Abnormal signal analysis and disposal method and system based on power grid alarm data

Abstract

The invention belongs to the technical field of power system automation, and discloses an abnormal signal analysis and disposal method and system based on power grid alarm data, and the method comprises the steps: completely collecting model data, operation data and alarm data of a D5000 system, and carrying out the data processing; according to the content, the type and the occurrence sequence of the alarm signals, abnormal signal analysis content is generated through an abnormal signal analysis method; unmatched alarm signal abnormity is predicted and handled, a data body is substituted into an expert database model, an abnormal signal handling method is studied and judged, and abnormal signal rapid handling is carried out in combination with edge calculation. The method has higher data processing speed and fault positioning accuracy, can effectively reduce the false alarm rate, improves the safe operation level of a power grid, improves the reliability, safety and operation efficiency of the power grid, and achieves the intelligent management of the power grid and the quick response of faults.

Description

Technical Field

[0001] The present invention relates to power system automation technology, and in particular to an abnormal signal analysis and processing method and system based on power grid alarm data. Background Art

[0002] With the continuous expansion of the scale of power grids and the improvement of their intelligence level, the safe and stable operation of power grid systems is of great significance to ensuring national energy security and economic and social development. As an important part of the power grid monitoring system, power grid alarm data plays an important role in timely discovering and handling power grid faults and ensuring the safe operation of the power grid. However, with the expansion of the scale of power grids and the increase of monitoring data, traditional power grid alarm data processing methods can no longer meet the operation needs of modern power grids.

[0003] In the past few decades, power grid monitoring systems and related technologies have developed rapidly. From the initial analog monitoring system to the modern digital monitoring system, the accuracy and efficiency of power grid monitoring have been significantly improved. In particular, with the development of technologies such as big data and artificial intelligence, the data processing capability and intelligence level of power grid monitoring systems have been continuously improved, providing a strong guarantee for the safe operation of the power grid. However, the existing power grid alarm data processing methods still have some shortcomings, such as slow data processing speed, inaccurate fault location, and high alarm false alarm rate.

[0004] Especially in the analysis of abnormal signals, most existing methods are based on fixed thresholds and rules, and cannot be adaptively adjusted according to the real-time changes in power grid operation, resulting in certain limitations in practical applications. In addition, most existing abnormal signal processing methods rely on centralized processing systems, resulting in slow data processing speed and serious network delays, which cannot meet the needs of rapid processing of power grid faults.

[0005] The analysis and processing of abnormal signals in the power grid alarm data currently involves statistics and analysis of the real-time abnormal alarm signals of the power grid and the processed abnormal alarm signals, and statistics of the abnormal alarm signals to be processed; the classification algorithm is used to analyze the abnormal alarm signals to be processed and the tripping record information of the fault recorder to screen out the key fault information; the "association" attribute in the typical abnormal alarm signals is manually processed, and the data analysis model is established in combination with the abnormal alarm information of the auxiliary control system and the inspection system information; combined with the historical fault cases of the power grid, fault handling plans, power grid stability regulations, and power grid protection regulations, the abnormal faults are accurately located and the handling strategies are given, so as to realize real-time analysis of the causes of typical signals, and generate relevant power grid events to be pushed to the monitors. Summary of the invention

[0006] In view of the above-mentioned existing problems, the present invention is aimed at the existing technology which mostly performs statistical analysis on abnormal alarm signals, but does not combine multiple power grid alarms to intelligently generate analysis and disposal solutions for abnormal signals. This patent provides a method for analyzing and disposing abnormal signals based on power grid alarm data. Through data collection and data processing, the alarm signal and model data are structured into a complete data body, which is then brought into the expert library model through the data body to obtain an abnormal signal disposal solution.

[0007] In order to solve the above technical problems, a method for analyzing and handling abnormal signals based on power grid alarm data is proposed, including:

[0008] The model data, operation data, and alarm data of the D5000 system are fully collected and processed; according to the content, type, and order of occurrence of the alarm signal, the abnormal signal analysis content is generated through the abnormal signal analysis method; the unmatched alarm signal anomalies are predicted and handled, and the data body is brought into the expert library model to determine the handling method of the abnormal signal, and the abnormal signal is quickly processed in combination with edge computing.

[0009] As a preferred solution of the abnormal signal analysis and disposal method based on power grid alarm data described in the present invention, wherein: the complete collection of model data, operation data, and alarm data of the D5000 system includes, the model data includes the substation location, type, voltage level, capacity, the rated current, impedance, capacitive reactance of the line, the type of interval, connected equipment, protection settings, and the equipment includes circuit breakers, switches, technical parameters of protection equipment, operating status, and historical maintenance records.

[0010] The operating data includes the telemetry data of the signal point, such as current, voltage, power, frequency, and power factor; the telesignaling data includes the equipment switch position, protection equipment action signal, fault signal, the status and result of the remote control command; and the environmental data, such as temperature, humidity, and wind speed.

[0011] The alarm data includes the recorded time, location, type, duration, affected equipment of the alarm events generated in the D5000 system, equipment overload, voltage anomaly, frequency fluctuation that deviates from normal operating parameters, warnings generated by the predictive maintenance system, and alarms generated by the D5000 system itself, such as communication interruption and hardware failure.

[0012] As a preferred solution of the abnormal signal analysis and disposal method based on power grid alarm data described in the present invention, the data processing includes collecting the data of the complete D5000 system and then filtering the interference signals that are not related to the alarm signal: comparing the collected data with the normal voltage range of 220V with a fluctuation of 10%, and treating the signals outside the fluctuation of 10% as interference signals and filtering them; for the filtered signals, analyzing the change trend: when the signal value has an upward trend over time and the signal value has a downward trend over time, it is not excluded and saved as an abnormal signal.

[0013] When the signal value fluctuates up and down in a short period of time, but there is no obvious upward or downward trend in the long term, this fluctuating signal is classified as an interference signal and subjected to secondary filtering.

[0014] For the signal after secondary filtering, check the correlation with the key signal: if the signal after secondary filtering is irrelevant to the key signal or the correlation is weak, it is judged as an irrelevant interference signal and final filtering is performed; if the signal after secondary filtering is strongly correlated with the key signal, it is judged as an abnormal signal and saved.

[0015] The signal point information of the alarm signal is associated with the model data through a data conversion method of multiple bit operations to generate: the data body of alarm signal + equipment + substation + interval:

[0016]

[0017] Among them, A(x) is the alarm signal after multiple bit operation conversions, A'(x) is the original alarm signal point information, B(x) is the conversion function of the bit operation including bitwise AND, bitwise OR, and bitwise XOR operations, n is the number of bit operations, and D(x) is the function of the model data that associates the bit operation result with the model data.

[0018] As a preferred solution of the abnormal signal analysis and disposal method based on power grid alarm data described in the present invention, the abnormal signal analysis method includes generating abnormal signal analysis content according to the alarm signal content, alarm signal type, and the order in which the alarm signal occurs through the abnormal signal analysis method:

[0019] Among them, F(x) is the comprehensive score of the abnormal signal, F(x)∈[0',10'], t0, t f is the start and end time of the time window for analysis, n is the number of alarm signals, ε i The weight factor of the ith alarm signal is dynamically adjusted according to the signal type and content, g i (x i ,t) the content and occurrence time of the ith alarm signal, hi (λ i (t)) is the order and historical data of the occurrence of the i-th warning signal, β is to ensure the comparability of the analysis results in different time windows, φ(x,t) is the difference between the current signal and the historical state, and θ(x,t) is a dynamic threshold adjustment function that adjusts according to the grid load change and seasonal factors.

[0020] According to the real-time data of grid operation and abnormal signal analysis, the warning threshold is adaptively adjusted and classified according to the type and severity of the warning: when F(x) ≤ 3', it is a type I emergency warning; when 3' < F(x) ≤ 7.5', it is a type II serious warning; when F(x) ≥ 7.5', it is a type III minor warning.

[0021] When the abnormal analysis is a type I emergency warning, continuously monitor the waveform analysis and trend prediction of the real-time data of the warning signal, automatically disconnect the fault area or switch to the standby system, the system starts the emergency response program to automatically conduct a secondary confirmation, generates an abnormal signal analysis report including the cause of the abnormality, lists the possible consequences in order from serious to minor, generates disposal steps and strategy guidelines for disposal requirements according to the consequences, and notifies the operator.

[0022] When the abnormal analysis is a type II serious warning, conduct a detailed analysis of the warning signal, including the comparison of historical data and real-time data, evaluate the impact of the warning on grid stability and equipment safety, and generate an abnormal signal analysis report.

[0023] When the abnormal analysis is a type III minor warning, track the development trend of the warning to judge whether it may be upgraded, and generate an abnormal signal analysis report.

[0024] As a preferred solution of the abnormal signal analysis and disposal method based on grid warning data described in the present invention, wherein: the bringing the data body into the expert library model includes splitting according to the keyword of the warning signal content, assigning different weight values to different keywords, combining the weight values of different keywords and the signal type weight to generate the warning signal weight value, and combining the signal weight value with the system expert library model to generate an abnormal signal disposal method: if the abnormal signal is a typical signal or has been entered in the expert library, the disposal strategy guidance is carried out according to the abnormal signal disposal method in the expert library model.

[0025] If the abnormal signal is an atypical signal and has not been entered in the expert library, predict the prediction result of the abnormal generation of the unmatched warning signal. If the prediction is correct, add the prediction result to the expert library model. If the prediction is incorrect, update the parameters until all the predicted values are calculated and can be entered into the expert library.

[0026] As a preferred solution of the abnormal signal analysis and disposal method based on power grid alarm data described in the present invention, the prediction and disposal of the abnormality of the unmatched alarm signal includes generating an alarm signal weight based on the alarm signal content, type and alarm data 10 minutes before and after the occurrence time, combining the alarm signal in the historical data and the signal 10 minutes before and after the alarm signal, and bringing the time and weight into the prediction model through the exponential smoothing algorithm, smoothing the time series data, capturing the trend and performing short-term prediction, and generating a predicted abnormal result: the initial smoothing value Y 1 , use the prediction model to recursively predict the value at the next time point:

[0027] Y t+1 =α·Y t +(1-α)·Y t

[0028] Among them, α is the smoothing parameter controlling the weight of the most recent observation, ranging from 0 to 1, and Y t+1 is the predicted value at time t+1, Y t is the actual observed value at time t, Y t is the predicted value at time t.

[0029] Update the smoothing parameter α according to the actual situation, repeat the recursion and update parameters of the prediction model until all the prediction values ​​are calculated, bring the prediction values ​​into the expert library model to match the expert library, and generate an abnormal signal analysis and disposal method.

[0030] As a preferred solution of the abnormal signal analysis and disposal method based on power grid alarm data described in the present invention, the rapid abnormal signal processing combined with edge computing includes deploying edge computing nodes on power grid equipment to collect power grid alarm data and equipment status data in real time and preprocessing the collected data to extract abnormal signal features:

[0031]

[0032] Among them, G(o,p,q) is, O i , P i , Q i is a complex information filtering function for feature parameters o, p, q, (o 2 +p 2 +q 2 ) 3 / 2 is to normalize the abnormal signals with different characteristic parameters, e -μ In order to evaluate the similarity between abnormal signals and normal signals, the fault type and fault location are determined according to the abnormal signal analysis and disposal method, and the suggestions are sent to the operation and maintenance personnel through the edge computing node for fault handling.

[0033] Another object of the present invention is to provide an abnormal signal analysis and processing system based on power grid alarm data. The present invention realizes intelligent analysis and processing of power grid alarm data through the collaborative work of the above-mentioned modules, improves the operating efficiency and safety of the power grid, reduces human intervention, improves the level of automation, and ensures the stable operation of the power grid system.

[0034] As a preferred solution of the abnormal signal analysis and disposal system based on power grid alarm data described in the present invention, it is characterized by including a data acquisition and processing module, an abnormal signal analysis module, an alarm signal storage module, and an abnormal signal disposal module.

[0035] The data acquisition and processing module collects the D5000 system model data, operation data, and alarm data on the power grid equipment in real time, pre-processes the collected data including filtering irrelevant interference signals, analyzing signal change trends, and performing correlation checks. Through a data conversion method of multiple bit operations, the signal point information of the alarm signal is associated with the model data to generate a comprehensive data body including alarm signal + equipment + substation + interval, cleans and prepares the data, and provides an accurate and comprehensive information basis for subsequent abnormal signal analysis.

[0036] The abnormal signal analysis module generates abnormal signal analysis content according to the content, type and occurrence sequence of the alarm signal through the abnormal signal analysis method, classifies the alarm signal, and generates an abnormal signal analysis report including the cause of the abnormality, possible consequences, handling steps and strategic guidance of handling requirements.

[0037] The alarm signal storage module enters the abnormal signal analysis results and handling methods into the expert library model for future query and reuse, predicts unmatched alarm signals, and adds correct prediction results to the expert library, updates the parameters in the expert library model, and improves the accuracy and adaptability of the prediction model.

[0038] The abnormal signal handling module provides handling strategy guidance to operators based on the abnormal signal handling methods in the expert library model, performs predictive analysis on atypical signals, and generates handling methods to quickly respond to unknown abnormal situations.

[0039] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the method described in the abnormal signal analysis and disposal based on power grid alarm data are implemented.

[0040] A computer-readable storage medium stores a computer program thereon, wherein when the computer program is executed by a processor, the steps of the method described in the abnormal signal analysis and disposal based on power grid alarm data are implemented.

[0041] Beneficial effects of the present invention: The present invention formats the textual alarm signal data into weight string data through the method of assigning weights to the alarm signal, brings the weight string data into the expert library model, matches the expert library model, uses the exponential smoothing algorithm, calculates the expert library matching missing data through the exponential smoothing algorithm, matches the expert library again, and if it matches, adds the expert library model, and if it fails to match, manually adds the expert library model. By combining the expert library with the exponential smoothing algorithm, after successful matching, the average accuracy rate is as high as 80% or more. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0043] Figure 1 An overall flow chart of an abnormal signal analysis and processing method based on power grid alarm data provided by one embodiment of the present invention.

[0044] Figure 2 An expert library processing strategy table of an abnormal signal analysis and processing method based on power grid alarm data is provided in one embodiment of the present invention.

[0045] Figure 3 An embodiment of the present invention provides an expert library processing strategy list for an abnormal signal analysis and processing method based on power grid alarm data.

[0046] Figure 4 An embodiment of the present invention provides a method for manually entering an expert database for an abnormal signal analysis and processing method based on power grid alarm data.

[0047] Figure 5 A diagram of abnormal signal handling content of a system for analyzing and handling abnormal signals based on power grid alarm data provided by an embodiment of the present invention.

[0048] Figure 6 An accuracy curve of an abnormal signal analysis and processing method based on power grid alarm data provided by an embodiment of the present invention.

[0049] Figure 7 An abnormal signal statistical diagram of an abnormal signal analysis and processing method based on power grid alarm data provided by an embodiment of the present invention.

[0050] Figure 8A system functional architecture diagram of an abnormal signal analysis and processing system based on power grid alarm data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0053] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is mutually exclusive with other embodiments, either individually or selectively.

[0054] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0055] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0056] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0057] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides an abnormal signal analysis and processing method based on power grid alarm data, comprising:

[0058] S1: Completely collect the model data, operation data, and alarm data of the D5000 system and perform data processing.

[0059] Furthermore, the model data, operation data, and alarm data of the D5000 system are fully collected: At this stage, the model data, operation data, and alarm data of the D5000 system are first comprehensively collected. The model data includes the location, type, voltage level, and capacity of the substation, the rated current, impedance, and capacitance of the line, the type of interval, the connected equipment, and the protection settings. The equipment includes the technical parameters, operating status, and historical maintenance records of circuit breakers, switches, and protection equipment. These data are a detailed description of the physical structure of the power grid, including the location, type, voltage level, and capacity of the substation, as well as the rated current, impedance, and capacitance of the line. These parameters are crucial to understanding the structure and operating characteristics of the power grid, and help to conduct accurate fault analysis and system optimization.

[0060] The operation data includes the telemetry data of the signal point, such as current, voltage, power, frequency, and power factor; the telesignaling data includes the equipment switch position, protection device action signal, fault signal, the status and result of the remote control command; and the environmental data, such as temperature, humidity, and wind speed. These data reflect the actual operation status of the power grid, including telemetry data (current, voltage, power, frequency, and power factor) and telesignaling data (equipment switch position, protection device action signal, and fault signal). Real-time monitoring of these data is crucial for timely detection of abnormalities and response to faults, and helps to improve the reliability and safety of the power grid.

[0061] The alarm data includes the recording time, location, type, duration, affected equipment of the alarm events generated in the D5000 system, equipment overload, voltage anomaly, frequency fluctuation that deviates from normal operating parameters, warnings generated by the predictive maintenance system, and alarm communication interruption and hardware failure generated by the D5000 system itself. These data include detailed information of the alarm events generated in the D5000 system, such as the recording time, location, type, duration, and affected equipment. These data are key information for power grid fault diagnosis and predictive maintenance, and play an important role in maintaining the stable operation of the power grid and preventing potential failures.

[0062] By collecting and processing the model data, operation data and alarm data of the power grid in real time, we can fully understand the status and trend of the power grid, and provide an accurate and comprehensive information basis for the subsequent abnormal signal analysis; by analyzing the content, type and occurrence sequence of the alarm signal, we can accurately identify the abnormal signal in the power grid and generate abnormal signal analysis content. This helps to timely discover and locate faults and improve the reliability and safety of the power grid; by predicting the abnormalities of unmatched alarm signals and using the expert library model to deal with them, we can predict and prevent potential faults in advance and reduce the occurrence and impact of faults.

[0063] It should be noted that after collecting the complete data of the D5000 system, the interference signals that are not related to the alarm signal are filtered out: the collected data is compared with the normal voltage range of 220V with a fluctuation of 10%, and the signals outside the fluctuation of 10% are regarded as interference signals and filtered. For the filtered signals, the change trend is analyzed: when the signal value has an upward trend over time, and the signal value has a downward trend over time, it is not excluded and saved as an abnormal signal.

[0064] When the signal value fluctuates up and down in a short period of time, but there is no obvious upward or downward trend in the long term, this fluctuating signal is classified as an interference signal and subjected to secondary filtering.

[0065] For the signal after secondary filtering, check the correlation with the key signal: if the signal after secondary filtering is irrelevant to the key signal or the correlation is weak, it is judged as an irrelevant interference signal and final filtering is performed; if the signal after secondary filtering is strongly correlated with the key signal, it is judged as an abnormal signal and saved.

[0066] The signal point information of the alarm signal is associated with the model data through a data conversion method of multiple bit operations to generate: the data body of alarm signal + equipment + substation + interval:

[0067] Device id = signal point id of alarm signal||18446462603027808255.

[0068] Substation id:

[0069] id=to_number("||substr(cast(TO_CHAR(id,'FMXXXXXXXXXXXXXXXX')asvarchar2(18)),1,3)||'0000'||substr(cast(TO_CHAR(id,'FMXXXXXXXXXXXXXXXX X')asvarchar2(18)),8,8),'FMXXXXXXXXXXXXXXXXXX').

[0070] Interval id = substation id & alarm signal id.

[0071] Data body = alarm signal id + device id + substation id + interval id.

[0072]

[0073] Among them, A(x) is the alarm signal after multiple bitwise operations conversion, A'(x) is the original alarm signal point information, B(x) is the conversion function of bitwise operations including AND, OR, and XOR operations, n is the number of bitwise operations, and D(x) is the function of model data that associates the bitwise operation result with the model data.

[0074] S2: According to the content, type, and occurrence sequence of the alarm signal, generate the analysis content of the abnormal signal through the abnormal signal analysis method.

[0075] Furthermore, according to the content, type, and occurrence sequence of the alarm signal, generate the analysis content of the abnormal signal through the abnormal signal analysis method:

[0076]

[0077] Among them, F(x) is the comprehensive score of the abnormal signal, F(x) ∈ [0', 10'], t0, t f are the start and end times of the analysis time window, n is the number of alarm signals, ε i is the weight factor of the i-th alarm signal dynamically adjusted according to the signal type and content, g i (x i , t) is the content and occurrence time of the i-th alarm signal, h i (λ i (t)) is the occurrence sequence and historical data of the i-th alarm signal, β is to ensure the comparability of the analysis results of different time windows, φ(x, t) is the difference between the current signal and the historical state, and θ(x, t) is the dynamic threshold adjustment function adjusted according to the grid load change and seasonal factors. This score is comprehensively calculated based on factors such as the content, type, and occurrence sequence of the alarm signal, and is used to evaluate the severity and influence range of the abnormal signal.

[0078] According to the real-time data of power grid operation and the analysis of abnormal signals, adaptively adjust the alarm threshold, and classify according to the type and severity of the alarm: when F(x) ≤ 3' it is a type I emergency alarm, when 3' < F(x) ≤ 7.5’ it is a type II serious alarm, and when F(x) ≥ 7.5’ it is a type III minor alarm.

[0079] When the abnormality analysis is a Class I emergency alarm, the waveform analysis and trend prediction of the real-time data of the alarm signal are continuously monitored, the faulty area is automatically disconnected or switched to the backup system, and the system starts the emergency response procedure to automatically perform a second confirmation, and generates an abnormal signal analysis report including the cause of the abnormality, the possible consequences listed in order from severe to minor, and the strategic guidance of the disposal steps and disposal requirements based on the consequences, and notifies the operator.

[0080] When the abnormal analysis is a Class II severe alarm, a detailed analysis of the alarm signal is performed, including comparison of historical data and real-time data, to evaluate the impact of the alarm on grid stability and equipment safety, and generate an abnormal signal analysis report.

[0081] When the abnormal analysis is a Class III minor alarm, track the development trend of the alarm to determine whether it is likely to be upgraded and generate an abnormal signal analysis report.

[0082] By comprehensively considering factors such as the content, type, and order of occurrence of alarm signals, the comprehensive score of abnormal signals is calculated, which helps to accurately assess the severity and impact range of abnormal signals; by setting the analysis time window, the alarm signals in different time windows can be compared to better understand the development trend and impact range of abnormal signals; by dynamically adjusting the weight factor according to the signal type and content, the impact of different alarm signals on abnormal analysis can be more accurately measured, which helps to improve the accuracy of abnormal analysis; by dynamically adjusting the alarm threshold, it can better adapt to the impact of power grid load changes and seasonal factors, and improve the accuracy and practicality of abnormal analysis; classifying alarm signals according to type and severity helps to take targeted processing measures and improve the efficiency and effectiveness of fault handling.

[0083] S3: Predict and handle unmatched alarm signal anomalies, bring the data body into the expert library model, determine the handling method of abnormal signals, and combine edge computing to quickly handle abnormal signals.

[0084] Furthermore, according to the keyword splitting of the alarm signal content, different keywords are assigned different weights, and the weights of different keywords are combined with the signal type weight to generate the alarm signal weight. The signal weight is combined with the system expert library model to generate the abnormal signal handling method: If the abnormal signal is a typical signal or has been entered in the expert library, the handling strategy guidance is provided according to the abnormal signal handling method in the expert library model. The alarm signal content is split by keywords to extract key information for subsequent analysis. By assigning different weights to different keywords and combining the signal type weight, the weight of the alarm signal is generated to reflect the severity and impact range of the alarm signal.

[0085] The alarm signal keyword split weights are shown in the following table:

[0086] Table 1

[0087] Signal Description Weight Main and secondary protection reclosing action 5 Protection device PT disconnection alarm 3 Overvoltage protection 5 Zero sequence overcurrent protection 5 Circuit breaker distance protection action 5 Longitudinal differential protection action 5 Circuit breaker spring is not charged 2 Circuit breaker open 5

[0088] If the abnormal signal is atypical and has not been recorded in the expert database, the unmatched alarm signal abnormality is predicted to generate a prediction result. If the prediction is correct, the prediction result is added to the expert database model. If the prediction is wrong, the parameters are updated until all the prediction values ​​are calculated and can be entered into the expert database.

[0089] It should be noted that the alarm signal weight is generated based on the alarm signal content, type and alarm data 10 minutes before and after the occurrence time, combined with the alarm signal in the historical data and the signal 10 minutes before and after the alarm signal. The time and weight are substituted into the prediction model through the exponential smoothing algorithm to smooth the time series data, capture the trend and make short-term predictions, and generate a prediction abnormality result: the initial smoothing value Y 1 , use the prediction model to recursively predict the value at the next time point:

[0090] Y t+1 =α·Y t +(1-α)·Y t

[0091] Among them, α is the smoothing parameter controlling the weight of the most recent observation, ranging from 0 to 1, and Y t+1 is the predicted value at time t+1, Y t is the actual observed value at time t, Y t is the predicted value at time t.

[0092] Update the smoothing parameter α according to the actual situation, repeat the recursion and update parameters of the prediction model until all the prediction values ​​are calculated, bring the prediction values ​​into the expert library model to match the expert library, and generate an abnormal signal analysis and disposal method.

[0093] By splitting keywords, key information in the alarm signal can be extracted more accurately, which is helpful for subsequent analysis and processing; by assigning different weights to different keywords and combining the signal type weights to generate the weights of the alarm signal, the severity and scope of the alarm signal can be more accurately reflected, which is helpful for subsequent analysis and processing; by matching the generated alarm signal weights with the expert library model in the system, the corresponding abnormal signal handling method can be quickly found to improve processing efficiency; through the exponential smoothing algorithm, combined with the alarm signal content, type and alarm data 10 minutes before and after the occurrence time, the unmatched alarm signal anomaly can be predicted, which is helpful for discovering potential faults; by updating the prediction results, the expert library model can be continuously optimized to improve the accuracy and practicality of the prediction.

[0094] In summary, the accuracy of abnormal signal analysis of power grid can be improved, rapid response and effective disposal of power grid faults can be achieved, the stable operation of the power grid can be ensured, and the reliability of power supply can be improved. This is of great significance for ensuring the safety, stability and efficiency of power supply.

[0095] It should also be noted that edge computing nodes are deployed on power grid equipment to collect power grid alarm data and equipment status data in real time and pre-process the collected data to extract abnormal signal features:

[0096]

[0097] Among them, G(o,p,q) is, O i , P i , Q i is a complex information filtering function for feature parameters o, p, q, (o 2 +p 2 +q 2 ) 3 / 2 is to normalize the abnormal signals of different characteristic parameters, e -μ In order to evaluate the similarity between abnormal signals and normal signals, the fault type and fault location are determined according to the abnormal signal analysis and disposal method, and the suggestions are sent to the operation and maintenance personnel through the edge computing node for fault handling.

[0098] Through the collaborative work of edge computing nodes and central servers, some abnormal signal analysis and processing tasks are distributed on power grid equipment, realizing rapid processing of abnormal signals and fault location, reducing network latency, and improving fault handling efficiency. Through the collaborative work of edge computing nodes and central servers, the fault handling efficiency is improved, network latency is reduced, and local algorithms are combined with big data analysis of central servers to improve the accuracy and comprehensiveness of fault analysis.

[0099] Example 2, reference Figure 2-Figure 7 , which is an embodiment of the present invention, provides an abnormal signal analysis and disposal method based on power grid alarm data. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0100] Figures 2 to 7 The operation process is performed according to an abnormal signal analysis and disposal method based on power grid alarm data, wherein Figure 5 A disposal report on the abnormal signal analysis and disposal methods of the power grid alarm data during operation.

[0101] Table 2

[0102]

[0103] From the above table data, we can clearly see that the abnormal signal detection time and processing time of system B (the system using the method of the present invention) are significantly lower than those of system A (the system using the prior art). Specifically, the abnormal signal detection time of system B is 50 seconds faster than that of system A, and the processing time is 100 seconds faster. This shows that the method of the present invention has a significant advantage in processing speed.

[0104] In addition, the prediction accuracy of system B is also better than that of system A, reaching 90%, which is 20% higher than that of system A. This shows that the method of the present invention not only has an advantage in processing speed, but also has a significant improvement in prediction accuracy.

[0105] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0106] Embodiment 3, the third embodiment of the present invention, is different from the first two embodiments in that:

[0107] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0108] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0109] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0110] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0111] Example 4, reference Figure 8 , which is the fourth embodiment of the present invention, provides an abnormal signal analysis and disposal system based on power grid alarm data, including a data acquisition and processing module, an abnormal signal analysis module, an alarm signal storage module, and an abnormal signal disposal module.

[0112] The data acquisition and processing module collects the D5000 system model data, operation data, and alarm data on the power grid equipment in real time, and pre-processes the collected data, including filtering irrelevant interference signals, analyzing signal change trends, and performing correlation checks. Through the data conversion method of multiple bit operations, the signal point information of the alarm signal is associated with the model data to generate a comprehensive data body containing alarm signal + equipment + substation + interval, clean and prepare the data, and provide an accurate and comprehensive information basis for subsequent abnormal signal analysis.

[0113] The abnormal signal analysis module generates abnormal signal analysis content according to the content, type, and occurrence sequence of the alarm signal through the abnormal signal analysis method, classifies the alarm signal, and generates an abnormal signal analysis report including the cause of the abnormality, possible consequences, handling steps, and strategic guidance on handling requirements.

[0114] The alarm signal storage module enters the abnormal signal analysis results and disposal methods into the expert library model for future query and reuse, predicts unmatched alarm signals, and adds correct prediction results to the expert library, updates the parameters in the expert library model, and improves the accuracy and adaptability of the prediction model.

[0115] The abnormal signal handling module provides handling strategy guidance to operators based on the abnormal signal handling methods in the expert library model, performs predictive analysis on atypical signals, and generates handling methods to quickly respond to unknown abnormal situations.

[0116] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for analyzing and handling abnormal signals based on power grid alarm data, characterized in that: include, Completely collect the model data, operation data, and alarm data of the D5000 system and process the data; According to the content, type and occurrence sequence of the alarm signal, the abnormal signal analysis content is generated through the abnormal signal analysis method; Predict unmatched alarm signal anomalies and handle them, bring the data body into the expert library model, determine the handling method of abnormal signals, and combine edge computing to quickly handle abnormal signals.

2. The abnormal signal analysis and processing method based on power grid alarm data according to claim 1, characterized in that: The complete collection of model data, operation data, and alarm data of the D5000 system includes: The model data includes the substation location, type, voltage level, capacity, line rated current, impedance, capacitive reactance, bay type, connected equipment, protection settings, equipment including circuit breakers, switches, technical parameters of protection equipment, operating status, and historical maintenance records; The operation data includes the telemetry data of the signal point, such as current, voltage, power, frequency, and power factor; the telemetering data includes the equipment switch position, protection equipment action signal, fault signal, the status and result of the remote control command; and the environmental data, such as temperature, humidity, and wind speed; The alarm data includes the recorded time, location, type, duration, affected equipment of the alarm events generated in the D5000 system, equipment overload, voltage anomaly, frequency fluctuation that deviates from normal operating parameters, warnings generated by the predictive maintenance system, and alarms generated by the D5000 system itself, such as communication interruption and hardware failure.

3. The abnormal signal analysis and processing method based on power grid alarm data according to claim 2 is characterized in that: The data processing includes collecting the complete data of the D5000 system and filtering the interference signals that are not related to the alarm signal: comparing the collected data with the normal voltage range of 220V with a fluctuation of 10%, treating the signals outside the fluctuation of 10% as interference signals and filtering them, and analyzing the change trend of the filtered signals: when the signal value has an upward trend over time, and the signal value has a downward trend over time, it is not excluded and saved as an abnormal signal; When the signal value fluctuates up and down in a short period of time, but there is no obvious upward or downward trend in the long term, this fluctuation signal is classified as an interference signal and filtered twice; For the signal after secondary filtering, check the correlation between it and the key signal: if the signal after secondary filtering is irrelevant to the key signal or the correlation is weak, it is determined to be an irrelevant interference signal and final filtering is performed; if the signal after secondary filtering is strongly correlated with the key signal, it is determined to be an abnormal signal and saved; The signal point information of the alarm signal is associated with the model data through a data conversion method of multiple bit operations to generate: the data body of alarm signal + equipment + substation + interval: Among them, A(x) is the alarm signal after multiple bit operation conversions, A'(x) is the original alarm signal point information, B(x) is the conversion function of the bit operation including bitwise AND, bitwise OR, and bitwise XOR operations, n is the number of bit operations, and D(x) is the function of the model data that associates the bit operation result with the model data.

4. The abnormal signal analysis and processing method based on power grid alarm data according to claim 3 is characterized in that: The abnormal signal analysis method comprises generating abnormal signal analysis content according to the alarm signal content, the alarm signal type, and the order in which the alarm signals occur, through the abnormal signal analysis method: Among them, F(x) is the comprehensive score of the abnormal signal, F(x)∈[0',10'], t0, t f is the start and end time of the time window for analysis, n is the number of alarm signals, ε i The weight factor of the ith alarm signal is dynamically adjusted according to the signal type and content, g i (x i ,t) the content and occurrence time of the ith alarm signal, h i (λ i (t)) is the order and historical data of the occurrence of the i-th alarm signal, β is to ensure that the analysis results of different time windows are comparable, φ(x,t) is the difference between the current signal and the historical state, and θ(x,t) is the dynamic threshold adjustment function adjusted according to the grid load changes and seasonal factors; According to the analysis of real-time data and abnormal signals in power grid operation, the alarm threshold is adaptively adjusted and classified according to the type and severity of the alarm: when F(x) ≤ 3', it is Class I emergency alarm; when 3' < F(x) ≤ 7.5', it is Class II serious alarm; when F(x) ≥ 7.5', it is Class III minor alarm; When the abnormal analysis is a Class I emergency alarm, continuously monitor the waveform analysis and trend prediction of the real-time data of the alarm signal, automatically disconnect the fault area or switch to the standby system, the system starts the emergency response program to automatically conduct a secondary confirmation, generates an abnormal signal analysis report including the cause of the abnormality, lists the possible consequences in order from serious to minor, generates disposal steps and strategy guidelines for disposal requirements according to the consequences, and notifies the operator; When the abnormal analysis is a Class II serious alarm, conduct a detailed analysis of the alarm signal, including the comparison of historical data and real-time data, evaluate the impact of the alarm on the power grid stability and equipment safety, and generate an abnormal signal analysis report; When the abnormal analysis is a Class III minor alarm, track the development trend of the alarm to judge whether it may be upgraded, and generate an abnormal signal analysis report.

5. The abnormal signal analysis and processing method based on power grid alarm data according to claim 4 is characterized in that: The above-mentioned bringing the data body into the expert library model includes splitting according to the keyword of the alarm signal content, assigning different weights to different keywords, combining the weights of different keywords and the signal type weight to generate the alarm signal weight, and combining the signal weight with the system expert library model to generate the abnormal signal disposal method: If the abnormal signal is a typical signal or has been entered into the expert library, the disposal strategy is guided according to the abnormal signal disposal method in the expert library model; If the abnormal signal is an atypical signal and has not been entered into the expert library, predict the prediction result of the abnormal generation of the unmatched alarm signal. If the prediction is correct, add the prediction result to the expert library model. If the prediction is incorrect, update the parameters until all prediction values are calculated and can be entered into the expert library.

6. The abnormal signal analysis and processing method based on power grid alarm data according to claim 5, characterized in that: The prediction and handling of the unmatched alarm signal anomaly includes generating an alarm signal weight based on the alarm signal content, type and alarm data 10 minutes before and after the occurrence time, combining the alarm signal in the historical data and the signal 10 minutes before and after the alarm signal, and bringing the time and weight into the prediction model through the exponential smoothing algorithm, smoothing the time series data, capturing the trend and making a short-term prediction, and generating a prediction anomaly result: the initial smoothing value Y 1 , use the prediction model to recursively infer the predicted value at the next time point: AND t+1 =α·Y t +(1-α)·Y t Among them, α is the smoothing parameter controlling the weight of the most recent observation, ranging from 0 to 1, and Y t+1 is the predicted value at time t+1, Y t is the actual observed value at time t, Y t is the predicted value at time t; Update the smoothing parameter α according to the actual situation, repeat the recursion of the prediction model and update the parameters until all prediction values are calculated, and bring the prediction values into the expert library model to match the expert library to generate the abnormal signal analysis and disposal method.

7. The abnormal signal analysis and processing method based on power grid alarm data according to claim 6, characterized in that: The above-mentioned combining edge computing for rapid processing of abnormal signals includes deploying edge computing nodes on power grid equipment to collect power grid alarm data and equipment status data in real time and preprocess the collected data to extract abnormal signal features: Among them, G(o,p,q) is, O i , P i , Q i is a complex information filtering function for feature parameters o, p, q, (o 2 +p 2 +q 2 ) 3 / 2 In order to normalize the abnormal signals with different characteristic parameters, e -μ In order to evaluate the similarity between abnormal signals and normal signals, the fault type and fault location are determined according to the abnormal signal analysis and disposal method, and the suggestions are sent to the operation and maintenance personnel through the edge computing node for fault handling.

8. A system using the abnormal signal analysis and processing method based on power grid alarm data as claimed in any one of claims 1 to 7, characterized in that: It includes a data collection and processing module, an abnormal signal analysis module, an alarm signal storage module, and an abnormal signal disposal module; The data collection and processing module collects the D5000 system model data, operation data, and alarm data on power grid equipment in real time, preprocesses the collected data, including filtering out irrelevant interference signals, analyzing the signal change trend, and conducting correlation checks. Through the data conversion method of multiple bit operations, associate the signal point information of the alarm signal with the model data to generate a comprehensive data body containing alarm signal + equipment + substation + interval, clean and prepare the data, and provide an accurate and comprehensive information basis for subsequent abnormal signal analysis; The abnormal signal analysis module generates abnormal signal analysis content according to the content, type and occurrence sequence of the alarm signal through the abnormal signal analysis method, classifies the alarm signal, and generates an abnormal signal analysis report including the cause of the abnormality, possible consequences, handling steps and strategic guidance of handling requirements; The alarm signal storage module enters the abnormal signal analysis results and treatment methods into the expert database model for future query and reuse, predicts the unmatched alarm signals, and adds the correct prediction results into the expert database, updates the parameters in the expert database model, and improves the accuracy and adaptability of the prediction model; The abnormal signal handling module provides handling strategy guidance to operators based on the abnormal signal handling methods in the expert library model, performs predictive analysis on atypical signals, and generates handling methods to quickly respond to unknown abnormal situations.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the abnormal signal analysis and processing method based on power grid alarm data described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of an abnormal signal analysis and processing method based on power grid alarm data described in any one of claims 1 to 7 are implemented.

Images