Electricity meter fault operation and maintenance method based on multi-dimensional data test analysis

Abstract

The present invention discloses an electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis, which relates to the field of electric power monitoring technology. The method includes: collecting the operating data of the target electric energy meter under different current, voltage and temperature conditions, recording the operating time of each condition, setting the test conditions and times; testing the electric energy meter, calculating the average error of the mutual inductor and sampling accuracy; configuring the operation and maintenance identification computing power according to the number of tests, performing fault identification based on the error data respectively, generating and integrating the fault operation and maintenance results under each operating condition, and obtaining the overall fault operation and maintenance identification results of the target electric energy meter. The present invention solves the technical problem that the traditional fault identification method in the prior art is difficult to capture the subtle changes and potential faults of the electric energy meter in a complex operating environment in real time and accurately, and achieves the technical effect of realizing accurate identification and efficient operation and maintenance of electric energy meter faults in a complex operating environment through multi-dimensional data test and analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power monitoring, and in particular to an electric energy meter fault operation and maintenance method based on multi-dimensional data testing and analysis. Background Art

[0002] In modern power systems, as the core equipment for power metering, electric energy meters undertake the key tasks of power metering and data collection. Their accuracy and reliability have a significant impact on the interests of power companies and users. In the actual operation of power systems, electric energy meters face challenges from a variety of current, voltage and temperature operating conditions. These complex and changeable operating environments often lead to unstable performance of electric energy meters and even cause failures. Traditional fault identification methods often rely on manual inspections and regular testing, which makes it difficult to capture subtle changes and potential faults of electric energy meters in complex operating environments in real time and accurately. Therefore, how to accurately identify faults of electric energy meters in complex and changeable operating environments has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present application provides an electric energy meter fault operation and maintenance method based on multi-dimensional data testing and analysis, which is used to solve the technical problem that traditional fault identification methods in the prior art are difficult to capture subtle changes and potential faults of electric energy meters in complex operating environments in real time and accurately.

[0004] The present application provides an electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis, the method comprising: collecting a current and voltage operating condition set and a temperature operating condition set in a target electric energy meter operating environment, and collecting a current and voltage operating duration set and a temperature operating duration set corresponding to the current and voltage operating condition set and the temperature operating condition set; setting a test condition set according to the current and voltage operating condition set and the temperature operating condition set, and setting a test number set according to the current and voltage operating duration set and the temperature operating duration set; testing the target electric energy meter according to the test condition set and the test number set, and calculating and obtaining a transformer average error set and a sampling accuracy average error set; configuring an operation and maintenance identification computing power set according to the test number set, performing electric energy meter fault operation and maintenance identification based on the transformer average error set and the sampling accuracy average error set, respectively, obtaining a conditional fault operation and maintenance identification result set, and fusing to obtain the fault operation and maintenance identification result of the target electric energy meter.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0006] The electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis provided in the present application relates to the field of electric power monitoring technology. By collecting the operating data of the electric energy meter under different current, voltage and temperature conditions, setting the test conditions and conducting multiple tests, the average error of the mutual inductor and sampling accuracy is calculated. The operation and maintenance identification computing power is configured based on the error and the number of tests, fault identification is performed, and the identification results under multiple conditions are integrated to generate an overall fault operation and maintenance report for the electric energy meter. The technical problem that the traditional fault identification method in the prior art is difficult to capture the subtle changes and potential faults of the electric energy meter in a complex operating environment in real time and accurately is solved. Through multi-dimensional data test and analysis, combined with the operation and maintenance identification computing power configuration and the fault operation and maintenance identification algorithm, the technical effect of accurate identification of electric energy meter faults in a complex operating environment and efficient operation and maintenance is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A schematic flow chart of an electric energy meter fault operation and maintenance method based on multi-dimensional data testing and analysis provided in an embodiment of the present application.

[0009] Figure 2 A schematic diagram of a process for obtaining a fault operation and maintenance identification result of a target electric energy meter in an electric energy meter fault operation and maintenance method based on multidimensional data test and analysis provided in an embodiment of the present application. DETAILED DESCRIPTION

[0010] The present application provides an electric energy meter fault operation and maintenance method based on multi-dimensional data testing and analysis, which is used to solve the technical problem that traditional fault identification methods in the prior art are difficult to capture subtle changes and potential faults of electric energy meters in complex operating environments in real time and accurately.

[0011] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0012] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0013] Examples, such as Figure 1 As shown, the present application provides an electric energy meter fault operation and maintenance method based on multi-dimensional data test analysis, the method comprising:

[0014] P10: Collect the current and voltage operating condition sets and the temperature operating condition sets in the target electric energy meter operating environment, and collect the current and voltage operating duration sets and the temperature operating duration sets corresponding to the current and voltage operating condition sets and the temperature operating condition sets.

[0015] Furthermore, step P10 of the embodiment of the present application further includes:

[0016] P11: In the target electric energy meter operating environment, the current and voltage are continuously monitored within a preset time period to obtain a current and voltage information set, and then a current and voltage operating condition set is obtained by dividing the current and voltage operating condition set; P12: The operating temperature is continuously monitored within the preset time period to obtain a temperature information set, and then a temperature operating condition set is obtained by dividing the temperature operating condition set; P13: The operating time of the target electric energy meter in the current and voltage operating condition set is separately collected to obtain a current and voltage operating duration set; P14: The operating time of the target electric energy meter in the temperature operating condition set is separately collected to obtain a temperature operating duration set.

[0017] It should be understood that by collecting multi-dimensional data of the target electric energy meter in the actual operating environment, different current, voltage and temperature conditions during the operation of the electric energy meter can be obtained, and their duration can be recorded to provide a basis for subsequent fault analysis and testing.

[0018] First, within a preset time period, the current and voltage in the target electric energy meter operating environment are continuously monitored to generate a current and voltage information set. The preset time period can be a regular operation and maintenance cycle, such as a month or a quarter. During this period, the electric energy meter records the current and voltage data at different time points to form a complete monitoring data set. To simplify the analysis, these data are divided, and the current and voltage values ​​with small differences are classified as the same operating condition to generate a current and voltage operating condition set. For example, if the difference in current or voltage values ​​at different time points is within 5%, these similar data are classified as a current and voltage operating condition. This division method can help identify the typical operating mode of the electric energy meter under different loads, which facilitates further analysis of the performance of the equipment under different load conditions.

[0019] Similarly, the operating temperature of the target power meter is continuously monitored within a preset time period to generate a temperature information set. The operating status of the power meter may be different under different ambient temperatures, so monitoring the temperature conditions is crucial for analyzing the performance of the power meter in different environments. Similar temperature values ​​are classified into the same type of operating temperature conditions to generate a set of temperature operating conditions to ensure that the effects of different temperatures can be accurately captured. For example, if the temperature changes within a certain range (such as ±3°C), these temperature conditions can be regarded as similar working environments. This temperature classification method can reflect the performance of the power meter in different environments such as low temperature and high temperature.

[0020] Furthermore, by continuously monitoring the working time of the target electric energy meter under different current and voltage operating conditions, the operating time of the electric energy meter under each condition is collected to generate a current and voltage operating time set, wherein the current and voltage operating time set includes multiple current and voltage operating time. These time data represent the working duration of the electric energy meter under a specific current and voltage combination. For example, if the operating time under a certain current and voltage combination accounts for a large part of the total time period, this may mean that the operating frequency of the electric energy meter under this operating condition is high and deserves special attention.

[0021] On the basis of temperature monitoring, the operation time of the energy meter under different temperature operating conditions is further collected to form a temperature operation time set, which includes multiple temperature operation time sets. These time data are similar to the current and voltage operation time, and are used to record the working time of the energy meter under different temperature environments. The operating characteristics of the energy meter under different temperature conditions may be different, so the collection of temperature operation time helps to evaluate the impact of temperature on the metering accuracy of the energy meter.

[0022] P20: According to the current and voltage operating condition set and the temperature operating condition set, a test condition set is set, and according to the current and voltage operating duration set and the temperature operating duration set, a test number set is set.

[0023] Furthermore, step P20 of the embodiment of the present application further includes:

[0024] P21: Use the current and voltage operating condition set as the standard test voltage and current parameter set, and combine it with the temperature operating condition set to obtain a test condition set; P22: Calculate and obtain a test number set based on the ratio of the current and voltage operating time and temperature operating time corresponding to each test condition to the sum of the current and voltage operating time set and the temperature operating time set, combined with the total number of tests.

[0025] Optionally, based on the collected current and voltage operating condition sets and temperature operating condition sets, the test conditions and test times are reasonably set to ensure that the multi-dimensional testing process of the electric energy meter can fully reflect various operating conditions in actual operation.

[0026] First, the current and voltage operating condition set is regarded as a standard test voltage and current parameter set. These parameters represent various current and voltage combinations that the electric energy meter may encounter in actual operation. Subsequently, these current and voltage parameters are combined one by one with the temperature operating condition set to obtain a test condition set, which contains a variety of test conditions. This combination ensures that the test can simulate the working conditions of the electric energy meter under various current, voltage and temperature combinations, and then evaluate its performance. For example, one test condition can be a combination of current 200A, voltage 220V, and temperature 25°C, and another test condition can be a combination of current 150A, voltage 240V, and temperature 35°C.

[0027] Next, according to the current voltage operation time and temperature operation time corresponding to each test condition, calculate its proportion of the total current voltage operation time and temperature operation time. This proportion reflects the relative frequency of the test condition in the actual operating environment. Then, combined with the preset total number of tests (for example, in order to obtain sufficient test data, it may be set to 1000 times or more), the above ratio is used to calculate the number of tests for each test condition. For example, if the operation time of a current voltage and temperature combination occupies 10% of the entire operation cycle, the corresponding duration ratio is 0.1. According to the total number of tests (such as 1000 times), combined with the duration ratio, the number of tests corresponding to each test condition is calculated to be 100 times (1000×0.1). In this way, the number of tests matches the actual working time ratio of the electric energy meter, ensuring the rationality of the test. Conditions with more tests usually correspond to working conditions with higher frequency in the actual operation of the electric energy meter, and conditions with fewer tests correspond to working conditions with lower frequency.

[0028] This process ensures the diversity of test conditions and the rationality of test times, so that it can fully cover various working conditions in the actual operation of the electricity meter, provide strong data support for the fault operation and maintenance of the electricity meter, and ensure the accuracy and reliability of the test results.

[0029] P30: Test the target electric energy meter according to the test condition set and the test number set, and calculate and obtain a mutual inductor average error set and a sampling accuracy average error set.

[0030] Furthermore, step P30 of the embodiment of the present application further includes:

[0031] P31: Obtain a standard transformer current set, a standard transformer voltage set, a standard current signal set and a standard voltage signal set under each test condition; P32: Test the target electric energy meter according to the test condition set and the test number set to obtain multiple test result sets, and calculate the transformer average error set and the sampling accuracy average error set by combining the standard transformer current set, the standard transformer voltage set, the standard current signal set and the standard voltage signal set.

[0032] Specifically, according to the set test conditions and test times, the target electric energy meter is tested multiple times, and the average mutual inductor error and the average sampling accuracy error of the electric energy meter under different conditions are calculated.

[0033] First, the standard transformer current set, standard transformer voltage set, standard current signal set and standard voltage signal set required under each test condition in the test condition set are obtained, including multiple standard transformer currents, multiple standard transformer voltages, multiple standard current signals and multiple standard voltage signals. These standard signals are important references in the test process and represent the current and voltage values ​​that the electric energy meter should output under ideal conditions. The acquisition of these standard signals usually relies on high-precision measurement equipment and calibration processes. Among them, the standard transformer current is a standard current signal measured by a precise current transformer (CT), representing the real current value under each test condition. The standard transformer has a high degree of accuracy and is used as a comparison target. The standard transformer voltage is a real voltage value measured by a standard voltage transformer (PT), representing the real voltage under each test condition. The standard current signal and the standard voltage signal are measured by precision testing equipment, further providing high-precision current and voltage data. These standard signals are used to compare with the signals collected by the electric energy meter to calculate the error.

[0034] Next, the target energy meter is tested according to the multiple test conditions and test times in the preset test condition set. Under each preset current voltage combination and temperature condition, the energy meter collects real-time current and voltage signals and performs multiple tests. The number of tests is allocated according to the actual running time of each condition to ensure the rationality of the number of tests. After each test, the energy meter will output a set of test results, including the current signal and voltage signal under the test conditions. These result sets include various data recorded by the energy meter under different working conditions.

[0035] Next, the test results of the electric energy meter are compared with the previously acquired standard transformer current, standard transformer voltage, standard current signal and standard voltage signal, and multiple error values ​​are calculated, including a transformer average error set and a sampling accuracy average error set. The transformer average error is calculated by comparing the current measured by the electric energy meter with the standard transformer current, and the same calculation is also applicable to the error of the voltage transformer. After the error under each test condition is calculated multiple times, the average value is taken to obtain the average error of the transformer. Sampling accuracy refers to the accuracy of the electric energy meter when converting analog signals into digital signals. The sampling accuracy error is calculated by comparing the current and voltage signals collected by the electric energy meter with the standard signals. After multiple tests, the system calculates the average value of the sampling errors to obtain the sampling accuracy average error. The transformer average error reflects the accuracy of the electric energy meter in current and voltage measurement, while the sampling accuracy average error evaluates the error level of the electric energy meter during the sampling process, providing a strong basis for fault diagnosis and maintenance of the electric energy meter.

[0036] P40: According to the test number set, an operation and maintenance identification computing power set is configured, and the electric energy meter fault operation and maintenance identification is performed based on the transformer average error set and the sampling accuracy average error set respectively, to obtain a conditional fault operation and maintenance identification result set, and to obtain the fault operation and maintenance identification result of the target electric energy meter by fusion.

[0037] It should be understood that based on the test condition set and the test number set, computing resources are configured to perform fault operation and maintenance identification of the electric energy meter. By analyzing the average error of the transformer and the average error of the sampling accuracy in multiple test results, faults under different conditions are judged, and finally the identification results under all conditions are integrated to generate the overall fault operation and maintenance identification result of the electric energy meter.

[0038] First, during the test process, since the number of tests under different test conditions is different, in order to ensure the efficiency and accuracy of fault identification, the operation and maintenance identification computing power is reasonably allocated according to the number of tests for each test condition. Operation and maintenance identification computing power refers to the computing resources used by the system when performing fault identification, including computing power and data processing capabilities. Exemplarily, the corresponding computing resources are configured according to the number of tests under each test condition and the complexity of the conditions. The more tests there are, the larger the amount of test data under that condition, and more computing power is required for processing. Doing so can ensure the accuracy of fault identification results under test conditions that occur frequently.

[0039] After configuring sufficient computing resources, use these resources to analyze the operating status of the energy meter based on the average error of the transformer and the average error of sampling accuracy calculated under each test condition, and identify potential faults. Exemplarily, by analyzing the average error of the current transformer and the voltage transformer, it is determined whether the performance of the transformer is normal. If the average error of the transformer exceeds the preset threshold range, this is identified as a potential fault. For example, excessive current or voltage errors may indicate that the transformer is aging or damaged. The sampling circuit inside the energy meter is detected to be working properly through error analysis of sampling accuracy. If the sampling accuracy error is large, it may indicate that there is a problem with the ADC (analog-to-digital converter) or other data processing modules of the energy meter, resulting in deviations in the digitization of current and voltage signals.

[0040] Through analysis, a conditional fault operation and maintenance identification result set is obtained, and the conditional fault operation and maintenance identification result set contains multiple conditional fault operation and maintenance identification results. These results correspond to different test conditions and error types, and they reflect the fault conditions of the electric energy meter under different working environments and performance parameters. In order to obtain more comprehensive and accurate fault operation and maintenance identification results, these conditional fault operation and maintenance identification results are fused. This fusion process comprehensively considers the mutual influence and correlation between different test conditions and error types, and can obtain the final fault operation and maintenance identification result of the target electric energy meter through strategies such as weighted average and voting decision.

[0041] Further, according to the test number set, configuring the operation and maintenance identification computing power set. Step P40 of the embodiment of the present application also includes:

[0042] P41: Obtain the integrated identification quantity for operation and maintenance identification of electric energy meter faults; P42: Multiply the ratio of each test number to the maximum test number in the test number set by the integrated identification quantity and round it to obtain the operation and maintenance identification quantity, configure it as the operation and maintenance identification computing power, and obtain the operation and maintenance identification computing power set.

[0043] Optionally, the operation and maintenance identification computing power is configured according to multiple test times in the test number set to ensure reasonable allocation of computing resources for fault operation and maintenance identification. This configuration process not only considers the number of test times, but also combines the requirements of the number of integrated identifications to achieve optimal allocation of computing power.

[0044] Specifically, first obtain the number of integrated identifications required for the operation and maintenance identification of electric energy meter faults. The number of integrated identifications refers to the total number of identification tasks that the system needs to perform in order to complete the operation and maintenance identification of electric energy meter faults. This number is calculated based on the identification tasks required under different test conditions and can be regarded as the total scale of the operation and maintenance tasks in the entire identification process. For example, under different test conditions, it may be necessary to analyze the error performance under multiple current and voltage combinations and different temperatures, and the sum of the number of identification tasks corresponding to each combination is the number of integrated identifications. The purpose of obtaining the number of integrated identifications is to determine the operation and maintenance computing power required in the entire identification process.

[0045] Next, the operation and maintenance identification computing power is allocated proportionally according to the number of tests for each test condition. For each test condition, first calculate the ratio of the number of tests under this condition to the maximum number of tests among all test conditions. This ratio reflects the importance of this condition relative to other conditions. Assuming that the number of tests for a certain condition is 300 times and the maximum number of tests is 600 times, the ratio is 300 / 600=0.5. Then multiply the ratio of the number of tests for each condition to the maximum number of tests by the number of integrated identifications to obtain the number of operation and maintenance identifications under this condition. This number represents the number of fault operation and maintenance identification tasks that the system needs to perform under this condition. In order to ensure that the number of tasks is an integer, the calculation result needs to be rounded. For example, if the number of integrated identifications is 1000 and the ratio under a certain condition is 0.5, the number of operation and maintenance identifications under this condition is 1000×0.5=500. Finally, configure the corresponding operation and maintenance identification computing power according to the number of operation and maintenance identifications under each condition. The allocation of computing power is based on the task complexity and computing requirements under the test conditions. Conditions with more tests receive more computing resources, while conditions with fewer tests receive fewer resources. This ensures efficient use of computing resources and avoids waste and excessive consumption of resources.

[0046] Further, such as Figure 2 As shown, based on the mutual inductor average error set and the sampling accuracy average error set, the electric energy meter fault operation and maintenance identification is performed respectively. The step P40 of the embodiment of the present application also includes:

[0047] P43: According to the integrated identification quantity, train the fault operation and maintenance identification branch of the integrated identification quantity to obtain a fault operation and maintenance identifier; P44: According to the operation and maintenance identification computing power set, respectively combine the transformer average error set and the sampling accuracy average error set and input them into the fault operation and maintenance identification branch of the operation and maintenance identification computing power set, identify and obtain a conditional fault operation and maintenance identification result set, and calculate the mean to obtain a conditional fault operation and maintenance identification result set; P45: According to the size of each operation and maintenance identification computing power, perform fusion weighted calculation on the conditional fault operation and maintenance identification result set to obtain the fault operation and maintenance identification result of the target electric energy meter.

[0048] In a possible embodiment of the present application, an integrated learning method is used to perform fault operation and maintenance identification of electric energy meters based on the average error set of mutual inductors and the average error set of sampling accuracy. This process not only improves the accuracy and robustness of identification, but also fully utilizes the advantages of the operation and maintenance identification computing power set.

[0049] Specifically, first, according to the number of integrated identifications obtained previously, a corresponding number of fault operation and maintenance identification branches are trained. Each branch represents a specific fault identification model, which is used to process transformer errors and sampling accuracy errors under different test conditions. The training data mainly comes from the error signals under various current, voltage and temperature conditions collected previously. When all identification branches have completed training, they are integrated into a complete fault operation and maintenance identifier, that is, an integrated model containing multiple identification branches. The fault operation and maintenance identifier can perform multi-dimensional fault identification based on the input test conditions and error signals.

[0050] Subsequently, according to the configuration of multiple operation and maintenance identification computing powers in the operation and maintenance identification computing power set, the transformer average error set and the sampling accuracy average error set are respectively combined and input into the fault operation and maintenance identification branch of the corresponding computing power. Each branch will output a conditional fault operation and maintenance identification result set, which contains the possible fault types and corresponding fault levels of the electric energy meter under different test conditions. In order to ensure the stability and accuracy of the identification results, the mean of the identification result set under each condition is calculated to obtain the fault operation and maintenance identification result under that condition.

[0051] Finally, according to the size of multiple operation and maintenance identification computing powers in the operation and maintenance identification computing power set, multiple conditional fault operation and maintenance identification results in the conditional fault operation and maintenance identification result set are fused and weighted. The identification result of each condition will be weighted according to its corresponding operation and maintenance identification computing power. The condition with greater operation and maintenance identification computing power indicates that it has been tested more times and has higher importance, so it occupies a larger weight in the weighted calculation. For example, the fault results identified under the condition of more tests will have a greater impact on the final judgment. By weighted fusion of the identification results under all conditions, the overall fault operation and maintenance identification result of the target electric energy meter is generated. This result not only reflects the fault condition of the electric energy meter under a certain specific working condition, but also comprehensively considers the operating status of the electric energy meter under different working environments, thereby providing comprehensive information for the final fault diagnosis.

[0052] Furthermore, step P43 of the embodiment of the present application also includes:

[0053] P43-1: According to the operation and maintenance data records of the electric energy meter, the sample transformer average error set and the sample sampling accuracy average error set are collected, and according to the fault level of the electric energy meter under different sample transformer average errors and sample sampling accuracy average errors, the sample fault operation and maintenance identification result set is marked; P43-2: Using the sample transformer average error set, sample sampling accuracy average error set and sample fault operation and maintenance identification result set, the fault operation and maintenance identification branches of the integrated identification quantity are trained to obtain a fault operation and maintenance identifier.

[0054] Optionally, sample data collection and model training for fault operation and maintenance identification are performed through the operation and maintenance data records of the electric energy meter, specifically including collecting sample data, marking fault levels, and using these data to train the fault operation and maintenance identification branch, and finally generating a fault operation and maintenance identifier.

[0055] First, based on the operation and maintenance data records of the electric energy meter, the sample transformer average error set and the sample sampling accuracy average error set are collected. These data sets represent various error conditions that may occur in the actual operation of the electric energy meter. At the same time, according to the fault level of the electric energy meter under different sample transformer average errors and sample sampling accuracy average errors, the sample data is labeled to obtain a set of sample fault operation and maintenance identification results. Each sample data will be labeled with a corresponding fault level, ranging from minor faults to serious faults, to help the system establish the association between errors and faults. For example, a sample with a larger transformer error may correspond to a higher fault level.

[0056] Next, these sample data sets, namely the sample transformer average error set, the sample sampling accuracy average error set and the sample fault operation and maintenance identification result set, are used to train the fault operation and maintenance identification branch. This training process can use machine learning or deep learning algorithms, such as decision trees, random forests, neural networks, etc. The previously generated sample transformer average error set, sample sampling accuracy average error set and sample fault operation and maintenance identification result set are used as training data sets. According to the number of integrated identifications, multiple identification branch models will be created, and supervised learning will be performed using the training data set, so that each identification branch can learn the characteristics and laws of the electric energy meter fault under different error conditions. After all identification branches have completed training, these branches are integrated to generate a complete fault operation and maintenance identifier. The identifier can analyze the transformer average error and sampling accuracy average error in real time during the operation of the electric energy meter, and predict the fault level of the electric energy meter based on these error data.

[0057] Furthermore, step P43-2 of the embodiment of the present application also includes:

[0058] P43-21: Divide the sample transformer average error set, sample sampling accuracy average error set and sample fault operation and maintenance identification result set to obtain the training data set of the integrated identification quantity; P43-22: Use the training data set of the integrated identification quantity to train and obtain the fault operation and maintenance identification branches of the integrated identification quantity respectively.

[0059] Specifically, multiple fault operation and maintenance identification branches are generated by dividing and training the sample data. First, the sample transformer average error set, the sample sampling accuracy average error set, and the sample fault operation and maintenance identification result set are divided. This division process can use cross-validation or random sampling methods to ensure that each training data set contains diverse samples and can represent the distribution of the overall data. Through division, multiple training data sets equal to the number of integrated identification are obtained, and each data set contains enough sample data to train a fault operation and maintenance identification branch.

[0060] Next, multiple training data sets obtained by the above division are used to train different fault operation and maintenance identification branches. Each branch learns based on its corresponding data set and extracts patterns related to the fault level from the average error of the transformer and the average error of the sampling accuracy. Through supervised learning, each branch can identify the fault characteristics under specific conditions and adjust its internal parameters to maximize the ability to predict faults. The training process usually involves machine learning or deep learning algorithms, such as support vector machines, neural networks, gradient boosting trees, etc. These algorithms gradually improve the accuracy and generalization ability of the identification branches by optimizing the loss function and iteratively updating the model parameters.

[0061] During the training process, regularization techniques and data enhancement methods can also be used to prevent model overfitting and improve the generalization performance of the model. Through sufficient training and verification, multiple fault operation and maintenance identification branches are obtained, which can identify electric energy meter faults under different working conditions and provide support for fault diagnosis.

[0062] In summary, the embodiments of the present application have at least the following technical effects:

[0063] The present application collects the operating data of the target electric energy meter under various current, voltage and temperature conditions, records the operating time of each condition, sets the test conditions and number of times, performs multiple tests on the electric energy meter, calculates the average error of the mutual inductor and sampling accuracy, configures the operation and maintenance identification computing power according to the number of tests, performs fault identification based on the error data, generates and integrates the fault operation and maintenance results under multiple conditions, and obtains the overall fault operation and maintenance identification result of the target electric energy meter.

[0064] The technical effect of accurately identifying electric energy meter faults and efficiently operating and maintaining them in complex operating environments has been achieved through multi-dimensional data testing and analysis, combined with operation and maintenance identification computing power configuration and fault operation and maintenance identification algorithms.

[0065] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0066] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0067] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis, characterized in that: The method comprises: Collecting a current and voltage operating condition set and a temperature operating condition set in the target electric energy meter operating environment, and collecting a current and voltage operating duration set and a temperature operating duration set corresponding to the current and voltage operating condition set and the temperature operating condition set; According to the current and voltage operating condition set and the temperature operating condition set, a test condition set is set, and according to the current and voltage operating duration set and the temperature operating duration set, a test number set is set; According to the test condition set and the test number set, the target electric energy meter is tested to obtain a transformer average error set and a sampling accuracy average error set by calculation; According to the test number set, an operation and maintenance identification computing power set is configured, and the electric energy meter fault operation and maintenance identification is performed based on the transformer average error set and the sampling accuracy average error set respectively, to obtain a conditional fault operation and maintenance identification result set, and then the fault operation and maintenance identification result of the target electric energy meter is obtained by fusion.

2. The electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis according to claim 1 is characterized in that: Collecting a current and voltage operating condition set and a temperature operating condition set in the target electric energy meter operating environment, and collecting a current and voltage operating duration set and a temperature operating duration set corresponding to the current and voltage operating condition set and the temperature operating condition set, including: In the target electric energy meter operating environment, continuously monitoring the current and voltage within a preset time period, obtaining a current and voltage information set, and dividing and obtaining a current and voltage operating condition set; Continuously monitoring the operating temperature within the preset time period, obtaining a temperature information set, and dividing to obtain a temperature operating condition set; The operation time of the target electric energy meter under the current and voltage operation condition set is collected respectively to obtain a current and voltage operation duration set; The operating time of the target electric energy meter under the temperature operation condition set is collected respectively to obtain a temperature operation duration set.

3. The electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis according to claim 1 is characterized in that: According to the current and voltage operating condition set and the temperature operating condition set, a test condition set is set, and according to the current and voltage operating duration set and the temperature operating duration set, a test number set is set, including: The current and voltage operating condition set is used as a standard test voltage and current parameter set, and combined with the temperature operating condition set to obtain a test condition set; The test number set is obtained by calculating the ratio of the current voltage operation time and the temperature operation time corresponding to each test condition to the sum of the current voltage operation time set and the temperature operation time set, combined with the total test number.

4. The electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis according to claim 1, characterized in that: According to the test condition set and the test number set, the target electric energy meter is tested, and a transformer average error set and a sampling accuracy average error set are calculated and obtained, including: Obtaining a standard transformer current set, a standard transformer voltage set, a standard current signal set, and a standard voltage signal set under each test condition; The target electric energy meter is tested according to the test condition set and the test number set to obtain multiple test result sets. The transformer average error set and the sampling accuracy average error set are calculated by combining the standard transformer current set, the standard transformer voltage set, the standard current signal set and the standard voltage signal set.

5. The electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis according to claim 1, characterized in that: According to the test number set, the operation and maintenance identification computing power set is configured, including: Obtain the number of integrated identifications for power meter fault operation and maintenance identification; The ratio of each test number to the maximum test number in the test number set is multiplied by the integrated identification number and rounded to obtain the operation and maintenance identification number, which is configured as the operation and maintenance identification computing power to obtain the operation and maintenance identification computing power set.

6. The electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis according to claim 5 is characterized in that: Based on the transformer average error set and the sampling accuracy average error set, the fault operation and maintenance identification of the electric energy meter is performed respectively to obtain a conditional fault operation and maintenance identification result set, and the fault operation and maintenance identification result of the target electric energy meter is obtained by fusion, including: According to the integrated identification quantity, training the fault operation and maintenance identification branches of the integrated identification quantity to obtain a fault operation and maintenance identifier; According to the operation and maintenance identification computing power set, the mutual inductor average error set and the sampling accuracy average error set are respectively combined and input into the fault operation and maintenance identification branch of the operation and maintenance identification computing power set, and the fault operation and maintenance identification result sets of multiple branches are identified to obtain, and the mean is calculated to obtain the conditional fault operation and maintenance identification result set; According to the size of each operation and maintenance identification computing power, a fusion weighted calculation is performed on the conditional fault operation and maintenance identification result set to obtain the fault operation and maintenance identification result of the target electric energy meter.

7. The electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis according to claim 6 is characterized in that: According to the integrated identification quantity, training the fault operation and maintenance identification branch of the integrated identification quantity to obtain a fault operation and maintenance identifier includes: According to the operation and maintenance data records of the electric energy meter, the sample transformer average error set and the sample sampling accuracy average error set are collected, and according to the fault level of the electric energy meter under different sample transformer average errors and sample sampling accuracy average errors, the sample fault operation and maintenance identification result set is marked; The sample mutual inductor average error set, the sample sampling accuracy average error set and the sample fault operation and maintenance identification result set are used to train the fault operation and maintenance identification branches of the integrated identification quantity to obtain a fault operation and maintenance identifier.

8. The electric energy meter fault operation and maintenance method based on multi-dimensional data test and analysis according to claim 7 is characterized in that: The sample transformer average error set, the sample sampling accuracy average error set and the sample fault operation and maintenance identification result set are used to train the fault operation and maintenance identification branch of the integrated identification quantity, including: Dividing the sample transformer average error set, the sample sampling accuracy average error set and the sample fault operation and maintenance identification result set to obtain the training data set of the integrated identification quantity; The training data sets of the integrated identification quantity are used to respectively train and obtain the fault operation and maintenance identification branches of the integrated identification quantity.

Images