Electric meter management method and device, computer equipment and readable storage medium

Abstract

The invention relates to an electric meter management method and device, computer equipment and a readable storage medium. The method comprises the following steps: acquiring operation data of a target ammeter and environment data of the target ammeter in real time; analyzing the current operation data, the current environment data, historical operation data before the current operation data and corresponding historical environment data by using a preset fault recognition model, and determining a current fault recognition result and a prediction recognition result of the target electric meter; and outputting a fault processing strategy of the target electric meter based on the current fault identification result and the prediction identification result. By adopting the method, a more accurate fault detection result can be obtained, so that the fault detection and repair difficulty and the maintenance cost are reduced.

Description

Technical Field

[0001] The present application relates to the technical field of fault detection, and particularly to a method and device for meter management, a computer device, and a readable storage medium. Background Art

[0002] With the gradual development of the power system, electric meters, as important tools for power metering and monitoring, have been widely used in various electricity-consuming places such as residential areas, commercial areas, and industrial facilities.

[0003] During the operation of an electric meter, faults may occur. In the related art, mainly relying on manual experience or simple status detection to determine whether there are faults during the operation of the electric meter.

[0004] However, the fault detection results of the related art methods are inaccurate, increasing the difficulty and repair cost of fault detection and repair. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a method and device for meter management, a computer device, and a readable storage medium, which can obtain more accurate fault detection results, thereby reducing the difficulty and repair cost of fault detection and repair.

[0006] In a first aspect, the present application provides a method for meter management, including:

[0007] Real-time collecting the operation data of the target electric meter and the environmental data where the target electric meter is located;

[0008] Using a preset fault recognition model to analyze the current operation data, the current environmental data, the historical operation data before the current operation data, and the corresponding historical environmental data to determine the current fault recognition result and the predicted recognition result of the target electric meter;

[0009] Based on the current fault recognition result and the predicted recognition result, outputting a fault handling strategy for the target electric meter.

[0010] In one of the embodiments, using a preset fault recognition model to analyze the current operation data, the current environmental data, the historical operation data before the current operation data, and the corresponding historical environmental data to determine the current fault recognition result and the predicted recognition result of the target electric meter includes:

[0011] Extracting the current data feature information of the current operation data, the current environmental feature information of the current environmental data, the historical data feature information of the historical operation data, and the historical environmental feature information of the historical environmental data;

[0012] Determine the current fault identification result of the target electric meter based on the current data feature information and the current environmental feature information; and determine the predicted identification result of the target electric meter based on the current data feature information, historical data feature information, current environmental feature information, and historical environmental feature information.

[0013] In one embodiment, determining the current fault identification result of the target electric meter based on the current data feature information and the current environmental feature information includes:

[0014] Analyze the fault type, faulty component, module to which the component belongs, and fault cause of the target electric meter based on the current data feature information and the current environmental feature information;

[0015] Take the fault type, faulty component, module to which the component belongs, and fault cause of the target electric meter as the current fault identification result of the target electric meter.

[0016] In one embodiment, determining the predicted identification result of the target electric meter based on the current data feature information, historical data feature information, current environmental feature information, and historical environmental feature information includes:

[0017] Fuse the current data feature information and the historical data feature information to obtain the data trend change during the operation of the target electric meter;

[0018] Fuse the current environmental feature information and the historical environmental feature information to obtain the environmental trend change during the operation of the target electric meter;

[0019] Determine the predicted identification result of the target electric meter at a future preset time point based on the data trend change and the environmental trend change.

[0020] In one embodiment, determining the predicted identification result of the target electric meter at a future preset time point based on the data trend change and the environmental trend change includes:

[0021] Obtain the data prediction result of the target electric meter at a future preset time point based on the data trend change; and determine the environmental prediction result of the target electric meter at a future preset time point based on the environmental trend change;

[0022] Adjust the data prediction result using the environmental prediction result to obtain the predicted identification result of the target electric meter at a future preset time point.

[0023] In one embodiment, the method further includes:

[0024] Remove the outliers from the operation data of the target electric meter and the environmental data where the target electric meter is located;

[0025] Perform data filling on the removed operation data and environmental data;

[0026] After standardizing the filled operation data and environment data, perform the step of analyzing the current operation data, current environment data, historical operation data before the current operation data, and corresponding historical environment data by using a preset fault identification model.

[0027] In one embodiment, the training process of the fault identification model includes:

[0028] Obtain sample operation data, sample environment data, reference fault identification results, and reference prediction identification results;

[0029] Input the sample operation data and sample environment data into the initial fault identification model, and use the initial fault identification model for analysis to obtain sample fault identification results and sample prediction identification results;

[0030] Based on the identification difference between the sample fault identification result and the reference fault identification result, and the prediction difference between the sample prediction identification result and the reference prediction identification result, adjust the parameter information of the initial fault identification model until the sample fault identification result and sample prediction identification result output by the adjusted initial fault identification model meet the preset conditions, and obtain the trained fault identification model.

[0031] In a second aspect, the present application further provides an electric meter management device, including:

[0032] An acquisition module, configured to acquire the operation data of the target electric meter and the environment data where the target electric meter is located in real time;

[0033] An analysis module, configured to analyze the current operation data, current environment data, historical operation data before the current operation data, and corresponding historical environment data by using a preset fault identification model, and determine the current fault identification result and prediction identification result of the target electric meter;

[0034] An output module, configured to output a fault handling strategy for the target electric meter based on the fault identification result and the prediction identification result.

[0035] In a third aspect, the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, it implements the content of any one of the electric meter management methods in the first aspect above.

[0036] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the content of any one of the electric meter management methods in the first aspect above.

[0037] In a fifth aspect, the present application further provides a computer program product, including a computer program which, when executed by a processor, implements the content of any one of the electricity meter management methods in the first aspect above.

[0038] The above electricity meter management method, device, computer device, and readable storage medium collect the operation data of the target electricity meter and the environmental data where the target electricity meter is located in real time; analyze the current operation data, current environmental data, historical operation data before the current operation data, and corresponding historical environmental data by using a preset fault identification model to determine the current fault identification result and prediction identification result of the target electricity meter; and output a fault handling strategy for the target electricity meter based on the current fault identification result and prediction identification result. By collecting operation data and environmental data in real time and comprehensively analyzing the current and historical operation and environmental data by using a preset fault identification model, this method can accurately determine the current fault identification result of the target electricity meter, and can deeply identify latent faults to identify subtle damages to internal components of the electricity meter, poor contact of lines, and other hidden faults. This makes the fault detection process more meticulous, can obtain more accurate fault detection results, thereby reducing the difficulty and repair cost of fault detection and repair. At the same time, by using the fault identification model, the faults of the electricity meter can be detected in real time, avoiding the limitations of manual inspection and experience judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0040] Figure 1 It is an application environment diagram of the electricity meter management method in an embodiment;

[0041] Figure 2 It is a flowchart of the electricity meter management method in an embodiment;

[0042] Figure 3 It is a flowchart of the electricity meter management method in an embodiment;

[0043] Figure 4 It is a flowchart of the electricity meter management method in an embodiment;

[0044] Figure 5 It is a flowchart of the electricity meter management method in an embodiment;

[0045] Figure 6 It is a flowchart of the electricity meter management method in an embodiment;

[0046] Figure 7 is a schematic flowchart of an electricity meter management method in an embodiment;

[0047] Figure 8 is a schematic flowchart of an electricity meter management method in an embodiment;

[0048] Figure 9 is a schematic flowchart of an electricity meter management method in an embodiment;

[0049] Figure 10 is a structural block diagram of an electricity meter management device in an embodiment. Detailed implementation manners

[0050] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0051] Before introducing the technical solutions of the present application in detail, the background technology of the present application will be briefly introduced.

[0052] With the gradual development of the power system, the electricity meter, as an important tool for power metering and monitoring, has been widely used in various electricity-consuming places such as residential areas, commercial areas and industrial facilities. The current smart electricity meters have functions such as remote meter reading, remote control and load monitoring.

[0053] However, although the smart electricity meter has many advantages, in actual applications, the fault diagnosis and judgment of the electricity meter are still bottlenecks at present. For example, the faults of the electricity meter can include sensor faults, data anomalies and communication problems, etc. The existing electricity meter fault detection mostly relies on manual experience or simple status detection to determine whether there are faults during the operation of the electricity meter.

[0054] However, the existing electricity meter fault detection relies on manual inspection and regular maintenance. For complex faults, a large amount of data often needs to be manually analyzed, making the fault detection process very time-consuming and resulting in low fault detection efficiency. At the same time, electricity meter faults often involve multiple components (for example, sensors, display modules and communication interfaces, etc.). The existing detection methods usually can only diagnose through a single monitoring parameter. The single electricity meter data cannot reflect more comprehensive fault information, lacks linkage with other power equipment or power grid systems, and cannot accurately locate the faulty components, which leads to many faults not being located in time, thus increasing the difficulty and maintenance cost of fault detection and repair. In addition, the existing methods mostly rely on historical data and rule judgment and cannot intelligently predict the occurrence of faults based on real-time data. And the faults of the electricity meter often accumulate gradually. If potential problems cannot be discovered in time, it may lead to serious faults and long outage times.

[0055] With the maturity of big data and artificial intelligence technologies, intelligent means have been introduced in many fields to improve the efficiency and accuracy of fault detection. The electricity meter industry has also started to apply these technologies to the field of electricity meter component fault detection.

[0056] However, there are still many problems with the intelligent fault detection method. For example, it lacks real-time monitoring and prediction functions, cannot dynamically evaluate the status of the electricity meter based on real-time data, and most existing systems rely on preset rules or manually entered data, lacking the ability of self-learning and automated optimization.

[0057] In view of the above problems, the present application provides an electricity meter management method, device, computer device, and readable storage medium, which can obtain more accurate fault detection results, thereby reducing the difficulty and maintenance cost of fault detection and repair. Of course, the technical solutions provided in the embodiments of the present application are not limited to only solving the above problems, and there are other technical effects. For specific details, please refer to the following embodiments. Next, the technical solutions of the present application will be introduced in detail. Next, the specific content of the method will be introduced in detail through specific embodiments.

[0058] The electricity meter management method provided in the embodiments of the present application can be applied to an application environment as Figure 1 shown. For example, the computer device can be a server, personal computer, laptop computer, smart phone, tablet computer, intelligent mobile phone, etc. The computer device may include a processor, a memory, and a network interface connected through a system bus or connected wirelessly. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store data during the electricity meter management process. The network interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements an electricity meter management method. Among them, the computer device can be implemented by an independent computer device or a computer device cluster composed of multiple computer devices. It should be noted that the memory of the computer device is not limited to the above-mentioned memory, and may also include high-speed random access memory, volatile solid-state memory, etc. In addition, the composition architecture of the computer device is not limited to the above situation, and some components can also be added or omitted.

[0059] In an exemplary embodiment, as Figure 2 shown, an electricity meter management method is provided, and this method is applied to Figure 1Taking the computer device in [it] as an example for illustration, it includes the following steps 201 to 203. Among them:

[0060] S101, Collect the operation data of the target electric meter and the environmental data where the target electric meter is located in real time.

[0061] Among them, the operation data of the target electric meter refers to the relevant data during the operation of the electric meter. For example, the operation data can be current, voltage, power consumption, power, power factor, etc. The environmental data where the target electric meter is located refers to the environmental information during the operation of the target electric meter. For example, the environmental data can be the temperature of the target electric meter, the humidity in the target electric meter, etc. The target electric meter can be any type of electric meter. For example, the target electric meter can be a returned electricity meter.

[0062] In the embodiment of the present application, an intelligent acquisition terminal, a temperature and humidity sensor, and an edge gateway can be set inside the target electric meter. The intelligent acquisition terminal is used to collect the operation data of the target electric meter in real time, and the temperature and humidity sensor is used to collect the environmental data where the target electric meter is located in real time. The target electric meter can send the collected operation data and environmental data to the computer device through the built-in edge gateway using the corresponding data transmission protocol (for example, MQTT protocol or HTTP protocol). In this way, the computer device can obtain the operation data and environmental data in real time.

[0063] S102, Analyze the current operation data, current environmental data, historical operation data before the current operation data, and the corresponding historical environmental data using a preset fault identification model to determine the current fault identification result and prediction identification result of the target electric meter.

[0064] Among them, the current operation data refers to the data of the target electric meter collected by the intelligent acquisition terminal at the current moment, and the current environmental data refers to the environmental data collected by the temperature and humidity sensor at the current moment.

[0065] The above-mentioned fault identification model can be used to identify faults in the current operation data, and judge the fault type, faulty component, module to which the component belongs, and fault cause of the target electric meter at the current moment. It can also be used to analyze the operation data (including current operation data and historical operation data) and environmental data (including current environmental data and historical environmental data) at the current moment and before the current moment to predict the fault type, faulty component, module to which the component belongs, and fault cause of the target electric meter in a period of time in the future.

[0066] The fault identification model can be a deep neural network model (Dynamic Neural Network, DNN), a long short-term memory network (Long Short-Term Memory, LSTM), a recurrent neural network model (Recurrent Neural Network, RNN), a convolutional neural network model (Convolutional Neural Networks, CNN), and so on. The network of CNN can be a ResNet series network, a CPN network, a SimpleBaseline network, a Posefix network, a High-Resolution Network (HRNet), and so on. The fault identification model can be a model composed of a single network or a model formed by combining multiple networks.

[0067] In the embodiment of the present application, after the computer device obtains the real-time collected operation data and environmental data, taking the current moment as an example, the computer device can obtain the current operation data and current environmental data corresponding to the current moment. It can also obtain the historical operation data of all moments before the current moment and the corresponding historical environmental data. During the process of using the fault identification model for analysis, the computer device can input the current operation data, current environmental data, historical operation data before the current operation data, and the corresponding historical environmental data into the fault identification model. For any one of the data, the fault identification model can extract the feature information of the data. And comprehensively analyze these feature information to determine the fault type, faulty component, module to which the component belongs, and fault cause of the target electric meter at the current moment, that is, the current fault identification result of the target electric meter. And determine the fault type, faulty component, module to which the component belongs, and fault cause of the predicted target electric meter in a future period of time, that is, the predicted identification result of the target electric meter.

[0068] S103, based on the current fault identification result and the predicted identification result, output a fault handling strategy for the target electric meter.

[0069] Among them, the fault handling strategy refers to the strategy for repairing the fault identification result of the target electric meter. For example, the fault handling strategy can be to replace the faulty component, check the health status of the power transmission line, or perform a firmware upgrade, and so on.

[0070] It should be noted that different types of fault types and severity levels correspond to different fault handling strategies. For example, if the fault type is a loose solder joint of a transformer and it is at the severe level, then the corresponding fault handling strategy is to re-weld. If it is at the normal level, then the corresponding fault handling strategy is to repair the solder joint. That is to say, different fault recognition results correspond to different fault handling strategies, and there is a corresponding mapping relationship between the fault recognition results and the fault handling strategies.

[0071] In the embodiment of the present application, the current fault recognition result indicates whether there is a fault in the target electric meter at the current moment, and the predicted recognition result indicates whether there is a fault in the target electric meter within a preset future time point. If the current fault recognition result is that there is no fault and the predicted recognition result is that there is a fault, then the computer device can determine the fault handling strategy corresponding to the predicted recognition result based on the mapping relationship. If the fault types and severity levels indicated by the current fault recognition result and the predicted recognition result are the same fault, then the computer device can determine the fault handling strategy corresponding to the current fault recognition result based on the mapping relationship.

[0072] In the above electric meter management method, the operation data of the target electric meter and the environmental data where the target electric meter is located are collected in real time; the preset fault recognition model is used to analyze the current operation data, current environmental data, historical operation data before the current operation data, and the corresponding historical environmental data to determine the current fault recognition result and the predicted recognition result of the target electric meter; based on the current fault recognition result and the predicted recognition result, the fault handling strategy of the target electric meter is output. By collecting operation data and environmental data in real time and using the preset fault recognition model to comprehensively analyze the current and historical operation and environmental data, this method can accurately determine the current fault recognition result of the target electric meter, and can deeply identify hidden faults to identify subtle damages of internal components of the electric meter, poor contact of lines and other hidden faults. This makes the fault detection process more meticulous, can obtain more accurate fault detection results, thereby reducing the difficulty and maintenance cost of fault detection and repair. At the same time, using the fault recognition model can detect the faults of the electric meter in real time and avoid the limitations of manual inspection and experience judgment.

[0073] In one embodiment, as Figure 3 shown, the specific content of analyzing the current operation data, current environmental data, historical operation data before the current operation data, and the corresponding historical environmental data by using the preset fault recognition model to determine the current fault recognition result and the predicted recognition result of the target electric meter is introduced. The specific content includes:

[0074] S201, extract the current data feature information of the current operation data, the current environmental feature information of the current environmental data, the historical data feature information of the historical operation data, and the historical environmental feature information of the historical environmental data.

[0075] In the embodiment of the present application, the computer device may respectively perform convolution operations on the current operation data and the current environment data by using multiple convolutional layers in the fault recognition model to obtain the current data feature information of the current operation data and the current environment feature information of the current environment data. Then, by extracting the temporal features of the historical operation data and the historical environment data, the trend features and periodic features of the historical operation data and the historical environment data are obtained, and the trend features and periodic features are used as the historical data feature information and the historical environment feature information.

[0076] S202. Determine the current fault recognition result of the target electric meter based on the current data feature information and the current environment feature information; and determine the predicted recognition result of the target electric meter based on the current data feature information, the historical data feature information, the current environment feature information, and the historical environment feature information.

[0077] In the embodiment of the present application, the computer device may analyze whether the target electric meter is abnormal at the current moment based on the current data feature information and the current environment feature information. If an abnormality occurs, continue to determine the cause and location of the abnormality based on these features. If no abnormality occurs, determine the data change trend of the target electric meter based on the current data feature information, the historical data feature information, the current environment feature information, and the historical environment feature information. Then, based on the data change trend, determine the data after a period of time in the future, that is, the predicted recognition result of the target electric meter.

[0078] In the above electric meter management method, the current data feature information of the current operation data, the current environment feature information of the current environment data, the historical data feature information of the historical operation data, and the historical environment feature information of the historical environment data are extracted; the current fault recognition result of the target electric meter is determined based on the current data feature information and the current environment feature information; and the predicted recognition result of the target electric meter is determined based on the current data feature information, the historical data feature information, the current environment feature information, and the historical environment feature information. By extracting features from the current operation data, the current environment data, the historical operation data, and the historical environment data, this method can construct an all-round description system of the operation state of the electric meter. The comprehensive extraction of such multi-dimensional features makes the fault diagnosis no longer limited to local or surface phenomena, and greatly improves the accuracy of the current fault recognition.

[0079] Next, a specific example is used to introduce the specific content of determining the current fault recognition result of the target electric meter based on the current data feature information and the current environment feature information, as Figure 4 shown, including:

[0080] S301. Analyze the fault type, faulty components, the module to which the components belong, and the cause of the fault of the target electricity meter based on the current data feature information and the current environmental feature information.

[0081] In the embodiment of the present application, the computer device can compare the current environmental feature information with the preset safety range to determine whether the current environmental feature is within the preset safety range. If it is, it is determined that the target electricity meter has not suffered a serious fault, and the current data feature information is further analyzed to further determine whether the target electricity meter is abnormal. If it is abnormal, analyze the fault type, faulty components, the module to which the components belong, and the cause of the fault of the target electricity meter.

[0082] S302. Take the fault type, faulty components, the module to which the components belong, and the cause of the fault of the target electricity meter as the current fault identification result of the target electricity meter.

[0083] In the embodiment of the present application, when the fault type, faulty components, the module to which the components belong, and the cause of the fault of the target electricity meter are obtained, the fault type, faulty components, the module to which the components belong, and the cause of the fault can be directly used as the current fault identification result of the target electricity meter at the current moment.

[0084] In the above electricity meter management method, based on the current data feature information and the current environmental feature information, analyze the fault type, faulty components, the module to which the components belong, and the cause of the fault of the target electricity meter; take the fault type, faulty components, the module to which the components belong, and the cause of the fault of the target electricity meter as the current fault identification result of the target electricity meter. This method is analyzed based on the current data feature information and the current environmental feature information, and can accurately judge the fault type, faulty components, the module to which the components belong, and the cause of the fault of the target electricity meter, so as to accurately determine the current fault identification result of the target electricity meter.

[0085] Next, a specific example is used to introduce the specific content of determining the predicted identification result of the target electricity meter based on the current data feature information, historical data feature information, current environmental feature information, and historical environmental feature information, as Figure 5 shown, and the specific content includes:

[0086] S401. Fuse the current data feature information and the historical data feature information to obtain the data trend change during the operation of the target electricity meter.

[0087] In an embodiment of the present application, the computer device may fuse the current data feature information and the historical data feature information according to a selected fusion method to obtain fused data features. Then, using the moving average method, a time window is set. For example, the time window can be 10 minutes. Calculate the average value of the fused data features within each time window to obtain a series of moving average values. By observing the changes in these moving average values, the data trend changes during the operation of the target electric meter can be intuitively understood. Among them, the fusion method can be time series weighted fusion or principal component analysis fusion, etc.

[0088] S402. Fuse the current environmental feature information and the historical environmental feature information to obtain the environmental trend changes during the operation of the target electric meter.

[0089] In an embodiment of the present application, the computer device may fuse the current environmental feature information and the historical environmental feature information according to time series weighted fusion or principal component analysis fusion to obtain fused environmental features. Then, use the moving average method to calculate the average value of the fused environmental features within each time window to obtain a series of moving average values. By observing the changes in these moving average values, the environmental trend changes during the operation of the target electric meter can be intuitively understood.

[0090] S403. Based on the data trend changes and the environmental trend changes, determine the prediction and identification result of the target electric meter at a future preset time point.

[0091] In an embodiment of the present application, the computer device may determine the operation data and environment at a future preset time point according to the data trend changes and the environmental trend changes. Then, analyze the operation data and environment at the future preset time point to determine whether there is a fault in the target electric meter at the future preset time point. If there is a fault, analyze information such as the cause of the fault and the location of the fault. If there is no fault, it means that the target electric meter is normal within a certain period in the future.

[0092] In one embodiment, as Figure 6 shown, the specific content of determining the prediction and identification result of the target electric meter at a future preset time point based on the data trend changes and the environmental trend changes includes:

[0093] S501. Obtain the data prediction result of the target electric meter at a future preset time point based on the data trend changes; and, based on the environmental trend changes, determine the environmental prediction result of the target electric meter at a future preset time point.

[0094] In the embodiments of the present application, the computer device can simulate a data trend change graph based on the data trend change, and determine the operation data at a future preset time point according to the data trend change graph, and use this operation data as the data prediction result. At the same time, it can also simulate an environmental trend change graph based on the environmental trend change, and determine the environmental data at a future preset time point according to the environmental trend change graph, and use this environmental data as the environmental prediction result.

[0095] S502. Adjust the data prediction result by using the environmental prediction result to obtain the predicted identification result of the target electric meter at a future preset time point.

[0096] In the embodiments of the present application, since the data prediction result takes less account of environmental changes during the prediction process, after the computer device obtains the environmental prediction result, it can use the environmental prediction result to adjust the data prediction result, and use the adjusted data prediction result as the predicted identification result of the target electric meter at a future preset time point.

[0097] In the above electric meter management method, the data prediction result of the target electric meter at a future preset time point is obtained based on the data trend change; and, based on the environmental trend change, the environmental prediction result of the target electric meter at a future preset time point is determined; the data prediction result is adjusted by using the environmental prediction result to obtain the predicted identification result of the target electric meter at a future preset time point. This method starts from two perspectives of environmental data and operation data, and based on the data trend changes from these two perspectives, it can obtain the data prediction result of the target electric meter at a future preset time point, which can provide a key basis for power system planning.

[0098] In the above electric meter management method, the current data feature information and the historical data feature information are fused to obtain the data trend change during the operation of the target electric meter; the current environmental feature information and the historical environmental feature information are fused to obtain the environmental trend change during the operation of the target electric meter; based on the data trend change and the environmental trend change, the predicted identification result of the target electric meter at a future preset time point is determined. This method can sensitively capture the subtle change trend of the electric meter operation data by fusing the current and historical data feature information. By synthesizing the data and environmental trend changes, it can improve the accuracy of electric meter fault prediction.

[0099] Before analyzing using the fault identification model, the operation data and environmental data collected in real time can also be standardized. Then, in one embodiment, the specific content of the standardization process is introduced, as Figure 7 shown, and the specific content includes:

[0100] S601. Remove the outliers in the operation data of the target electric meter and the environmental data where the target electric meter is located.

[0101] Among them, the outlier can be a value in the operation data or the environmental data that is significantly different from other values.

[0102] In the embodiment of the present application, the computer device can analyze the operation data of the target electric meter and the environmental data where the target electric meter is located through an outlier detection algorithm to determine the outliers in the operation data and the outliers in the environmental data. Then, the outliers in the operation data and the outliers in the environmental data are deleted. Among them, the outlier detection algorithm can be the standard deviation method, the interquartile range method, the isolation forest algorithm, etc.

[0103] S602, perform data filling on the operation data and environmental data after removal.

[0104] In the embodiment of the present application, the computer device can perform data filling on the operation data after removal through a related filling algorithm to obtain the filled operation data. And perform data filling on the environmental data after removal to obtain the filled environmental data. Among them, the filling algorithm can be filling based on a time series model, filling based on a similar day pattern, or interpolation method filling, etc.

[0105] S603, after performing standardization processing on the filled operation data and environmental data, perform the step of analyzing the current operation data, the current environmental data, the historical operation data before the current operation data, and the corresponding historical environmental data by using a preset fault identification model.

[0106] In the embodiment of the present application, the computer device can perform standardization processing on the filled operation data and environmental data, that is, map the operation data and environmental data to a specific range, and the mean of each data feature is 0 and the variance is 1 after standardization processing. In this way, when using a preset fault identification model for fault identification, the stability and convergence speed of the fault identification process can be improved.

[0107] In the above electric meter management method, remove the outliers in the operation data of the target electric meter and the environmental data where the target electric meter is located; perform data filling on the operation data and environmental data after removal; after performing standardization processing on the filled operation data and environmental data, perform the step of analyzing the current operation data, the current environmental data, the historical operation data before the current operation data, and the corresponding historical environmental data by using a preset fault identification model. This method reduces the noise interference in the operation data and environmental data by removing outliers. After removing the outliers and then performing data filling, the integrity of the data can be restored, ensuring that important information will not be missed due to data loss during the analysis process, and guaranteeing a comprehensive detection of the operation status of the electric meter and environmental changes.

[0108] The above embodiments are all introductions to the actual application process of the fault identification model. Next, a detailed introduction to the training process of the fault identification model will be given through an embodiment, as follows Figure 8 It also includes:

[0109] S701, obtaining sample operation data, sample environment data, reference fault identification results, and reference prediction identification results.

[0110] In the embodiments of the present application, the historical operation data and corresponding environment data of different types of electric meters during operation are stored in the database. The computer device can search for the historical operation data of different types and the corresponding environment data in the database, and use these historical operation data as sample operation data, and use the corresponding environment data as sample environment data. Whether different types of electric meters have faults during operation is used as the reference fault identification result, and whether there will be faults in a future period is used as the reference prediction identification result.

[0111] S702, inputting the sample operation data and sample environment data into the initial fault identification model, and using the initial fault identification model for analysis to obtain sample fault identification results and sample prediction identification results.

[0112] In the embodiments of the present application, since CNN can automatically learn data features, and LSTM is suitable for processing time series data and can capture long-term dependencies in historical data for future trend prediction. Therefore, CNN and LSTM can be combined to obtain the initial fault identification model.

[0113] The computer device can input the sample operation data and sample environment data into the initial fault identification model. The initial fault identification model can extract some more meaningful features from the sample operation data and sample environment data. For example, through time series analysis of sensor data such as voltage, current, and temperature of the electric meter, the features that can be extracted can be the frequency feature of current fluctuation, the temperature change trend, the correlation between voltage and load, the periodicity and abnormal patterns of power data. Then, based on these features, the sample fault identification results and sample prediction identification results corresponding to the sample operation data are analyzed.

[0114] S703, adjusting the parameter information of the initial fault identification model based on the identification difference between the sample fault identification result and the reference fault identification result, and the prediction difference between the sample prediction identification result and the reference prediction identification result, until the sample fault identification result and sample prediction identification result output by the adjusted initial fault identification model meet the preset conditions, and obtaining the trained fault identification model.

[0115] In the embodiments of the present application, the reference fault recognition result and the reference prediction recognition result can be used as the gold standard. After obtaining the sample fault recognition result and the sample prediction recognition result, the computer device can calculate the recognition difference between the sample fault recognition result and the reference fault recognition result, and calculate the prediction difference between the sample prediction recognition result and the reference prediction recognition result. And according to the recognition difference and the prediction difference, the parameter information of the initial fault recognition model is adjusted. Through multiple iterative adjustment processes, the recognition difference and the prediction difference approach zero. When the preset conditions are met, the training process is completed, and a trained fault recognition model is obtained.

[0116] In the above meter management method, sample operation data, sample environment data, a reference fault recognition result, and a reference prediction recognition result are obtained; the sample operation data and the sample environment data are input into the initial fault recognition model, and the initial fault recognition model is used for analysis to obtain a sample fault recognition result and a sample prediction recognition result; based on the recognition difference between the sample fault recognition result and the reference fault recognition result, and the prediction difference between the sample prediction recognition result and the reference prediction recognition result, the parameter information of the initial fault recognition model is adjusted until the sample fault recognition result and the sample prediction result output by the adjusted initial fault recognition model meet the preset conditions, and a trained fault recognition model is obtained. The rich and diverse sample data in this method comes from the actual meter monitoring scenario and can truly reflect various situations that may be encountered during the operation of the meter. This makes the trained fault recognition model have stronger real-world adaptability, can accurately identify the faults that occur in the meter under complex actual environments, and can avoid the model from failing in actual applications due to single or unrealistic training data.

[0117] As a specific embodiment of the present application, the meter management process will be specifically introduced next. In one embodiment, as Figure 9 shown, the meter management method includes:

[0118] S801, real-time collect the operation data of the target meter and the environment data where the target meter is located;

[0119] S802, remove the outliers in the operation data of the target meter and the environment data where the target meter is located;

[0120] S803, perform data filling on the removed operation data and environment data;

[0121] S804, after standardizing the filled operation data and environment data, extract the current data feature information of the current operation data, the current environment feature information of the current environment data, the historical data feature information of the historical operation data, and the historical environment feature information of the historical environment data;

[0122] S805. Analyze the fault type, faulty components, modules to which the components belong, and fault causes of the target electricity meter based on the current data characteristic information and the current environmental characteristic information;

[0123] S806. Use the fault type, faulty components, modules to which the components belong, and fault causes of the target electricity meter as the current fault identification result of the target electricity meter;

[0124] S807. Integrate the current data characteristic information and the historical data characteristic information to obtain the data trend change during the operation of the target electricity meter;

[0125] S808. Integrate the current environmental characteristic information and the historical environmental characteristic information to obtain the environmental trend change during the operation of the target electricity meter;

[0126] S809. Obtain the data prediction result of the target electricity meter at a future preset time point based on the data trend change; and, determine the environmental prediction result of the target electricity meter at the future preset time point based on the environmental trend change;

[0127] S810. Use the environmental prediction result to adjust the data prediction result to obtain the prediction identification result of the target electricity meter at the future preset time point;

[0128] S811. Output the fault handling strategy of the target electricity meter based on the current fault identification result and the prediction identification result.

[0129] It should be understood that although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the arrows, these steps do not necessarily need to be executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily need to be executed at the same moment, but can be executed at different moments. The execution order of these steps or stages does not necessarily need to be sequential, but can be executed alternately or in turn with at least some of the steps or stages in other steps or other steps.

[0130] Based on the same inventive concept, the embodiments of the present application also provide an electricity meter management device for implementing the above-mentioned electricity meter management method. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the following electricity meter management device can refer to the limitations on the electricity meter management method in the above text, and will not be repeated here.

[0131] In an exemplary embodiment, as Figure 10As shown, a power meter management device is provided, including: a collection module 11, an analysis module 12, and an output module 13, where:

[0132] The collection module 11 is configured to collect the operation data of the target power meter and the environmental data where the target power meter is located in real time;

[0133] The analysis module 12 is configured to analyze the current operation data, the current environmental data, the historical operation data before the current operation data, and the corresponding historical environmental data by using a preset fault identification model to determine the current fault identification result and the prediction identification result of the target power meter;

[0134] The output module 13 is configured to output a fault handling strategy for the target power meter based on the current fault identification result and the prediction identification result.

[0135] In an exemplary embodiment, the above analysis module 12 includes a feature extraction unit and a result determination unit, where:

[0136] The feature extraction unit is configured to extract the current data feature information of the current operation data, the current environmental feature information of the current environmental data, the historical data feature information of the historical operation data, and the historical environmental feature information of the historical environmental data;

[0137] The result determination unit is configured to determine the current fault identification result of the target power meter based on the current data feature information and the current environmental feature information; and determine the prediction identification result of the target power meter based on the current data feature information, the historical data feature information, the current environmental feature information, and the historical environmental feature information.

[0138] In an exemplary embodiment, the above result determination unit is further configured to analyze the fault type, the faulty component, the module to which the component belongs, and the fault cause of the target power meter based on the current data feature information and the current environmental feature information; and use the fault type, the faulty component, the module to which the component belongs, and the fault cause of the target power meter as the current fault identification result of the target power meter.

[0139] In an exemplary embodiment, the above result determination unit is further configured to fuse the current data feature information and the historical data feature information to obtain the data trend change during the operation of the target power meter; fuse the current environmental feature information and the historical environmental feature information to obtain the environmental trend change during the operation of the target power meter; and determine the prediction identification result of the target power meter at a preset future time point based on the data trend change and the environmental trend change.

[0140] In an exemplary embodiment, the above result determination unit is further configured to obtain a data prediction result of the target electric meter at a preset future time point based on the change of the data trend; and, determine an environmental prediction result of the target electric meter at the preset future time point based on the change of the environmental trend; and adjust the data prediction result by using the environmental prediction result to obtain a prediction recognition result of the target electric meter at the preset future time point.

[0141] In an exemplary embodiment, the above electric meter management device further includes: a removal module, a filling module, and a processing module, where:

[0142] The removal module is configured to remove outliers from the operation data of the target electric meter and the environmental data where the target electric meter is located;

[0143] The filling module is configured to perform data filling on the operation data and environmental data after removal;

[0144] The processing module is configured to perform standardization processing on the operation data and environmental data after filling, and then execute the step of analyzing the current operation data, the current environmental data, the historical operation data before the current operation data, and the corresponding historical environmental data by using a preset fault recognition model.

[0145] In an exemplary embodiment, the above electric meter management device further includes: an acquisition module, an initial analysis module, and an adjustment module, where:

[0146] The acquisition module is configured to acquire sample operation data, sample environmental data, a reference fault recognition result, and a reference prediction recognition result;

[0147] The initial analysis module is configured to input the sample operation data and the sample environmental data into an initial fault recognition model, and perform analysis by using the initial fault recognition model to obtain a sample fault recognition result and a sample prediction recognition result;

[0148] The adjustment module is configured to adjust the parameter information of the initial fault recognition model based on the recognition difference between the sample fault recognition result and the reference fault recognition result, and the prediction difference between the sample prediction recognition result and the reference prediction recognition result, until the sample fault recognition result and the sample prediction recognition result output by the adjusted initial fault recognition model meet the preset conditions, so as to obtain a trained fault recognition model.

[0149] Each module in the above electric meter management device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the computer device in the form of hardware or be independent of the processor, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0150] In an exemplary embodiment, a computer device is provided, which includes a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the content of any one of the above embodiments of the electricity meter management method is implemented.

[0151] In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the content of any one of the above embodiments of the electricity meter management method is implemented.

[0152] In an embodiment, a computer program product is provided, which includes a computer program. When the computer program is executed by a processor, the content of any one of the above embodiments of the electricity meter management method is implemented.

[0153] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0154] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0155] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in the present application.

[0156] The above embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A method for managing an electric meter, characterized in that: The method comprises: Collecting the operating data of the target electric meter and the environmental data of the target electric meter in real time; Analyze the current operation data, the current environment data, the historical operation data before the current operation data, and the corresponding historical environment data using a preset fault identification model to determine the current fault identification result and the predicted identification result of the target electric meter; Based on the current fault identification result and the predicted identification result, a fault handling strategy for the target electric meter is output.

2. The method according to claim 1, characterized in that: The method of analyzing the current operation data, the current environment data, the historical operation data before the current operation data and the corresponding historical environment data by using the preset fault identification model to determine the current fault identification result and the predicted identification result of the target electric meter includes: Extracting current data feature information of the current operation data, current environment feature information of the current environment data, historical data feature information of the historical operation data, and historical environment feature information of the historical environment data; Based on the current data characteristic information and the current environment characteristic information, determine the current fault identification result of the target meter; and based on the current data characteristic information, the historical data characteristic information, the current environment characteristic information and the historical environment characteristic information, determine the predicted identification result of the target meter.

3. The method according to claim 2, characterized in that The determining, based on the current data characteristic information and the current environment characteristic information, a current fault identification result of the target electric meter includes: Analyze the fault type, faulty components, modules to which the components belong, and fault causes of the target electric meter based on the current data characteristic information and the current environment characteristic information; The fault type, faulty components, modules to which the components belong, and fault causes of the target electric meter are used as the current fault identification result of the target electric meter.

4. The method according to claim 2, characterized in that: The determining the prediction identification result of the target electric meter based on the current data feature information, the historical data feature information, the current environment feature information and the historical environment feature information includes: Merging the current data characteristic information with the historical data characteristic information to obtain data trend changes during the operation of the target electric meter; The current environmental characteristic information and the historical environmental characteristic information are integrated to obtain the environmental trend change during the operation of the target electric meter; Based on the data trend change and the environmental trend change, a prediction identification result of the target electric meter at a preset time point in the future is determined.

5. The method according to claim 4, characterized in that The determining, based on the data trend change and the environmental trend change, a prediction identification result of the target electric meter at a preset time point in the future, comprises: Acquire a data prediction result of the target electric meter at a preset time point in the future based on the data trend change; and determine an environmental prediction result of the target electric meter at a preset time point in the future based on the environmental trend change; The data prediction result is adjusted using the environmental prediction result to obtain a prediction identification result of the target electric meter at a preset time point in the future.

6. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Removing abnormal values ​​in the operating data of the target electric meter and the environmental data of the target electric meter; Perform data filling for the removed operation data and environmental data; After the filled operation data and environmental data are standardized, the step of using a preset fault identification model to analyze the current operation data, the current environmental data, the historical operation data before the current operation data, and the corresponding historical environmental data is executed.

7. The method according to any one of claims 1 to 4, characterized in that: The training process of the fault identification model includes: Obtaining sample operation data, sample environment data, reference fault identification results, and reference prediction identification results; Inputting the sample operation data and the sample environment data into an initial fault identification model, and using the initial fault identification model to perform analysis to obtain a sample fault identification result and a sample prediction identification result; Based on the recognition difference between the sample fault recognition result and the reference fault recognition result, and the prediction difference between the sample prediction recognition result and the reference prediction recognition result, the parameter information of the initial fault recognition model is adjusted until the sample fault recognition result and the sample prediction recognition result output by the adjusted initial fault recognition model meet preset conditions, thereby obtaining a trained fault recognition model.

8. An electric meter management device, characterized in that: The device comprises: A collection module, used for collecting the operating data of the target electric meter and the environmental data of the target electric meter in real time; An analysis module, used to analyze the current operation data, the current environment data, the historical operation data before the current operation data and the corresponding historical environment data using a preset fault identification model, to determine the current fault identification result and the predicted identification result of the target electric meter; An output module is used to output a fault handling strategy of the target electric meter based on the fault identification result and the predicted identification result.

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 method according to 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 the method according to any one of claims 1 to 7 are implemented.

Images