, a power outage cause of the low-voltage power grid is determined.

[0012] Furthermore, a historical data rule base is constructed based on the historical data of the low-voltage power grid, including:

[0013] According to the historical data of the low-voltage power grid, initial rules corresponding to the historical data are obtained, wherein the initial rules include trigger conditions, attribution results, basic weights, dynamic weights, and environmental sensitivity;

[0014] updating the basic weight, the dynamic weight, and the environmental sensitivity in each of the initial rules to obtain an updated rule;

[0015] According to the update rule, a historical data rule base corresponding to the historical data is constructed.

[0016] Furthermore, constructing a real-time feature vector based on the real-time data includes:

[0017] Determining characteristic information in the real-time data according to the real-time data;

[0018] Comparing each of the feature information with the preset feature requirements to obtain feature comparison results;

[0019] According to the feature comparison results, time features, environment features, device status features and user behavior features are obtained;

[0020] constructing a real-time feature vector according to the time feature, the environment feature, the device state feature, and the user behavior feature;

[0021] The characteristic information includes time information, weather information, device status information and user behavior information.

[0022] Furthermore, obtaining a target rule that matches the real-time feature vector from the historical data rule base includes:

[0023] Matching the real-time feature vector with the rules in the historical data rule base to determine the rule to be verified;

[0024] Determining the basic weight and dynamic weight of each rule to be verified according to the basic weight and dynamic weight of each rule in the historical data rule base;

[0025] Linear processing is performed on the basic weight and dynamic weight of each of the rules to be verified to determine a target rule that meets the preset rule requirements among the rules to be verified.

[0026] Furthermore, linear processing is performed on the basic weight and dynamic weight of each of the rules to be verified to determine the target rule that meets the preset rule requirements among the rules to be verified, including:

[0027] Performing linear processing on the basic weight and dynamic weight of each of the rules to be verified to obtain a comprehensive weight of each of the rules to be verified;

[0028] Compare the comprehensive weight of each rule to be verified with the preset weight requirement to obtain a weight comparison result;

[0029] According to the weight comparison result, a target rule that meets the preset rule requirements among the rules to be verified is determined.

[0030] Furthermore, the real-time feature vector is matched with the rules in the historical data rule base to determine the rule to be verified, and the method further includes:

[0031] Matching the real-time feature vector with the rules in the historical data rule base to obtain a rule matching result;

[0032] If the rule matching result indicates that the rule in the historical data rule base does not match the real-time feature vector, similarity calculation is performed on the triggering condition of the real-time feature vector and the triggering condition of each rule in the historical data rule base to obtain a triggering condition similarity;

[0033] performing an updating process on the historical data rule base according to the trigger condition similarity to obtain an updated historical data rule base;

[0034] According to the updated historical data rule base, a rule to be verified that matches the real-time feature vector is determined.

[0035] Furthermore, the historical data rule base is updated according to the trigger condition similarity to obtain an updated historical data rule base, including:

[0036] Summing the trigger condition similarity with the basic weight, dynamic weight and environmental sensitivity of each rule in the historical data rule base to obtain an updated basic weight, an updated dynamic weight and an updated environmental sensitivity;

[0037] An updated historical data rule base is obtained according to the updated basic weight, the updated dynamic weight, the updated environmental sensitivity, the attribution results in the historical data rule base, and the triggering condition of the real-time feature vector.

[0038] In a second aspect, the present application provides a low-voltage power grid data processing device, comprising:

[0039] A data acquisition module is used to collect historical and real-time data of the low-voltage power grid when a low-voltage power outage event is detected;

[0040] A rule base construction module, configured to construct a historical data rule base according to historical data of the low-voltage power grid, wherein the historical data rule base includes at least one rule for determining the low-voltage power outage event based on the historical data;

[0041] A feature vector acquisition module, configured to construct a real-time feature vector based on the real-time data;

[0042] A target rule acquisition module is used to acquire a target rule that matches the real-time feature vector from the historical data rule library;

[0043] The power outage cause determination module is used to determine the power outage cause of the low-voltage power grid according to the target rule.

[0044] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0045] The memory stores computer-executable instructions;

[0046] The processor executes the computer-executable instructions stored in the memory to implement the method according to the first aspect.

[0047] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used for the method described in the first aspect.

[0048] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method as described in any one of the first aspects.

[0049] The present application provides a method, apparatus, device, medium, and product for processing low-voltage power grid data. These methods include collecting historical and real-time data from the low-voltage power grid upon detecting a low-voltage power outage; constructing a historical data rule base based on the historical data of the low-voltage power grid, the historical data rule base including at least one rule for determining low-voltage power outage events based on the historical data; constructing a real-time feature vector based on the real-time data; obtaining a target rule from the historical data rule base that matches the real-time feature vector; and determining the cause of the low-voltage power outage based on the target rule. This technical solution improves the real-time and accuracy of power grid data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] Figure 1 A flowchart of a first embodiment of a method for processing low-voltage power grid data provided in this application;

[0052] Figure 2 A flowchart of a second embodiment of a method for processing low-voltage power grid data provided in this application;

[0053] Figure 3 This is a flow chart of Embodiment 3 of the method for processing low-voltage power grid data proposed in this application;

[0054] Figure 4 This is a flow chart of a fourth embodiment of the method for processing low-voltage power grid data proposed in this application;

[0055] Figure 5 This is a flowchart of a fifth embodiment of the method for processing low-voltage power grid data proposed in this application;

[0056] Figure 6 This is a schematic diagram of the structure of the low-voltage power grid data processing device proposed in this application;

[0057] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0058] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0059] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0060] Existing low-voltage power grid data processing methods rely on manual experience and static rule bases. Data processing is mainly based on fixed threshold judgment, with long update cycles and poor adaptability. Feature extraction only focuses on equipment operating parameters, ignoring user behavior and environmental information, resulting in rigid rule matching and high misjudgment rate, which limits the timeliness and accuracy of fault location.

[0061] To address these technical challenges, the present invention achieves intelligent processing of low-voltage power grid data by constructing a dynamic rule base that integrates historical and real-time data, combined with feature vector matching, adaptive weight fusion, and a similarity-driven rule evolution mechanism. This method supports dynamic rule updates and self-learning, enhancing responsiveness to diverse load scenarios and sudden faults, improving the accuracy and timeliness of fault attribution, and effectively supporting the intelligent upgrade of power grid data processing.

[0062] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0063] Figure 1 This is a flow chart of Example 1 of the method for processing low-voltage power grid data provided in this application.

[0064] like Figure 1 Shown, including:

[0065] S101. When a low-voltage power outage event is detected, historical data and real-time data of the low-voltage power grid are collected.

[0066] The low-voltage grid refers to a distribution network with a rated voltage of no more than 1 kV. Historical data refers to various operational data stored in the low-voltage grid before a power outage occurs, including current, voltage, load, and environmental parameters. Real-time data refers to grid operational data collected immediately during or after a power outage.

[0067] In this step, two key data need to be pre-processed and calculated, including the calculation of the dynamic overload current threshold and the calculation of the user's repair keyword weight. Specifically, the calculation formula for the dynamic overload current threshold is:

[0068] I max，t =μI+K·σ I

[0069] Among them, I max,t Indicates the dynamic overload current threshold, μ I Represents the mean current in the sliding window of the preset time, σ I It represents the variance of the current within the sliding window of the preset time, and K represents the safety factor, which is used to adjust the alarm sensitivity.

[0070] The calculation formula for the user repair keyword weight is:

[0071]

[0072] Among them, ω j A represents the weight of the jth user's repair keyword, j A represents the number of occurrences of the jth user's repair keyword, total Indicates the total number of repair reports submitted by users.

[0073] For example, when a low-voltage power outage is detected, historical data and real-time data are collected in real time, and data cleaning is performed on the historical data and real-time data, such as removing missing values ​​and outliers, and repairing erroneous data. Then, based on the historical current data and the set time sliding window, the dynamic overload current threshold is calculated in real time. This threshold is used to determine whether the current current fluctuates abnormally and assist in attributing the power outage. At the same time, keyword weight statistics are performed. Based on the user's repair record, keywords (such as "tripping", "flash disconnection", "transformer", etc.) are extracted, the frequency of occurrence of each keyword is counted, and its weight is calculated for subsequent association analysis. Finally, all types of feature data are normalized to ensure that different features are comparable under the same dimension. Through the above steps, a high-quality data set that can be used for library construction and feature extraction is obtained.

[0074] S102: Construct a historical data rule base based on historical data of the low-voltage power grid.

[0075] The historical data rule base includes at least one rule for determining a low-voltage power outage event based on historical data.

[0076] For example, preprocessed historical data is used to construct a historical data rule base. This involves extracting key features from historical power outage events, such as overcurrent, undervoltage, and equipment tripping, and then generating rules based on machine learning methods. Finally, these generated rules are stored, for example, in a structured format in a database for subsequent real-time access.

[0077] S103: Construct a real-time feature vector based on the real-time data.

[0078] Among them, the real-time feature vector is a multi-dimensional feature set extracted from real-time data to describe the current power grid state for use in subsequent rule matching.

[0079] For example, based on real-time data collected during a power outage, features highly correlated with the outage's impact can be selected, such as instantaneous current change, voltage fluctuation, and trip number. These selected features are then normalized to generate standardized values. Finally, the standardized features are arranged in a uniform order to form a one-dimensional or multi-dimensional vector, which serves as a real-time feature description of the current event.

[0080] S104: Obtain target rules that match the real-time feature vector from the historical data rule library.

[0081] The target rule refers to the rule in the historical data rule base that has the highest matching degree with the current real-time feature vector and meets the triggering conditions.

[0082] Specifically, in this step, the target rules in the historical data rule base are determined by comparing the real-time feature vector with each rule in the historical data rule base. For example, rules that meet the trigger conditions are compared one by one to select the rules that meet the requirements. Then, the combined weight of the selected candidate rules is calculated by combining their base weight and dynamic weight. Finally, the candidate rules are ranked according to their combined weight, and one or more of the rules with the highest scores are selected as the final target rules.

[0083] S105. Determine the cause of the power outage in the low-voltage power grid according to the target rule.

[0084] The power outage cause refers to the specific technical reason that caused the low-voltage power outage, such as a short circuit, equipment failure, or external damage. For example, based on the matched target rule, the power outage cause is inferred. Specifically, this may include reading the attribution results recorded in the target rule as a preliminary judgment result. If multiple rules have similar weights, a weighted average or voting mechanism is used for comprehensive judgment to improve accuracy.

[0085] The embodiment of the present application combines the deep integration of historical data and real-time data, builds a rule base and dynamically adjusts the weights, which significantly improves the accuracy and response speed of determining the cause of power outages, effectively supports intelligent decision-making and rapid fault location in power grid operation and maintenance, and improves the reliability and power supply continuity of the power grid.

[0086] Figure 2 This is a flow chart of the second embodiment of the method for processing low-voltage power grid data provided by this application. Figure 2 As shown, in Figure 1 On the basis of the embodiment, a historical data rule base is constructed according to the historical data of the low-voltage power grid, including:

[0087] S201 : According to historical data of the low-voltage power grid, initial rules corresponding to the historical data are obtained.

[0088] Among them, the initial rules are constructed based on historical power outage data, including trigger conditions, attribution results, basic weights, dynamic weights and environmental sensitivity.

[0089] In the embodiment of the present application, the trigger condition refers to the specific judgment condition for triggering the rule in the historical data, such as load mutation, voltage drop, etc.; the attribution result refers to the cause of the power outage inferred based on the trigger condition, such as short circuit or external force damage; the basic weight represents the frequency of occurrence or confidence index of the rule in the historical data, and the initial index can be set to 0.5; the dynamic weight is a confidence index that can be adjusted over time, which takes into account the matching accuracy and attenuation factor; environmental sensitivity refers to the sensitivity index of the rule to different environmental conditions (such as extreme weather).

[0090] For example, after analyzing historical low-voltage grid data, initial rules were extracted using a rule-based approach. Each rule includes a set of trigger conditions, such as current exceeding a threshold, voltage sag, and frequent repair keyword reporting; an attribution result, such as "line aging" or "concentrated load"; and initial assigned basic weights, dynamic weights, and environmental sensitivity, all with a uniform initial value of 0.5. All rules are stored in a structured rule base for subsequent dynamic optimization.

[0091] S202: Update the basic weight, dynamic weight and environmental sensitivity in each initial rule to obtain an updated rule.

[0092] In this step, the update formula for the basic weight is as follows:

[0093]

[0094] Among them, ω 基础 ,i represents the basic weight of the i-th rule, Δt k Indicates the time interval between the kth correct match of the i-th rule and the current time, Δt m Indicates the time interval between the mth match of the i-th rule and the current time, Nt otal,i Indicates the total number of times the i-th rule is matched, N hit,i It represents the total number of correct matches of the i-th rule, and β represents the time decay coefficient.

[0095] The update formula for dynamic weights is as follows:

[0096] ω 动态，i =α1·ACC i +α2·Sens i +α3·e -β·Δt

[0097]

[0098] Among them, ω 动态,i represents the dynamic weight of the i-th rule, α1, α2, α3 represent the dynamic adjustment coefficients, ACC i represents the recent accuracy of the i-th rule, N correct,i Indicates the number of correct matches of the i-th rule in the past 24 hours, N match,i Indicates the total number of matches for the i-th rule in the past 24 hours, Sens i represents the environmental sensitivity of the i-th rule, Δt represents the time interval between the last match of the i-th rule and the current time, and β represents the time attenuation coefficient.

[0099] Among them, the updated calculation formula of environmental sensitivity is:

[0100]

[0101] Among them, c represents the cth extreme weather type, C represents the total number of extreme weather types, N total,i,c N represents the total number of times the i-th rule matches under the c-th extreme weather type, hit,i,c represents the total number of correct matches of the i-th rule under the c-th extreme weather type, and λc represents the type weight coefficient.

[0102] Specifically, based on the historical matching records of the rules, the weight ratio of successful matches is calculated, and the time decay coefficient is used to weaken the impact of old data and enhance the reference value of recent data; the matching accuracy (ACC), environmental adaptability (Sens), and time decay factor of the latest events of the rules in the past 24 hours are considered to combine and evaluate their immediate credibility; these update processes can optimize the processing results.

[0103] S203: Construct a historical data rule base corresponding to the historical data according to the update rule.

[0104] In this step, a structured historical data rule base is constructed based on the updated rule set. All fields are recorded in a unified format and can be quickly called in an in-memory database or a local relational database. This rule base will be used to efficiently match real-time feature vectors, improving the accuracy and timeliness of power outage attribution.

[0105] By periodically updating the basic weights, dynamic weights, and environmental sensitivity, this embodiment of the application can dynamically reflect the actual operating environment and regular changes of the low-voltage power grid. Compared with traditional static rule models, this technical solution significantly improves the real-time reliability, anti-interference ability, and adaptability of the attribution method, providing stronger support for subsequent intelligent diagnosis and decision support.

[0106] Figure 3 This is a flow chart of the third embodiment of the method for processing low-voltage power grid data proposed in this application. Figure 3 As shown, in Figure 1 Based on the embodiment, a real-time feature vector is constructed according to real-time data, including:

[0107] S301: Determine feature information in the real-time data based on the real-time data.

[0108] Among them, feature information is a structured information unit extracted from real-time data, including time information, weather information, device status information and user behavior information.

[0109] For example, after collecting real-time data, it is first divided into four feature sources: time information, weather information, device status information, and user behavior information. Time information refers to the current time, day of the week, or holiday. Weather information refers to the presence of extreme weather conditions such as thunderstorms, strong winds, and high temperatures. Device status information refers to the ratio of current or voltage to its corresponding threshold. User behavior information refers to the text of user-submitted repair tickets, keywords, and their frequency. Extracting this information facilitates subsequent feature comparison.

[0110] S302: Compare each feature information with the preset feature requirements to obtain feature comparison results.

[0111] The preset feature requirements refer to predefined threshold conditions or patterns, such as "current > threshold is abnormal" or "load is high during the morning peak period." The feature comparison results are quantitative outputs indicating whether each feature meets the preset requirements and are used to encode the feature value.

[0112] For example, the extracted feature information is sequentially compared with preset logic or thresholds. For example, time information comparison determines whether the current time is during peak electricity consumption (e.g., 7:00-9:00 a.m. and 18:00-21:00 p.m. on weekdays); weather information comparison identifies whether the current weather is extreme through sensor data; current comparison calculates the ratio of the current current to a dynamic threshold to determine whether it exceeds the standard; and user behavior comparison matches the repair work order for high-frequency keywords such as "flash outage" and "tripping." Each of these comparisons generates a Boolean or numerical result, which is used as the basis for constructing feature values ​​in the next step.

[0113] S303: Obtain time features, environment features, device status features, and user behavior features based on the feature comparison results.

[0114] Among them, the time feature is used to determine whether the current electricity consumption is at its peak, the environmental feature is used to determine whether the current weather is extreme, the equipment status feature is used to determine the degree of current exceeding the standard, and the user behavior feature is used to determine the intensity of the user's repair report information.

[0115] Specifically, for the time feature f1, if the current time is during peak electricity consumption, the time feature is recorded as 1, otherwise it is recorded as 0; for the environmental feature f2, an identification value is generated based on weather conditions (such as thunderstorms and typhoons), using standardized numbers or intensity scores; for the equipment status feature f3, the ratio of the actual current value to the dynamic overload threshold is output as a continuous value; and for the user behavior feature f4, keywords are extracted based on the repair work order and summed up according to the corresponding weights to obtain the user behavior feature.

[0116] Extract keywords based on user repair work orders in real-time data, and sum and calculate the weights of user repair keywords corresponding to the extracted keywords to obtain user behavior feature f4;

[0117] S304: Construct a real-time feature vector based on time features, environment features, device status features, and user behavior features.

[0118] The real-time feature vector is a one-dimensional vector composed of the four types of feature values ​​obtained in S303, which represents the status of the current event in terms of time, environment, device, and user behavior dimensions.

[0119] For example, the above four types of features f1, f2, f3, and f4 are arranged in a preset order to form a final real-time feature vector:

[0120] F t =[f1, f2, f3, f4]

[0121] Each feature can be a binary value, a floating point value, or a categorical value; the vector F t This is the core input for subsequent comparisons with the rule base and is used for fault attribution analysis. Furthermore, this real-time feature vector supports dynamic updates, meaning it can be periodically recalculated as real-time data changes, enabling sliding window matching analysis.

[0122] The embodiment of the present application effectively characterizes the operating status and external environment characteristics of the low-voltage power grid at a specific time point by extracting multi-dimensional features from real-time data and structuring them into feature vectors, significantly improving the accuracy and adaptability of subsequent rule matching and fault attribution.

[0123] Figure 4 This is a flow chart of the fourth embodiment of the method for processing low-voltage power grid data proposed in this application. Figure 4 As shown, in Figure 1 Based on the embodiment, the target rule that matches the real-time feature vector is obtained from the historical data rule library, including:

[0124] S401: Match the real-time feature vector with the rules in the historical data rule library to determine the rules to be verified.

[0125] Among them, the rules to be verified refer to a set of rules that match the current real-time feature vector but have not yet been confirmed to be appropriate attribution results. Specifically, after receiving the real-time feature vector, it is matched with the trigger conditions of each rule in the rule base to determine whether the feature conditions meet the preset trigger conditions. Among them, the matching strategy can adopt threshold matching, fuzzy matching (such as keyword similarity) or multi-dimensional vector distance judgment (such as Euclidean distance is less than the set threshold). Finally, all rules that meet the matching conditions are marked as a set of rules to be verified for subsequent weight scoring and sorting.

[0126] S402: Determine the basic weight and dynamic weight of each rule to be verified based on the basic weight and dynamic weight of each rule in the historical data rule library.

[0127] In this step, each rule to be verified is queried to extract its most recently updated base weight and most recently updated dynamic weight. Environmental sensitivity is already integrated into the dynamic weight, eliminating the need for repeated extraction. This weight information serves as input for the next step, calculating the comprehensive weight to select target rules.

[0128] S403: Perform linear processing on the basic weight and dynamic weight of each rule to be verified, and determine the target rule that meets the preset rule requirements among the rules to be verified.

[0129] In this step, the basic weight and dynamic weight of each rule to be verified need to be linearly processed to obtain the comprehensive weight of each rule to be verified;

[0130] Compare the comprehensive weight of each rule to be verified with the preset weight requirement to obtain the weight comparison result;

[0131] According to the weight comparison results, the target rule that meets the preset rule requirements among the rules to be verified is determined.

[0132] The preset weight requirement is a pre-set minimum trustworthy threshold, and rules below this threshold are deemed to be ineligible.

[0133] For example, the adaptive balancing coefficient λ is first calculated using the following formula:

[0134]

[0135] Among them, λ represents the adaptive balance coefficient, V represents the environmental change rate, γ represents the adjustment coefficient, θ represents the environmental stability threshold, E1 represents the number of low-voltage power outage events that occurred in the past hour, and E 24 represents the average number of low-voltage power outages over the past 24 hours. Specifically, when the environment fluctuates drastically (i.e., when V is high), λ approaches 0, indicating a greater reliance on dynamic weights. When the environment is stable, λ approaches 1, indicating a greater reliance on historical base weights.

[0136] Then, calculate the comprehensive weight, the formula is as follows:

[0137]

[0138] Among them, W i represents the comprehensive weight of the i-th candidate rule, ω 基础 ,i represents the basic weight of the i-th rule, ω 动态,i represents the dynamic weight of the i-th rule. Specifically, the first term is the product term, which considers the interaction between the basic weight and the dynamic weight; the second term is the mean of the two, which avoids the influence of unilateral extreme weight on the comprehensive weight; the square root processing makes the comprehensive weight W i Maintaining a moderate scaling range enhances the discrimination of weight scores in attribution ranking. Therefore, the combined weight W of all rules to be verified is i All can be calculated one by one through this formula.

[0139] Finally, the comprehensive weight of each rule to be verified is compared with the pre-set weight threshold to determine the target rule that meets the weight requirement among the rules to be verified. For example, the pre-set weight threshold is W min is 0.6, then 0.6 is the minimum weight requirement of the “trusted rule”; then the combined weight W of all the rules to be verified is i and weight threshold W min Compare and mark the comprehensive weight W i ≥ weight threshold W min In addition, if there are multiple target rules, you can i Sort and select the rule with the highest comprehensive weight as the final target rule.

[0140] The embodiment of the present application realizes the intelligent fusion of basic weights and dynamic weights by introducing an adaptive balancing coefficient, and adopts a nonlinear comprehensive scoring model to evaluate and screen the verification rules, thereby improving the accuracy and environmental adaptability of rule selection, and significantly enhancing the dynamic response capability, reliability and judgment accuracy of the low-voltage power grid power outage attribution system under complex working conditions, providing a solid foundation for realizing intelligent and efficient fault analysis.

[0141] Figure 5 This is a flow chart of the fifth embodiment of the method for processing low-voltage power grid data proposed in this application. Figure 5 ,exist Figure 4 Based on the embodiment, the real-time feature vector is matched with the rules in the historical data rule library to determine the rule to be verified. The method further includes:

[0142] S501: Match the real-time feature vector with the rules in the historical data rule library to obtain a rule matching result.

[0143] The rule matching result refers to the result of determining whether the real-time feature vector meets the triggering condition of any rule in the rule base, which can be "matched successfully" or "no match".

[0144] Specifically, the real-time feature vector is compared one by one with the trigger conditions of each rule in the historical data rule library. If a rule meets the conditions, it is recorded as a pending rule and enters the attribution process. If no rule is matched during the comparison process, it means that a reasonable attribution cannot be made for the event at this time, and the similar rule identification and rule library update mechanism will be entered.

[0145] S502: If the rule matching result indicates that the rule in the historical data rule base does not match the real-time feature vector, similarity calculation is performed on the triggering condition of the real-time feature vector and the triggering condition of each rule in the historical data rule base to obtain the triggering condition similarity.

[0146] Among them, the trigger condition similarity refers to the numerical value that measures the "feature pattern similarity" between the current event and the historical rules, and is generally expressed using vector similarity, such as cosine similarity.

[0147] For example, when no rule match is detected, the event is reported to the operation and maintenance personnel for manual annotation to confirm its actual triggering conditions and attribution results. The similarity between the manually annotated triggering conditions and the triggering conditions of each rule in the rule base is then calculated. The similarity calculation can use cosine similarity, and the formula is:

[0148]

[0149] According to the above similarity calculation, the similarity value of each rule relative to the trigger condition of the current event can be obtained, and the above similarity value is used as a measurement indicator to facilitate the initialization processing of subsequent rules.

[0150] S503: Update the historical data rule base according to the trigger condition similarity to obtain an updated historical data rule base.

[0151] Among them, the updated historical data rule base refers to a complete structured rule set including the original rules and the newly added learning rules.

[0152] Specifically, the trigger condition similarity is summed with the basic weight, dynamic weight and environmental sensitivity of each rule in the historical data rule base to obtain the updated basic weight, updated dynamic weight and updated environmental sensitivity;

[0153] According to the update basic weight, the update dynamic weight, the update environment sensitivity, the attribution result in the historical data rule base and the triggering condition of the real-time feature vector, the update historical data rule base is obtained.

[0154] In this step, the similarity result is weighted and summed with the basic weight, dynamic weight, and environmental sensitivity of the existing rules to construct a new initial value. The formula is as follows:

[0155]

[0156] Among them, sim i The updated value is combined with the manually annotated trigger conditions and attribution results to generate a new rule structure: the trigger is the feature vector of the current event, the attribution result is the expert annotation, and the weight parameter is the superposition of the similarity values. Finally, this new rule is added to the historical data rule library to form a "self-learning update rule."

[0157] S504: Determine a rule to be verified that matches the real-time feature vector according to the updated historical data rule base.

[0158] In this step, the original real-time feature vector is re-matched with all the rules in the updated rule base to obtain a new set of rules to be verified. After this, you can re-enter Figure 4 The attribution process shown can realize a closed-loop process of rule self-update through the update process, thereby improving the data processing capability of the low-voltage power grid when facing unknown scenarios.

[0159] The embodiment of the present application introduces a self-learning update framework that combines a similarity-driven rule generation mechanism with manual annotation, so that when faced with unknown or sudden power outages, the rule base can be continuously expanded, adaptively enhanced, and knowledge accumulated, thereby improving the accuracy, adaptability, and maintainability of the long-term operation of the low-voltage power grid and building a smart grid fault attribution platform with cognitive capabilities.

[0160] Figure 6 This is a schematic diagram of the structure of the low-voltage power grid data processing device proposed in this application. Figure 6 As shown, the low-voltage grid data processing device 60 includes:

[0161] The data acquisition module 601 is used to collect historical data and real-time data of the low-voltage power grid when a low-voltage power outage event is detected;

[0162] A rule base construction module 602 is configured to construct a historical data rule base based on historical data of the low-voltage power grid, wherein the historical data rule base includes at least one rule for determining a low-voltage power outage event based on the historical data;

[0163] The feature vector acquisition module 603 is used to construct a real-time feature vector based on real-time data;

[0164] The target rule acquisition module 604 is used to obtain the target rule that matches the real-time feature vector from the historical data rule library;

[0165] The power outage cause determination module 605 is configured to determine the power outage cause of the low-voltage power grid according to target rules.

[0166] Furthermore, the rule base construction module 602 is further specifically configured to:

[0167] Based on the historical data of the low-voltage power grid, initial rules corresponding to the historical data are obtained. The initial rules include trigger conditions, attribution results, basic weights, dynamic weights, and environmental sensitivity.

[0168] The basic weight, dynamic weight and environmental sensitivity of each initial rule are updated respectively to obtain the updated rule;

[0169] According to the update rules, a historical data rule base corresponding to the historical data is constructed.

[0170] Furthermore, the feature vector acquisition module 603 is further specifically configured to:

[0171] According to the real-time data, determining characteristic information in the real-time data;

[0172] Compare each feature information with the preset feature requirements to obtain feature comparison results;

[0173] According to the feature comparison results, time features, environmental features, device status features and user behavior features are obtained;

[0174] Construct real-time feature vectors based on time features, environmental features, device status features, and user behavior features;

[0175] Among them, feature information includes time information, weather information, device status information and user behavior information.

[0176] Furthermore, the target rule acquisition module 604 is further specifically configured to:

[0177] Match the real-time feature vector with the rules in the historical data rule base to determine the rules to be verified;

[0178] Determine the basic weight and dynamic weight of each rule to be verified based on the basic weight and dynamic weight of each rule in the historical data rule base;

[0179] Perform linear processing on the basic weight and dynamic weight of each rule to be verified to determine the target rule that meets the preset rule requirements among the rules to be verified.

[0180] Furthermore, the target rule acquisition module 604 is further specifically configured to:

[0181] Perform linear processing on the basic weight and dynamic weight of each rule to be verified to obtain the comprehensive weight of each rule to be verified;

[0182] Compare the comprehensive weight of each rule to be verified with the preset weight requirement to obtain the weight comparison result;

[0183] According to the weight comparison results, the target rule that meets the preset rule requirements among the rules to be verified is determined.

[0184] Furthermore, the target rule acquisition module 604 is further specifically configured to:

[0185] Match the real-time feature vector with the rules in the historical data rule base to obtain the rule matching result;

[0186] If the rule matching result indicates that the rule in the historical data rule base does not match the real-time feature vector, similarity calculation is performed on the triggering condition of the real-time feature vector and the triggering condition of each rule in the historical data rule base to obtain the triggering condition similarity;

[0187] According to the similarity of the trigger conditions, the historical data rule base is updated to obtain an updated historical data rule base;

[0188] According to the updated historical data rule base, the rules to be verified that match the real-time feature vector are determined.

[0189] Furthermore, the target rule acquisition module 604 is further specifically configured to:

[0190] The trigger condition similarity is summed with the basic weight, dynamic weight and environmental sensitivity of each rule in the historical data rule base to obtain the updated basic weight, updated dynamic weight and updated environmental sensitivity;

[0191] According to the update basic weight, the update dynamic weight, the update environment sensitivity, the attribution result in the historical data rule base and the triggering condition of the real-time feature vector, the update historical data rule base is obtained.

[0192] Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 7 As shown, the electronic device 70 includes:

[0193] The electronic device 70 may include one or more processors 701 , one or more computer-readable storage media memories 702 , a communication component 703 , and other components. The processor 701 , the memory 702 , and the communication component 703 are connected via a bus 704 .

[0194] In a specific implementation process, at least one processor 701 executes the computer-executable instructions stored in the memory 702 , so that the at least one processor 701 executes the above-mentioned method for processing low-voltage power grid data.

[0195] The specific implementation process of the processor 701 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0196] In the above Figure 7 In the illustrated embodiment, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASICs), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules in the processor.

[0197] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0198] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be classified as address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0199] In some embodiments, a computer program product is further provided, comprising a computer program or instructions, which implement the steps of any of the above-mentioned methods for processing low-voltage power grid data when executed by a processor.

[0200] The specific implementation of the above operations can be found in the previous embodiments and will not be repeated here.

[0201] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be accomplished by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0202] To this end, an embodiment of the present application provides a computer-readable storage medium, which stores multiple instructions, and the instructions can be loaded by a processor to execute the steps in any low-voltage power grid data processing method provided in the embodiment of the present application.

[0203] The storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0204] According to one aspect of the present application, a computer program product or a computer program is provided. The computer program product or the computer program comprises computer instructions stored in a computer-readable storage medium.

[0205] Since the instructions stored in the storage medium can execute the steps in any low-voltage power grid data processing method provided in the embodiments of the present application, the beneficial effects that can be achieved by any low-voltage power grid data processing method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0206] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0207] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all optional embodiments, and the actions and modules involved are not necessarily required by this application.

[0208] It should be further noted that, although the various steps in the flowchart are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps may be performed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily performed at the same time, but may be performed at different times. The execution order of these sub-steps or stages is not necessarily to be performed in sequence, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0209] It should be understood that the above-described device embodiments are merely illustrative, and the device of the present application may also be implemented in other ways. For example, the division of units / modules in the above-described embodiments is merely a logical functional division, and actual implementations may employ other division methods. For example, multiple units, modules, or components may be combined or integrated into another system, or some features may be omitted or not implemented.

[0210] In addition, unless otherwise specified, the functional units / modules in the various embodiments of the present application may be integrated into a single unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The aforementioned integrated units / modules may be implemented in the form of hardware or software program modules.

[0211] If the integrated unit / module is implemented in hardware, the hardware may be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor may be any appropriate hardware processor, such as a CPU, GPU, FPGA, DSP, and ASIC. Unless otherwise specified, the storage unit may be any appropriate magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc.

[0212] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0213] In the above embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a particular embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined in any way. To keep the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0214] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0215] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for processing low-voltage power grid data, characterized in that: include: When a low-voltage power outage is detected, historical and real-time data of the low-voltage power grid are collected; Building a historical data rule base according to the historical data of the low-voltage power grid, wherein the historical data rule base includes at least one rule for determining the low-voltage power outage event based on the historical data; constructing a real-time feature vector according to the real-time data; Obtaining a target rule that matches the real-time feature vector from the historical data rule library; According to the target rule, a power outage cause of the low-voltage power grid is determined.

2. The processing method according to claim 1, characterized in that Constructing a historical data rule base based on the historical data of the low-voltage power grid, including: According to the historical data of the low-voltage power grid, initial rules corresponding to the historical data are obtained, wherein the initial rules include trigger conditions, attribution results, basic weights, dynamic weights, and environmental sensitivity; updating the basic weight, the dynamic weight, and the environmental sensitivity in each of the initial rules to obtain an updated rule; According to the update rule, a historical data rule base corresponding to the historical data is constructed.

3. The processing method according to claim 1, characterized in that Constructing a real-time feature vector based on the real-time data includes: Determining characteristic information in the real-time data according to the real-time data; Comparing each of the feature information with the preset feature requirements to obtain feature comparison results; According to the feature comparison results, time features, environment features, device status features and user behavior features are obtained; constructing a real-time feature vector according to the time feature, the environment feature, the device state feature, and the user behavior feature; The characteristic information includes time information, weather information, device status information and user behavior information.

4. The processing method according to claim 1, characterized in that Obtaining a target rule that matches the real-time feature vector from the historical data rule library includes: Matching the real-time feature vector with the rules in the historical data rule base to determine the rule to be verified; Determining the basic weight and dynamic weight of each rule to be verified according to the basic weight and dynamic weight of each rule in the historical data rule base; Linear processing is performed on the basic weight and dynamic weight of each of the rules to be verified to determine a target rule that meets the preset rule requirements among the rules to be verified.

5. The processing method according to claim 4, characterized in that: Performing linear processing on the basic weight and dynamic weight of each of the rules to be verified to determine the target rule that meets the preset rule requirements among the rules to be verified, including: Performing linear processing on the basic weight and dynamic weight of each of the rules to be verified to obtain a comprehensive weight of each of the rules to be verified; Compare the comprehensive weight of each rule to be verified with the preset weight requirement to obtain a weight comparison result; According to the weight comparison result, a target rule that meets the preset rule requirements among the rules to be verified is determined.

6. The processing method according to claim 4, characterized in that Matching the real-time feature vector with the rules in the historical data rule base to determine the rule to be verified, the method further comprising: Matching the real-time feature vector with the rules in the historical data rule base to obtain a rule matching result; If the rule matching result indicates that the rule in the historical data rule base does not match the real-time feature vector, similarity calculation is performed on the triggering condition of the real-time feature vector and the triggering condition of each rule in the historical data rule base to obtain a triggering condition similarity; performing an updating process on the historical data rule base according to the trigger condition similarity to obtain an updated historical data rule base; According to the updated historical data rule base, a rule to be verified that matches the real-time feature vector is determined.

7. The processing method according to claim 6, characterized in that The historical data rule base is updated according to the trigger condition similarity to obtain an updated historical data rule base, including: Summing the trigger condition similarity with the basic weight, dynamic weight and environmental sensitivity of each rule in the historical data rule base to obtain an updated basic weight, an updated dynamic weight and an updated environmental sensitivity; An updated historical data rule base is obtained according to the updated basic weight, the updated dynamic weight, the updated environmental sensitivity, the attribution results in the historical data rule base, and the triggering condition of the real-time feature vector.

8. A low-voltage power grid data processing device, characterized in that: include: A data acquisition module is used to collect historical and real-time data of the low-voltage power grid when a low-voltage power outage event is detected; A rule base construction module, configured to construct a historical data rule base according to historical data of the low-voltage power grid, wherein the historical data rule base includes at least one rule for determining the low-voltage power outage event based on the historical data; A feature vector acquisition module, configured to construct a real-time feature vector based on the real-time data; A target rule acquisition module is used to acquire a target rule that matches the real-time feature vector from the historical data rule library; The power outage cause determination module is used to determine the power outage cause of the low-voltage power grid according to the target rule.

9. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

11. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 7 when the computer program is executed by a processor.

Images