Distribution network equipment transaction monitoring analysis method and system based on rule template

Abstract

The invention discloses a distribution network equipment transaction monitoring analysis method and system based on a rule template, and relates to the technical field of data transaction analysis, and the system comprises a data acquisition module, a model training module, a rule template matching module, a rule template adjustment module and a visual display module. The data acquisition module is used for acquiring historical meteorological temperature and historical data of user side load power, and acquiring a rule template set by the distribution network equipment based on the size of the user side load power. The method has the beneficial effects that the response efficiency of the distribution network equipment is improved and manual intervention is reduced by training and establishing the time sequence model, predicting the load power of the user side, matching the distribution network equipment with the corresponding rule template and performing pre-switching preparation, so that the stability and efficiency of the system are improved, the abnormal condition is monitored in real time, the rule template is optimized and adjusted, and the user experience is improved. The load power change trend is visually displayed, the accuracy of load power prediction and rule template matching is improved, and electric energy distribution is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of data anomaly analysis, and in particular to a method and system for monitoring and analyzing distribution network equipment anomalies based on rule templates. Background Art

[0002] Distribution network equipment is the core infrastructure for the distribution and transmission of electrical energy in power systems. It primarily consists of substations, distribution lines, switchgear, transformers, reactive power compensation devices, and monitoring and protection devices. These devices work together to build a power supply network connecting the power grid and end users. Its operational performance directly determines the stability, economy, and safety of power supply. Rule templates are standardized sets of judgment criteria based on business logic, historical data mining, or expert experience. They can quickly capture abnormal characteristics during equipment operation, such as parameter limits exceeding, sudden state changes, load fluctuations, and other key information, providing a crucial basis for equipment status assessment and fault warning.

[0003] Traditional distribution network equipment mostly uses a single fixed rule template. When faced with seasonal and time-based load fluctuations, such as peak electricity consumption caused by high temperatures in summer, this method cannot rationally adjust electricity distribution according to user needs. The problem of unbalanced electricity distribution often occurs, resulting in waste of electricity resources and power shortages in some areas, which has an adverse impact on residents' lives and industrial production. Summary of the Invention

[0004] The purpose of this section is to provide a distribution network equipment abnormality monitoring and analysis method and system based on rule templates, which can flexibly adjust the power distribution strategy according to the user's real-time power demand and improve the efficiency of power resource utilization.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: a method for monitoring and analyzing distribution network equipment anomalies based on rule templates, comprising the following steps: S100, obtaining historical meteorological temperature and historical data of user-side load power, and obtaining a rule template set by the distribution network equipment based on the user-side load power; S200, training a time series model based on historical meteorological temperature and historical data, predicting the user-side load power after a certain period of time, and matching the rule template of the distribution network equipment according to the size of the load power prediction value; S300, monitoring the actual load power on the user side, analyzing the data anomalies between the actual value and the predicted value of the load power, adjusting the rule template of the distribution network equipment, and sending anomaly reminders; S400, displaying the actual value and predicted value of the load power, as well as the trend chart of the actual load power changes in real time on a visual screen.

[0006] As a preferred solution of the method for monitoring and analyzing anomalies of distribution network equipment based on a rule template according to the present invention, wherein: in S100, the historical meteorological temperature refers to the temperature value of the area where the user is located released by the meteorological bureau, the load power on the user side refers to the real-time power obtained by the user from the power supply system during the power consumption process, the historical data refers to the load power corresponding to each time point in each day of the user, and the rule template refers to a template for adjusting power distribution preset according to the user's power consumption requirements.

[0007] As a preferred solution of the method for monitoring and analyzing anomalies of distribution network equipment based on a rule template according to the present invention, wherein: training the time series model according to the historical meteorological temperature and historical data includes the following steps: dividing 24 hours in a day into n time periods at an interval of 30 minutes each, arranging the n time periods in chronological order, and naming them A1, A2,..., An time periods in sequence; within each time period, sampling the load power on the user side at an interval of one minute, and calculating the average load power within this time period; using the n time periods and the corresponding average load power data as the model training set; fitting and establishing a SARIMA prediction model using the model training set and the historical meteorological temperature data, and predicting the load power on the user side.

[0008] As a preferred solution of the method for monitoring and analyzing anomalies of distribution network equipment based on a rule template according to the present invention, wherein: the SARIMA model is a prediction model for analyzing time series data with trend, seasonality and autocorrelation by adding seasonal difference to the ARIMA model, and the ARIMA model is a prediction model for analyzing non-stationary time series data with trend and autocorrelation through autoregression, differencing and moving average processing.

[0009] As a preferred solution of the method for monitoring and analyzing anomalies of distribution network equipment based on a rule template according to the present invention, wherein: matching the rule template of the distribution network equipment according to the magnitude of the load power prediction value includes the following steps: setting power thresholds T1, T2, T3, where T1 < T2 < T3, dividing the load power into three power intervals [0, T1), [T1, T2), [T2, T3] according to the power thresholds, and each power interval corresponds to a rule template; taking every 1 hour as a special time point, calculating the load power prediction value Pnt of the next special time point, and judging whether Pnt belongs to the power interval corresponding to the current rule template; when Pnt does not belong to the above power interval, matching the corresponding rule template according to this prediction value, and preparing for pre-switching of the rule template; the pre-switching preparation refers to the equipment status check, parameter preloading and resource preallocation of the distribution network equipment before switching the rule template.

[0010] As a preferred solution of the distribution network equipment abnormality monitoring and analysis method based on rule templates described in the present invention, wherein: the analysis of the data abnormality of the load power forecast value includes the following steps: in the pre-switching preparation process, increasing the load power forecast frequency, calculating the abnormality threshold U at a certain time point i i , the actual load power value at time point i is X i , the corresponding predicted value is Y i , calculate the abnormal index V at time point i i , abnormal threshold U i The calculation formula is: U i =k×ln(X i +e), where e is a natural constant, k represents the threshold coefficient, and the abnormal index V i The calculation formula is:

[0011]

[0012] In the formula, m represents the correction coefficient. At the same time point, the anomaly threshold and the anomaly index are compared. When the anomaly index is greater than or equal to the anomaly threshold, the rule template is re-matched based on the predicted value at the next time point, and an anomaly reminder is issued.

[0013] As a preferred solution of the distribution network equipment abnormality monitoring and analysis system based on rule templates described in the present invention, it includes: a data acquisition module, a model training module, a rule template matching module, a rule template adjustment module and a visualization display module;

[0014] The data acquisition module is used to obtain historical meteorological temperature and user-side load power data, and obtain the rule template set by the distribution network equipment based on the user-side load power size;

[0015] The model training module is used to train a time series model based on historical meteorological temperature and historical data. The 24 hours of a day are evenly divided into n time periods with each interval being 30 minutes. The n time periods are arranged in chronological order and named as time periods A1, A2, ..., An. Within each time period, user-side load power data is sampled at one-minute intervals to calculate the average load power within the time period. The n time periods and the corresponding average load power data are used as a model training set. The SARIMA prediction model is established using the model training set and the historical meteorological temperature data for fitting, and user-side load power is predicted.

[0016] The rule template matching module is used to match the rule templates of distribution network equipment according to the magnitude of the load power prediction value, and set power thresholds T1, T2, and T3, where T1 < T2 < T3. The load power is divided into three power intervals [0, T1), [T1, T2), and [T2, T3] according to the power thresholds, and each power interval corresponds to a rule template; every 1 hour is used as a special time point, and the load power prediction value P of the next special time point is calculated. nt , and determine whether P nt belongs to the power interval corresponding to the current rule template; when P nt does not belong to the above power interval, match the corresponding rule template according to the prediction value, and prepare for the pre-switching of the rule template; the pre-switching preparation refers to the equipment status inspection, parameter preloading, and resource preallocation of the distribution network equipment before switching the rule template.

[0017] The rule template adjustment module is used to monitor the actual load power on the user side, analyze the data anomaly between the actual value and the prediction value of the load power, adjust the rule template of the distribution network equipment, and send an anomaly reminder.

[0018] The visualization display module is used to display the actual value and prediction value of the load power, as well as the change trend chart of the actual load power, in real time through a visualization screen.

[0019] Advantages of the present invention:

[0020] 1. By training and establishing a time series model to predict the load power on the user side, the distribution network equipment matches the corresponding rule templates and prepares for pre-switching, which improves the response efficiency of the distribution network equipment, reduces manual intervention, thereby enhancing the stability and efficiency of the system, monitors the anomaly situation in real time, optimizes and adjusts the rule templates, visualizes the change trend of the load power, improves the accuracy of load power prediction and rule template matching, and optimizes the power distribution.

[0021] [[ID=​​​​​​​​The present invention is a flow chart of a method for monitoring and analyzing abnormal changes in distribution network equipment based on rule templates. DETAILED DESCRIPTION

[0024] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0025] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from this description. The "embodiment" referred to here refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention.

[0026] Example 1

[0027] Reference Figure 1 This embodiment provides a distribution network equipment anomaly monitoring and analysis method based on a rule template, which specifically includes the following steps: S100, obtaining historical meteorological temperature and historical data of user-side load power, and obtaining a rule template set by the distribution network equipment based on the user-side load power; S200, training a time series model based on historical meteorological temperature and historical data, predicting the user-side load power after a certain period of time, and matching the rule template of the distribution network equipment according to the size of the load power prediction value; S300, monitoring the actual load power on the user side, analyzing the data anomaly between the actual value and the predicted value of the load power, adjusting the rule template of the distribution network equipment, and sending an anomaly reminder; S400, displaying the actual value and predicted value of the load power and the trend chart of the actual load power in real time on a visual screen.

[0028] In S100, the historical meteorological temperature refers to the temperature value of the user's area published by the Meteorological Bureau, the user-side load power refers to the real-time power obtained by the user from the power supply system during the power usage process, the historical data refers to the load power corresponding to each time point of the user every day, and the rule template refers to a template pre-set according to the user's power demand for adjusting power distribution.

[0029] The time series model is trained based on historical meteorological temperature and historical data, including the following steps: 1. Divide the 24 hours of a day into n time periods with each period of 30 minutes, and arrange the n time periods in chronological order, naming them A1, A2, ..., A n Time period; within each time period, user-side load power data is sampled at one-minute intervals, and the average load power within the time period is calculated; n time periods and the corresponding average load power data are used as the model training set; the model training set and historical meteorological temperature data are used to fit and establish a SARIMA prediction model, and user-side load power is predicted.

[0030] The SARIMA model is a prediction model that adds seasonal differencing to the ARIMA model and is used to analyze time series data with trends, seasonality, and autocorrelation. The ARIMA model is a prediction model that analyzes non-stationary time series data with trends and autocorrelation through autoregression, differencing, and moving average processing.

[0031] The rule template for matching distribution network equipment according to the magnitude of the load power prediction value includes the following steps: Set power thresholds T1, T2, and T3, where T1 < T2 < T3. Divide the load power into three power intervals [0, T1), [T1, T2), and [T2, T3] according to the power thresholds, and each power interval corresponds to a rule template; Take every 1 h as a special time point, and calculate the load power prediction value P of the next special time point nt , and judge whether P nt belongs to the power interval corresponding to the current rule template; When P nt does not belong to the above power interval, match the corresponding rule template according to this prediction value and perform the preparation for pre-switching of the rule template; The pre-switching preparation refers to the equipment status check, parameter preloading, and resource preallocation of the distribution network equipment before switching the rule template.

[0032] The purpose of the pre-switching preparation is to ensure a smooth transition when switching the rule template, reduce the impact on residential or industrial electricity consumption, improve the efficiency and reliability of the distribution network, and pre-adjust and prepare the operating parameters, control logic, or resource allocation of the distribution network equipment in advance, so as to ensure that when the actual load power reaches the corresponding interval, the equipment can quickly and smoothly switch to the target rule template and avoid delays, impacts, or control failures that may be caused by real-time switching.

[0033] Analyze the data anomaly situation of the load power prediction value, including the following steps: During the pre-switching preparation process, increase the load power prediction frequency, and calculate the anomaly threshold U at a certain time point i i , the actual value of the load power at time point i is X i and the corresponding prediction value is Y i , calculate the anomaly index V at time point i i , and the calculation formula of the anomaly threshold U i is: U i = k × ln(X i + e), where e is the natural constant and k represents the threshold coefficient. The calculation formula of the anomaly index V i is:

[0034]

[0035] In the formula, m represents the correction coefficient. At the same time point, the anomaly threshold and the anomaly index are compared. When the anomaly index is greater than or equal to the anomaly threshold, the rule template is re-matched based on the predicted value at the next time point, and an anomaly reminder is issued.

[0036] Historical meteorological temperature data is connected to the Meteorological Bureau interface in real time through the API, and regional temperature values ​​are automatically matched according to the user's geographic location (such as the substation area and substation coverage) to ensure data timeliness.

[0037] The seasonal difference period is set to 24 (corresponding to a 24-h daily period), and the training set uses a rolling forecast method to improve the real-time forecast accuracy.

[0038] Power thresholds (T1, T2, T3) support user-defined configuration. Industrial users can set stricter overload thresholds (such as T3 = 110% of rated power), while residential users focus on energy-saving templates (such as T1 = base load + 20%).

[0039] The pre-switching preparation process includes three checks: device communication status detection, backup power capacity verification, and adjacent device load balancing calculation to ensure uninterrupted template switching.

[0040] The trend chart uses a dual-axis coordinate system to support multi-dimensional correlation analysis (such as load surge warning during high temperature periods).

[0041] Example 2

[0042] This embodiment provides a distribution network equipment abnormality monitoring and analysis system based on rule templates, which specifically includes a data acquisition module, a model training module, a rule template matching module, a rule template adjustment module, and a visualization display module;

[0043] The data acquisition module is used to obtain historical meteorological temperature and user-side load power data, and obtain the rule template set by the distribution network equipment based on the user-side load power size;

[0044] The model training module is used to train a time series model based on historical meteorological temperature and historical data. The module divides a 24-hour day into n time periods, each 30 minutes long. These time periods are then arranged in chronological order and named A1, A2, ..., An. Within each time period, user-side load power data is sampled at one-minute intervals to calculate the average load power within that time period. The n time periods and the corresponding average load power data are used as the model training set. The SARIMA forecasting model is then built using the model training set and historical meteorological temperature data to predict user-side load power.

[0045] The rule template matching module is used to match the rule templates of distribution network devices according to the magnitude of the predicted load power value. Power thresholds T1, T2, and T3 are set, where T1 < T2 < T3. The load power is divided into three power intervals [0, T1), [T1, T2), and [T2, T3] according to the power thresholds, and each power interval corresponds to a rule template; every 1h is used as a special time point, and the predicted load power value Pnt for the next special time point is calculated, and it is judged whether Pnt belongs to the power interval corresponding to the current rule template; when Pnt does not belong to the above power interval, the corresponding rule template is matched according to this predicted value, and the preparation for pre-switching of the rule template is carried out; the preparation for pre-switching refers to the equipment status check, parameter pre-loading, and resource pre-allocation of the distribution network device before switching the rule template.

[0046] The rule template adjustment module is used to monitor the actual load power on the user side, analyze the data anomalies between the actual value and the predicted value of the load power, adjust the rule template of the distribution network device, and send anomaly alerts.

[0047] The visualization display module is used to display the actual value and the predicted value of the load power, as well as the change trend graph of the actual load power, in real time through a visualization screen.

[0048] Importantly, although only a few embodiments are described in detail in this disclosure, those who refer to this disclosure should easily understand that many modifications are possible without substantially departing from the subject matter described in this application. For example, the dimensions, structures, shapes, and proportions of various components, as well as temperature, pressure, installation arrangements, the use of materials, color, orientation changes, etc.; for example, a component shown as integrally formed can be composed of multiple parts or components, and the position of the component can be inverted or changed in other ways; therefore, all such modifications should be included within the scope of the present invention. Other substitutions, modifications, changes, and omissions can be made in the design, operating conditions, and arrangements of the exemplary embodiments without departing from the scope of the present invention.

Claims

1. A distribution network equipment abnormality monitoring and analysis method based on rule templates, characterized in that: It includes the following steps: S100. Obtain historical meteorological temperature and historical data of the user-side load power, and obtain a rule template set by the distribution network equipment based on the magnitude of the user-side load power; S200. Train a time series model according to the historical meteorological temperature and historical data to predict the user-side load power after a certain period of time, and match the rule template of the distribution network equipment according to the magnitude of the predicted load power value; S300. Monitor the actual load power on the user side, analyze the data anomaly between the actual value and the predicted value of the load power, adjust the rule template of the distribution network equipment, and send an anomaly reminder; S400. Real-time display the actual value and predicted value of the load power, as well as the change trend chart of the actual load power through a visualization screen.

2. The distribution network equipment abnormality monitoring and analysis method based on rule templates according to claim 1 is characterized in that: In S100, the historical meteorological temperature refers to the temperature value of the area where the user is located released by the meteorological bureau, the user-side load power refers to the real-time power obtained from the power supply system during the user's power consumption process, the historical data refers to the load power corresponding to each time point in each day of the user, and the rule template refers to a template preset according to the user's power consumption demand for adjusting power distribution.

3. The distribution network equipment abnormality monitoring and analysis method based on rule templates according to claim 1 is characterized in that: The training of the time series model according to the historical meteorological temperature and historical data includes the following steps: Divide the 24 hours of a day into n time periods with each period being 30 minutes. Arrange the n time periods in chronological order and name them A1, A2, ..., A n Time period; In each time period, sample the user-side load power data at intervals of one minute, and calculate the average load power in this time period; Use n time periods and the corresponding average load power data as the model training set; Use the model training set and historical meteorological temperature data to fit and establish a SARIMA prediction model, and predict the user-side load power.

4. The distribution network equipment abnormality monitoring and analysis method based on rule templates according to claim 3 is characterized in that: The SARIMA model is a prediction model that adds seasonal differences to the ARIMA model and is used to analyze time series data with trends, seasonality, and autocorrelation. The ARIMA model is a prediction model that analyzes non-stationary time series data with trends and autocorrelation through autoregression, differencing, and moving average processing.

5. The distribution network equipment abnormality monitoring and analysis method based on rule templates according to claim 1 is characterized in that: The matching of the rule template of the distribution network equipment according to the magnitude of the predicted load power value includes the following steps: Set power thresholds T1, T2, and T3, where T1 < T2 < T3, and divide the load power into three power intervals [0, T1), [T1, T2), and [T2, T3], and each power interval corresponds to a rule template; Each 1h interval is taken as a special time point, and the load power forecast value P at the next special time point is calculated. nt , judge P nt Whether it belongs to the power range corresponding to the current rule template; When P nt If the power range does not fall within the above range, the corresponding rule template is matched according to the predicted value, and pre-switching preparation of the rule template is performed; The pre-switching preparation refers to the equipment status check, parameter preloading, and resource preallocation of the distribution network equipment before switching the rule template.

6. The distribution network equipment abnormality monitoring and analysis method based on rule templates according to claim 5 is characterized in that: The analysis of the data anomaly of the predicted load power value includes the following steps: In the pre-switching preparation process, the load power forecast frequency is increased and the abnormal threshold U at a certain time point i is calculated. i , the actual load power value at time point i is X i , the corresponding predicted value is Y i , calculate the abnormal index V at time point i i , abnormal threshold U i The calculation formula is: U i =k×ln(X i +e), where e is a natural constant, k represents the threshold coefficient, and the abnormal index V i The calculation formula is: In the formula, m represents the correction coefficient. At the same moment, compare the anomaly threshold with the anomaly index. When the anomaly index is greater than or equal to the anomaly threshold, re-match the rule template according to the predicted value of the next moment and send an anomaly reminder.

7. A distribution network equipment abnormality monitoring and analysis system based on rule templates, using the distribution network equipment abnormality monitoring and analysis method based on rule templates as claimed in claim 1, characterized in that: It includes a data acquisition module, a model training module, a rule template matching module, a rule template adjustment module, and a visualization display module; The data acquisition module is used to obtain historical meteorological temperature and user-side load power data, and obtain the rule template set by the distribution network equipment based on the user-side load power size; The model training module is used to train a time series model based on historical meteorological temperature and historical data. The 24 hours of a day are evenly divided into n time periods with each interval being 30 minutes. The n time periods are arranged in chronological order and named as time periods A1, A2, ..., An. Within each time period, user-side load power data is sampled at one-minute intervals to calculate the average load power within the time period. The n time periods and the corresponding average load power data are used as a model training set. The SARIMA prediction model is established using the model training set and the historical meteorological temperature data for fitting, and user-side load power is predicted. The rule template matching module is used to match the rule templates of distribution network equipment according to the magnitude of the load power prediction value. Power thresholds T1, T2, and T3 are set, where T1 < T2 < T3. The load power is divided into three power intervals [0, T1), [T1, T2), and [T2, T3] according to the power thresholds, and each power interval corresponds to a rule template; every 1 h is used as a special time point, and the load power prediction value P of the next special time point is calculated nt , and it is judged whether P nt belongs to the power interval corresponding to the current rule template; when P nt does not belong to the above power interval, the corresponding rule template is matched according to this prediction value to prepare for the pre-switching of the rule template; the pre-switching preparation refers to the equipment status check, parameter preloading, and resource pre-allocation of the distribution network equipment before switching the rule template The rule template adjustment module is used to monitor the actual load power on the user side, analyze the data changes between the actual value and the predicted value of the load power, adjust the rule template of the distribution network equipment, and send a change reminder; The visualization display module is used to display the actual value and predicted value of the load power and the change trend diagram of the actual load power in real time through a visualization screen.

Images