Intelligent early warning method and system for digital electric energy meter

Abstract

The invention discloses an intelligent early warning method and system for a digital electric energy meter, and relates to the field of power systems. The problem that an existing electric energy meter is low in early warning accuracy is mainly solved. The method mainly comprises the steps of obtaining a transformation level of a to-be-warned electric energy meter in a voltage network, and collecting voltage data at different transformation nodes and instrument data of the electric energy meter; performing anomaly screening on the instrument data according to a preset time interval and the dynamic parameters of the electric energy meter; if abnormal data of the electric energy meters are screened out, calling use labels of associated electric energy meters having an association relationship with the electric energy meters and adjacent electric energy meters having an adjacent relationship, performing classification processing on the transformation levels, the voltage data and the instrument data on the basis of a classification model which completes model training, and determining a classification result; and carrying out weight integration on the use label and the classification result, determining an early warning classification of the electric energy meter, and sending early warning information according to the early warning classification. The method is used for early warning abnormity of the digital electric energy meter.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power systems, and in particular to an intelligent early warning method and system for a digital electric energy meter. Background Art

[0002] In the power system, smart energy meters, as important power transmission detection equipment in the power system, need to detect abnormal conditions in a timely manner to ensure normal use. At present, the existing early warning method of energy meters usually uses threshold judgment on the data collected by the energy meters as the basis for whether the energy meters are operating normally. However, due to the different environments in which the energy meters are located, such as the humidity level and temperature, the operation of individual energy meters will be greatly affected, and the unstable voltage transformation of the transmission line will also cause abnormal detection of the energy meters. Therefore, a digital energy meter intelligent early warning method is urgently needed to solve the above problems. Summary of the invention

[0003] In view of this, the present invention provides a digital electric energy meter intelligent early warning method and system, the main purpose of which is to solve the problem that in the existing electric energy meter detection, there are many abnormal detections of electric energy meters and the accuracy of electric energy meter detection is low.

[0004] According to one aspect of the present invention, there is provided a digital electric energy meter intelligent early warning method, comprising:

[0005] Obtaining the voltage transformation level of the electric energy meter to be warned in the voltage network, and collecting voltage data at different voltage transformation nodes and instrument data of the electric energy meter;

[0006] Screening the meter data for abnormalities according to a preset time interval and a dynamic parameter of the electric energy meter, wherein the dynamic parameter is generated based on the historical record data of the electric energy meter and the meter status information;

[0007] If abnormal data of the electric energy meter is screened, the usage labels of the associated electric energy meters associated with the electric energy meter and the adjacent electric energy meters in an adjacent relationship are retrieved, and the transformer level, the voltage data, and the meter data are classified based on the classification model for which model training has been completed to determine the classification result;

[0008] The usage tag and the classification result are weighted and integrated to determine the warning classification of the electric energy meter, and the warning information is sent according to the warning classification.

[0009] Furthermore, the abnormality screening of the meter data according to the preset time interval and the dynamic parameters of the electric energy meter includes:

[0010] Determining an instrument frequency of the instrument data, and determining a screening segment of the instrument data according to the preset time interval and the instrument frequency;

[0011] Retrieving the updated dynamic parameters, and determining the data consumption value of the screening section according to the dynamic parameters, wherein the data consumption value is used to characterize the abnormal consumption of data by the electric energy meter;

[0012] If the data consumption value is greater than the average value of the screening segment, the data in the screening segment is determined to be abnormal data.

[0013] Furthermore, before calling the updated dynamic parameters, the method further includes:

[0014] Acquire historical record data and updated meter status information of the electric energy meter, where the meter status information is collected by the inspection terminal;

[0015] The instrument status information is quantized to obtain a quantized instrument quantization value, and the historical record data and dynamic parameters corresponding to the instrument quantization value are calculated based on a preset dynamic function.

[0016] Furthermore, before retrieving the usage tags of the associated electric energy meters associated with the electric energy meter and the adjacent electric energy meters in an adjacent relationship, the method further includes:

[0017] Searching for multiple related electric energy meters that match the distance range, device information, and inspection task group of the electric energy meter;

[0018] The relevance options and the adjacency options of the related electric energy meters are obtained, and the usage tags used by the inspection terminal to mark the relevance options and the related electric energy meters corresponding to the adjacency options are obtained.

[0019] Furthermore, the weighting of the usage tag and the classification result to determine the warning classification of the electric energy meter, and sending the warning information according to the warning classification includes:

[0020] Obtaining a weight parameter matching the voltage transformation level, and numerically weighting the usage label and the classification result according to the weight parameter to obtain an integrated result;

[0021] According to the integration result, query the warning list of the transformer level to determine the warning classification, and the different transformer levels match different warning lists;

[0022] The early warning terminal corresponding to the voltage transformation level is queried, and the early warning information generated according to the early warning classification is sent to the early warning terminal.

[0023] Furthermore, before the classification model based on the completed model training performs classification processing on the transformer level, the voltage data, and the instrument data and determines the classification result, the method further includes:

[0024] Constructing a multi-level decision tree model and obtaining a classification training sample set, wherein the classification training sample set includes transformer level samples, voltage samples, and instrument samples with different classification labels;

[0025] The multi-level decision tree model is trained based on the classification training sample set to obtain a trained classification model, and the binary classification decision weights in the multi-level decision tree model are dynamically adjusted based on the sample classification in the transformer grade sample.

[0026] According to another aspect of the present invention, there is provided a digital electric energy meter intelligent early warning system, comprising:

[0027] An acquisition module is used to acquire the voltage transformation level of the electric energy meter to be warned in the voltage network, and collect voltage data at different voltage transformation nodes and instrument data of the electric energy meter;

[0028] A screening module, used to screen the meter data for abnormality according to a preset time interval and a dynamic parameter of the electric energy meter, wherein the dynamic parameter is generated based on the historical record data of the electric energy meter and the meter status information;

[0029] A retrieval module, for retrieving the usage labels of the associated electric energy meters associated with the electric energy meter and the adjacent electric energy meters in an adjacent relationship if the abnormal data of the electric energy meter is screened, and classifying the transformer level, the voltage data, and the meter data based on the classification model for which model training has been completed to determine the classification result;

[0030] The sending module is used to weight the usage tag and the classification result, determine the warning classification of the electric energy meter, and send the warning information according to the warning classification.

[0031] Furthermore, the screening module includes:

[0032] A first determining unit, configured to determine an instrument frequency of the instrument data, and determine a screening segment of the instrument data according to the preset time interval and the instrument frequency;

[0033] A retrieving unit, used to retrieve the updated dynamic parameters, and determine the data consumption value of the screening section according to the dynamic parameters, wherein the data consumption value is used to characterize the abnormal consumption of data by the electric energy meter;

[0034] The second determination method unit is used to determine that the data in the screening segment is abnormal data if the data consumption value is greater than the average value of the screening segment.

[0035] Furthermore, the system further comprises:

[0036] The acquisition module is further used to acquire the historical record data of the electric energy meter and the updated meter status information, wherein the meter status information is collected by the inspection terminal;

[0037] The calculation module is used to quantify the instrument status information to obtain the quantized instrument quantization value, and calculate the historical record data and the dynamic parameters corresponding to the instrument quantization value based on a preset dynamic function.

[0038] Furthermore, the system further comprises:

[0039] A search module, used to search for multiple related electric energy meters that match the distance range, device information and inspection task group of the electric energy meter;

[0040] The acquisition module is further used to acquire the relevance options and the adjacency options of the relevant electric energy meters, and to acquire usage tags used by the inspection terminal to mark the relevant electric energy meters corresponding to the relevance options and the adjacency options.

[0041] Furthermore, the sending module includes:

[0042] A weighting unit, used to obtain a weight parameter matching the transformer level, and numerically weight the usage label and the classification result according to the weight parameter to obtain an integrated result;

[0043] A query unit, used for querying the warning list of the transformer level according to the integration result, determining the warning classification, and matching different transformer levels with different warning lists;

[0044] A generating unit is used to query the warning terminal corresponding to the voltage transformation level, and send the warning information generated according to the warning classification to the warning terminal.

[0045] Furthermore, the system further comprises:

[0046] A construction module is used to construct a multi-level decision tree model and obtain a classification training sample set, wherein the classification training sample set includes transformer level samples, voltage samples, and instrument samples with different classification labels;

[0047] A training module is used to train the multi-level decision tree model based on the classification training sample set to obtain a trained classification model, and the binary classification decision weights in the multi-level decision tree model are dynamically adjusted based on the sample classification in the transformer grade sample.

[0048] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned digital electric energy meter intelligent early warning method.

[0049] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0050] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned digital electric energy meter intelligent early warning method.

[0051] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0052] The present invention provides a digital electric energy meter intelligent early warning method and system. The embodiment of the present invention obtains the transformer level of the electric energy meter to be warned in the voltage network, and collects the voltage data at different transformer nodes and the meter data of the electric energy meter; the meter data is screened for abnormality according to the preset time interval and the dynamic parameters of the electric energy meter, and the dynamic parameters are generated based on the historical record data of the electric energy meter and the meter status information; if the abnormal data of the electric energy meter is screened, the use tags of the associated electric energy meter with an associated relationship and the adjacent electric energy meter with an adjacent relationship are retrieved, and the transformer level, voltage data, and meter data are classified and processed based on the classification model of the completed model training to determine the classification result; the use tag and the classification result are weighted and integrated to determine the early warning classification of the electric energy meter, and the early warning information is sent according to the early warning classification. The screening of abnormal data can be combined with the actual state of the electric energy meter, and the early warning classification can be identified in combination with the manual inspection results and the monitoring data analysis results, which can avoid the abnormality of the electric energy meter monitoring, thereby improving the accuracy of the electric energy meter early warning.

[0053] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0055] Figure 1 A flow chart of an intelligent early warning method for a digital electric energy meter provided by an embodiment of the present invention is shown;

[0056] Figure 2 Another flow chart of an intelligent early warning method for a digital electric energy meter provided by an embodiment of the present invention is shown;

[0057] Figure 3 A block diagram of a digital electric energy meter intelligent early warning system provided by an embodiment of the present invention is shown;

[0058] Figure 4 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0059] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0060] The embodiment of the present invention provides a digital electric energy meter intelligent early warning method, such as Figure 1 As shown, the method includes:

[0061] 101. Obtain the voltage transformation level of the electric energy meter to be warned in the voltage network, and collect voltage data at different voltage transformation nodes and instrument data of the electric energy meter.

[0062] In an embodiment of the present invention, the electric energy meter to be warned is a digital electric energy meter that needs to be monitored and warned for abnormalities, and may be a multifunctional electric energy meter, an intelligent electric energy meter, etc. The voltage network is the power supply network where the electric energy meter is located, and its transformation levels include low voltage, medium voltage and high voltage. In order to realize the monitoring of the electric energy meter, the voltage data of different transformers (transformation nodes) in the voltage network and the data collected by the electric energy meter to be warned, that is, the instrument data, are collected. Among them, the instrument data may include total power consumption, time-sharing power consumption, current, voltage, frequency, power, and current, voltage harmonic components, harmonic fundamental variable rate, etc. used to evaluate the quality of electric energy, which are not specifically limited in the embodiment of the present invention.

[0063] 102. Screen the meter data for abnormalities according to a preset time interval and dynamic parameters of the electric energy meter.

[0064] In the embodiment of the present invention, in order to determine whether the electric energy meter has an abnormal operation, the collected instrument data is used as the basis for judging whether it is abnormal or not, and the instrument data is identified according to the preset time interval and the dynamic parameters of the electric energy meter. When the abnormal data of the electric energy meter is screened, the electric energy meter is preliminarily judged to be abnormal. Specifically, the instrument data can be extracted based on the preset time interval, and the abnormal part of the instrument data can be identified based on the dynamic parameters to screen out the abnormal data. Among them, the preset time interval can be 10 minutes, 30 minutes, 1 natural day, and can also be customized according to actual application requirements, which is not specifically limited in the embodiment of the present invention. The dynamic parameters are parameters generated based on the historical record data and the instrument status information of the electric energy meter, that is, the dynamic parameters can characterize the dynamic changes of the current state of the electric energy meter compared with the historical state and the real-time equipment status. The instrument data is screened according to the dynamic parameters, and the data can be judged in combination with the actual state of the electric energy meter to avoid misjudgment of abnormal data, thereby improving the accuracy of abnormal screening.

[0065] 103. If abnormal data of the electric energy meter is screened out, the usage labels of the associated electric energy meters that are associated with the electric energy meter and the adjacent electric energy meters that are adjacent to each other are retrieved, and the transformer level, the voltage data, and the meter data are classified based on the classification model that has completed model training to determine the classification result.

[0066] In the embodiment of the present invention, when abnormal data is screened from the meter data of the electric energy meter, it indicates that the electric energy meter is abnormal and has the necessity of early warning. Then, it is necessary to classify the transformer level, voltage data, and meter data through the classification model that has completed model training, and predict the classification result of the electric energy meter to make a finer granularity judgment on the state of the electric energy meter. Among them, the classification model that has completed model training is obtained by training based on samples corresponding to different transformer levels, different voltage data, and different meter data samples. The classification model can be a binary classification model or a multi-classification model. The model can be constructed based on a random forest model, a binary tree model, or a support vector machine model, which is not specifically limited in the embodiment of the present invention. In addition, it is also necessary to retrieve the usage labels of the associated electric energy meter and the adjacent electric energy meter. Among them, the associated electric energy meter can be an electric energy meter that has a certain association with the current electric energy meter, such as an electric energy meter corresponding to the same user or the same payment contact person, and an electric energy meter belonging to the same inspection task group or the same station area as the current electric energy meter. The adjacent electric energy meters are electric energy meters that are adjacent to the current electric energy meter in the distance dimension, such as all electric energy meters within a preset radius with the current electric energy meter as the center.

[0067] 104. Integrate the weights of the usage tag and the classification result to determine the warning classification of the electric energy meter, and send warning information according to the warning classification.

[0068] In an embodiment of the present invention, a usage label is a numerical value marked based on the inspection results of the inspection personnel, and is used to characterize the energy consumption, equipment status, data collection accuracy and other operating conditions of the corresponding electric energy meter. In the process of integrating the classification results with the usage labels, the classification results need to be converted into numerical values ​​and then integrated. According to the integrated numerical values, the warning classification is matched from the correspondence between the pre-constructed numerical values ​​and the warning classification to achieve the warning of the electric energy meter monitoring. By integrating the usage labels marked based on the inspection results of adjacent and related electric energy meters with the classification results of the electric energy meters determined based on data analysis, a comprehensive analysis between the electric energy meter to be warned and its related electric energy meters is achieved, and the results of the manual monitoring dimension and the intelligent analysis dimension of the monitoring data are also achieved, which can effectively avoid the abnormality of the electric energy meter monitoring, thereby improving the accuracy of the electric energy meter warning.

[0069] It should be noted that the receiving end (warning terminal) of the warning information can be different according to the transformer level or warning classification. For example, the warning classification represents the warning of medium and low levels, and the receiving end can be the portable terminal of the grassroots inspection personnel or the fixed terminal of the inspection station; the warning classification represents the high-level warning, and the receiving terminal can be the terminal of the inspection team leader; the warning classification represents the high-level warning, and the receiving end can be the warning system server, the terminal equipment of the main control center, etc. According to the different warning classifications, selecting different receiving ends for warning information can improve the effectiveness and timeliness of the warning, thereby improving the efficiency of the maintenance of the electric energy meter.

[0070] In one embodiment of the present invention, for further explanation and limitation, as Figure 2 As shown, the step of screening the meter data for abnormality according to the preset time interval and the dynamic parameters of the electric energy meter includes:

[0071] 201. Determine an instrument frequency of the instrument data, and determine a screening segment of the instrument data according to the preset time interval and the instrument frequency.

[0072] 202. Retrieve the updated dynamic parameters, and determine the data consumption value of the screening segment according to the dynamic parameters.

[0073] 203. If the data consumption value is greater than the average value of the screening segment, determine that the data in the screening segment is abnormal data.

[0074] In an embodiment of the present invention, the instrument frequency is the frequency of collecting instrument data. The screening segment is a data segment extracted from the instrument data for determining whether the instrument data is abnormal. The length of each data segment extracted from the instrument data can be determined according to the quotient of the preset time interval and the instrument frequency, the instrument data is segmented according to this length, and the screening segment is extracted from the segmented data end according to the preset interval or randomly. Among them, the length of each data segment can be the quotient of the preset time interval and the instrument frequency, or the product of the quotient and a correction coefficient less than 1. For example, if the preset time interval is 10 minutes, the instrument frequency is 20s / time, and the correction coefficient is 0.5, then the data segment length is 10min×60s÷20s / time×0.5=15 times, that is, every 15 sampling data of the instrument data is segmented into one data segment.

[0075] Due to the influence of factors such as the accuracy of the energy meter equipment, ambient temperature and humidity, and electromagnetic interference, the data collected by the energy meter is not completely usable data, and there are also some error data. The data consumption value is the error data with no reference value in each screening segment, that is, the data consumption value is used to characterize the abnormal consumption of data by the energy meter. The dynamic parameter is used to characterize the error degree of the data collected by the energy meter. According to the dynamic data, the data consumption value of each screening segment can be calculated. The formula is:

[0076] (1);

[0077] in, is the data consumption value, is the amount of data contained in the filter segment, is a dynamic parameter, is a coefficient greater than or equal to 1, and can be 1.

[0078] If the data consumption value of the current filtering segment is greater than the average value of the data consumption values ​​of other data filtering segments of this electric energy meter, the data in this filtering segment is determined to be abnormal data, indicating that the current electric energy meter has an abnormal situation.

[0079] In one embodiment of the present invention, for further explanation and limitation, before calling the updated dynamic parameter, the method further includes:

[0080] Obtaining historical record data of the electric energy meter and updated meter status information;

[0081] The instrument status information is quantized to obtain a quantized instrument quantization value, and the historical record data and dynamic parameters corresponding to the instrument quantization value are calculated based on a preset dynamic function.

[0082] In the embodiment of the present invention, the historical record data of the electric energy meter includes the historical abnormal data, historical warning data, historical maintenance data and other numerical values ​​that can characterize the operation quality of the meter, and is updated in real time with the occurrence, warning and maintenance of abnormal conditions of the electric energy meter. The inspection terminal is a terminal device for recording the meter status of the inspected electric energy meter, that is, the meter status information is collected by the inspection terminal, and the meter status information responds to the data input of the inspection terminal to make information updates. In order to calculate the dynamic parameters, it is necessary to quantify the meter status information. The meter status information includes multiple status levels such as very good, good, general, and poor, and different status scores are assigned to different status levels to convert different status levels into numerical representations. For example, the quantization value of the meter with a status of "very good" is 10, the quantization value of the meter with a status of "good" is 8, the quantization value of the meter with a status of "general" is 6, and the quantization value of the meter with a status of "poor" is 5. The preset dynamic function is a function that changes dynamically with the update of the historical record data and the meter quantization value, and the calculation result of the function is the dynamic parameter. The larger the value of the dynamic parameter, the worse the operation quality of the electric energy meter. The formula is:

[0083] (2);

[0084] in, is a dynamic parameter, is the number of historical anomalies, is the number of historical warnings, Indicates the number of historical maintenance times, a<b<c<1, a+b+c=1, Quantize the value for the meter.

[0085] It should be noted that by determining dynamic parameters based on the historical data of the electric energy meter and the real-time updated meter status, and calculating the data consumption value based on the dynamic parameters, the power data collection status of the electric energy meter can be accurately evaluated. In the process of abnormal data screening of the meter data, the influence of the real-time status of the electric energy meter can be eliminated, thereby improving the accuracy of abnormal data screening.

[0086] In one embodiment of the present invention, for further explanation and limitation, before calling the usage tags of the associated electric energy meter having an associated relationship with the electric energy meter and the adjacent electric energy meter having an adjacent relationship, the method further includes:

[0087] Searching for multiple related electric energy meters that match the distance range, device information, and inspection task group of the electric energy meter;

[0088] The relevance options and the adjacency options of the related electric energy meters are obtained, and the usage tags used by the inspection terminal to mark the relevance options and the related electric energy meters corresponding to the adjacency options are obtained.

[0089] In an embodiment of the present invention, the related electric energy meters include associated electric energy meters and adjacent electric energy meters. In order to determine the related electric energy meters of the electric energy meter, the search is performed according to the distance range, equipment information and inspection task group corresponding to the current electric energy meter, and the electric energy meter that meets any of the above items is used as the related electric energy meter. Among them, the distance range is the distance range pre-configured for the current electric energy meter, and the distance ranges corresponding to different electric energy meters may be different. The electric energy meter whose distance from the current electric energy meter meets the corresponding distance range is determined as the adjacent electric energy meter of the corresponding proximity option. The equipment information includes the model and equipment type of the electric energy meter. The electric energy meter with the same equipment model or the same equipment type as the current electric energy meter can be determined as the associated electric energy meter of the corresponding correlation option. In addition, the electric energy meter belonging to the same inspection task group as the current electric energy meter is also determined as the associated electric energy meter of the corresponding correlation option. Among them, the inspection task group includes the task of inspecting multiple electric energy meters within a certain period of time. For example, all electric energy meters inspected within a natural day constitute an inspection task group.

[0090] After inspecting the energy meter, the inspector will mark the energy meter with a usage tag through the inspection terminal. The usage tag is used to characterize the usage of the corresponding energy meter. For example, the usage is divided into 10 levels from 1 to 10. The larger the value of the usage tag, the better the usage of the energy meter. The inspector can determine the value of the usage tag based on the age of the energy meter, the number of failures, and the detection data. Therefore, after determining the adjacent energy meters and the associated energy meters, their usage tags can be obtained.

[0091] In one embodiment of the present invention, for further explanation and limitation, the weighting integration of the usage tag and the classification result, determining the warning classification of the electric energy meter, and sending the warning information according to the warning classification includes:

[0092] Obtaining a weight parameter matching the voltage transformation level, and numerically weighting the usage label and the classification result according to the weight parameter to obtain an integrated result;

[0093] According to the integration result, query the warning list of the transformer level to determine the warning classification;

[0094] The early warning terminal corresponding to the voltage transformation level is queried, and the early warning information generated according to the early warning classification is sent to the early warning terminal.

[0095] In the embodiment of the present invention, different transformer levels are pre-configured with different weight parameters, and the weight parameters include the weight ratio of the usage label and the classification result. The integration result is the sum of the usage label multiplied by the corresponding weight and the classification result multiplied by the corresponding weight. Since the usage label is a mark for manual inspection, the manual inspection equipment in the high-voltage section is more susceptible to interference, and its detection accuracy is lower than that in the low-voltage section. Therefore, the weight of the usage label is set lower than the classification result in the high-voltage section, and the weight of the usage label is set higher than the classification result in the low-voltage section. For example, the weight ratio of the transformer level is 4:6 for high voltage; the weight ratio of the transformer level is 5:5 for medium voltage; and the weight ratio of the transformer level is 6:4 for low voltage. Of course, it can also be customized according to actual application requirements, and the embodiment of the present invention does not make specific limitations.

[0096] After obtaining the integration result, query the warning list that matches the voltage level of the current electric energy meter, and match the corresponding warning classification from the warning list according to the integration result. Among them, different transformer levels match different warning lists, that is, the corresponding relationship between the integration results and the warning classification in the warning list corresponding to different voltage levels is different. For example, for the same integration result, the warning classification matched in the warning list corresponding to the high voltage level is a first-level warning, but the warning classification matched in the warning list corresponding to the low voltage level is a third-level warning. Thereby, the fine identification of the warning classification is realized and the accuracy of the warning is improved. Since the urgency of intervention for abnormal electric energy meters of different transformer levels is different, the personnel responsible for monitoring and operation and maintenance may also be different. Therefore, the warning terminals corresponding to different transformer levels are also different. It is necessary to determine the warning terminal according to the transformer level, and then generate warning information according to the warning classification to realize the warning. Among them, the warning information includes one or more of the warning frequency, warning time, and target warning terminal. The warning frequency is the sending frequency of the warning prompt. If the warning level of the warning classification is higher, the sending frequency is higher. The warning time is the time point for sending the warning prompt. For warnings of lower levels, they can be sent according to the working time of the warning terminal instead of sending them immediately. The target warning terminal is one or more terminals selected from the warning terminals corresponding to the transformer level according to the importance of the warning classification.

[0097] In one embodiment of the present invention, for further explanation and limitation, the classification model based on the completed model training performs classification processing on the transformer level, the voltage data, and the instrument data, and before determining the classification result, the method further includes:

[0098] Build a multi-level decision tree model and obtain a classification training sample set;

[0099] The multi-level decision tree model is trained based on the classification training sample set to obtain a trained classification model.

[0100] In an embodiment of the present invention, the classification model is a multi-level decision tree model, and is trained based on a classification training sample set. The classification training sample set includes transformer level samples, voltage samples, and instrument data samples with different classification labels. The binary classification decision weights in the multi-level decision tree model are dynamically adjusted based on the sample classification in the transformer level sample. In other words, the decision weights of the binary tree are determined according to the number of classifications of the transformer level sample. For example, when the number of classifications is 3, the decision weight is set to 1 / 3, and when the number of classifications is 5, the decision weight is set to 1 / 5. The size of the decision weight determines the depth of the binary tree, which in turn affects the robustness of the model. By dynamically adjusting the decision weight according to the sample classification in the transformer level sample, the more refined the sample classification of the transformer level sample is, the deeper the decision tree can be, thereby improving the accuracy of model recognition.

[0101] In an embodiment of the present invention, for further explanation and limitation, after sending the warning information according to the warning classification, the method further includes:

[0102] Record the number of warnings, warning terminal information and warning transformer level to generate historical warning information; according to the preset voltage transmission conditions, count the abnormal warning status of the electric energy meter in the historical warning information; if the abnormal warning status matches the transmission warning condition, generate transmission warning information matching the electric energy meter warning.

[0103] In an embodiment of the present invention, the preset voltage transmission conditions may include time, power, current, etc. The statistical period may be determined according to whether the operation time of the electric energy meter reaches the preset interval length, whether the power statistics reach the threshold or whether the current value reaches the threshold, and the abnormal warning state of the electric energy meter in the historical warning information within this time period is counted. The abnormal warning state of the electric energy meter can be judged based on the number of warning electric energy meters in the same transformer level and the time range as conditions. If the number of warning electric energy meters in the same transformer level is greater than the number threshold in the transmission warning condition, it indicates that the power supply network of this transformer level is abnormally transmitted, and transmission warning information is generated. If the warning time interval of the warning electric energy meter in the same transformer level is less than the time range, that is, the power supply network of this transformer level is frequently warned, then transmission warning information is generated. Among them, the number threshold and the time range can be customized according to actual application requirements, and the embodiment of the present invention does not make specific limitations.

[0104] The present invention provides a digital electric energy meter intelligent early warning method. The embodiment of the present invention obtains the transformer level of the electric energy meter to be warned in the voltage network, and collects the voltage data at different transformer nodes and the meter data of the electric energy meter; the meter data is screened for abnormality according to the preset time interval and the dynamic parameters of the electric energy meter, and the dynamic parameters are generated based on the historical record data of the electric energy meter and the meter status information; if the abnormal data of the electric energy meter is screened, the use tags of the associated electric energy meter with an associated relationship and the adjacent electric energy meter with an adjacent relationship are retrieved, and the transformer level, voltage data, and meter data are classified and processed based on the classification model of the completed model training to determine the classification result; the use tags and the classification results are weighted and integrated to determine the early warning classification of the electric energy meter, and the early warning information is sent according to the early warning classification. The screening of abnormal data can be combined with the actual state of the electric energy meter, and the early warning classification can be identified in combination with the manual inspection results and the monitoring data analysis results, which can avoid the abnormality of the electric energy meter monitoring, thereby improving the accuracy of the electric energy meter early warning.

[0105] Furthermore, as a response to the above Figure 1 The embodiment of the present invention provides a digital electric energy meter intelligent early warning system, such as Figure 3 As shown, the system includes:

[0106] The acquisition module 31 is used to acquire the voltage transformation level of the electric energy meter to be warned in the voltage network, and collect the voltage data at different voltage transformation nodes and the instrument data of the electric energy meter;

[0107] A screening module 32, for screening the meter data for abnormality according to a preset time interval and a dynamic parameter of the electric energy meter, wherein the dynamic parameter is generated based on the historical record data of the electric energy meter and the meter status information;

[0108] The retrieval module 33 is used for retrieving the usage labels of the associated electric energy meters associated with the electric energy meter and the adjacent electric energy meters in an adjacent relationship if the abnormal data of the electric energy meter is screened, and classifying the transformer level, the voltage data, and the meter data based on the classification model for which model training has been completed to determine the classification result;

[0109] The sending module 34 is used to weight the usage tag and the classification result, determine the warning classification of the electric energy meter, and send the warning information according to the warning classification.

[0110] Furthermore, the screening module 32 includes:

[0111] A first determining unit, configured to determine an instrument frequency of the instrument data, and determine a screening segment of the instrument data according to the preset time interval and the instrument frequency;

[0112] A retrieving unit, used to retrieve the updated dynamic parameters, and determine the data consumption value of the screening section according to the dynamic parameters, wherein the data consumption value is used to characterize the abnormal consumption of data by the electric energy meter;

[0113] The second determination method unit is used to determine that the data in the screening segment is abnormal data if the data consumption value is greater than the average value of the screening segment.

[0114] Furthermore, the system further comprises:

[0115] The acquisition module 31 is further used to acquire the historical record data of the electric energy meter and the updated meter status information, wherein the meter status information is collected by the inspection terminal;

[0116] The calculation module is used to quantify the instrument status information to obtain the quantized instrument quantization value, and calculate the historical record data and the dynamic parameters corresponding to the instrument quantization value based on a preset dynamic function.

[0117] Furthermore, the system further comprises:

[0118] A search module, used to search for multiple related electric energy meters that match the distance range, device information and inspection task group of the electric energy meter;

[0119] The acquisition module 31 is further used to acquire the relevance options and the proximity options of the relevant electric energy meters, and acquire usage tags used by the inspection terminal to mark the relevant electric energy meters corresponding to the relevance options and the proximity options.

[0120] Furthermore, the sending module 34 includes:

[0121] A weighting unit, used to obtain a weight parameter matching the transformer level, and numerically weight the usage label and the classification result according to the weight parameter to obtain an integrated result;

[0122] A query unit, used for querying the warning list of the transformer level according to the integration result, determining the warning classification, and matching different transformer levels with different warning lists;

[0123] A generating unit is used to query the warning terminal corresponding to the voltage transformation level, and send the warning information generated according to the warning classification to the warning terminal.

[0124] Furthermore, the system further comprises:

[0125] A construction module is used to construct a multi-level decision tree model and obtain a classification training sample set, wherein the classification training sample set includes transformer level samples, voltage samples, and instrument samples with different classification labels;

[0126] A training module is used to train the multi-level decision tree model based on the classification training sample set to obtain a trained classification model, and the binary classification decision weights in the multi-level decision tree model are dynamically adjusted based on the sample classification in the transformer grade sample.

[0127] The present invention provides a digital electric energy meter intelligent early warning system. The embodiment of the present invention obtains the transformer level of the electric energy meter to be warned in the voltage network, and collects the voltage data at different transformer nodes and the meter data of the electric energy meter; the meter data is screened for abnormality according to the preset time interval and the dynamic parameters of the electric energy meter, and the dynamic parameters are generated based on the historical record data of the electric energy meter and the meter status information; if the abnormal data of the electric energy meter is screened, the use tags of the associated electric energy meter with an associated relationship and the adjacent electric energy meter with an adjacent relationship are retrieved, and the transformer level, voltage data, and meter data are classified and processed based on the classification model of the completed model training to determine the classification result; the use tags and the classification results are weighted and integrated to determine the early warning classification of the electric energy meter, and the early warning information is sent according to the early warning classification. The screening of abnormal data can be combined with the actual state of the electric energy meter, and the early warning classification can be identified in combination with the manual inspection results and the monitoring data analysis results, so as to avoid the abnormality of the electric energy meter monitoring, thereby improving the accuracy of the electric energy meter early warning.

[0128] According to one embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the digital electric energy meter intelligent early warning method in any of the above method embodiments.

[0129] Figure 4 A schematic diagram of the structure of a terminal provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the terminal.

[0130] like Figure 4 As shown, the terminal may include: a processor (processor) 402 , a communication interface (Communications Interface) 404 , a memory (memory) 406 , and a communication bus 408 .

[0131] The processor 402 , the communication interface 404 , and the memory 406 communicate with each other via a communication bus 408 .

[0132] The communication interface 404 is used for network communication with other devices such as a client or other servers.

[0133] The processor 402 is used to execute the program 410, and specifically can execute the relevant steps in the above-mentioned embodiment of the digital electric energy meter intelligent early warning method.

[0134] Specifically, the program 410 may include program codes, which include computer operation instructions.

[0135] The processor 402 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the terminal may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0136] The memory 406 is used to store the program 410. The memory 406 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0137] The program 410 may be specifically configured to enable the processor 402 to perform the following operations:

[0138] Obtaining the voltage transformation level of the electric energy meter to be warned in the voltage network, and collecting voltage data at different voltage transformation nodes and instrument data of the electric energy meter;

[0139] Screening the meter data for abnormalities according to a preset time interval and a dynamic parameter of the electric energy meter, wherein the dynamic parameter is generated based on the historical record data of the electric energy meter and the meter status information;

[0140] If abnormal data of the electric energy meter is screened, the usage labels of the associated electric energy meters associated with the electric energy meter and the adjacent electric energy meters in an adjacent relationship are retrieved, and the transformer level, the voltage data, and the meter data are classified based on the classification model for which model training has been completed to determine the classification result;

[0141] The usage tag and the classification result are weighted and integrated to determine the warning classification of the electric energy meter, and the warning information is sent according to the warning classification.

[0142] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing system, they can be concentrated on a single computing system, or distributed on a network composed of multiple computing systems, and optionally, they can be implemented by a program code executable by a computing system, so that they can be stored in a storage system and executed by the computing system, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0143] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A digital electric energy meter intelligent early warning method, characterized in that: include: Obtaining the voltage transformation level of the electric energy meter to be warned in the voltage network, and collecting voltage data at different voltage transformation nodes and instrument data of the electric energy meter; Screening the meter data for abnormalities according to a preset time interval and a dynamic parameter of the electric energy meter, wherein the dynamic parameter is generated based on the historical record data of the electric energy meter and the meter status information; If abnormal data of the electric energy meter is screened, the usage labels of the associated electric energy meters associated with the electric energy meter and the adjacent electric energy meters in an adjacent relationship are retrieved, and the transformer level, the voltage data, and the meter data are classified based on the classification model for which model training has been completed to determine the classification result; The usage tag and the classification result are weighted and integrated to determine the warning classification of the electric energy meter, and the warning information is sent according to the warning classification.

2. The method according to claim 1, characterized in that The abnormality screening of the meter data according to the preset time interval and the dynamic parameters of the electric energy meter includes: Determining an instrument frequency of the instrument data, and determining a screening segment of the instrument data according to the preset time interval and the instrument frequency; Retrieving the updated dynamic parameters, and determining the data consumption value of the screening section according to the dynamic parameters, wherein the data consumption value is used to characterize the abnormal consumption of data by the electric energy meter; If the data consumption value is greater than the average value of the screening segment, the data in the screening segment is determined to be abnormal data.

3. The method according to claim 2, characterized in that Before calling the updated dynamic parameters, the method further includes: Acquire historical record data and updated meter status information of the electric energy meter, where the meter status information is collected by the inspection terminal; The instrument status information is quantized to obtain a quantized instrument quantization value, and the historical record data and dynamic parameters corresponding to the instrument quantization value are calculated based on a preset dynamic function.

4. The method according to claim 1, characterized in that: Before retrieving the usage tags of the associated electric energy meters associated with the electric energy meter and the adjacent electric energy meters in an adjacent relationship, the method further includes: Searching for multiple related electric energy meters that match the distance range, device information, and inspection task group of the electric energy meter; The relevance options and the adjacency options of the related electric energy meters are obtained, and the usage tags used by the inspection terminal to mark the relevance options and the related electric energy meters corresponding to the adjacency options are obtained.

5. The method according to claim 1, characterized in that The step of weighting the usage tag and the classification result to determine the warning classification of the electric energy meter and sending the warning information according to the warning classification includes: Obtaining a weight parameter matching the voltage transformation level, and numerically weighting the usage label and the classification result according to the weight parameter to obtain an integrated result; According to the integration result, query the warning list of the transformer level to determine the warning classification, and the different transformer levels match different warning lists; The early warning terminal corresponding to the voltage transformation level is queried, and the early warning information generated according to the early warning classification is sent to the early warning terminal.

6. The method according to claim 1, characterized in that The classification model based on the completed model training performs classification processing on the transformer level, the voltage data, and the instrument data, and before determining the classification result, the method further includes: Constructing a multi-level decision tree model and obtaining a classification training sample set, wherein the classification training sample set includes transformer level samples, voltage samples, and instrument samples with different classification labels; The multi-level decision tree model is trained based on the classification training sample set to obtain a trained classification model, and the binary classification decision weights in the multi-level decision tree model are dynamically adjusted based on the sample classification in the transformer grade sample.

7. A digital electric energy meter intelligent early warning system, characterized in that: include: An acquisition module is used to acquire the voltage transformation level of the electric energy meter to be warned in the voltage network, and collect voltage data at different voltage transformation nodes and instrument data of the electric energy meter; A screening module, used to screen the meter data for abnormality according to a preset time interval and a dynamic parameter of the electric energy meter, wherein the dynamic parameter is generated based on the historical record data of the electric energy meter and the meter status information; A retrieval module, for retrieving the usage labels of the associated electric energy meters associated with the electric energy meter and the adjacent electric energy meters in an adjacent relationship if the abnormal data of the electric energy meter is screened, and classifying the transformer level, the voltage data, and the meter data based on the classification model for which model training has been completed to determine the classification result; The sending module is used to weight the usage tag and the classification result, determine the warning classification of the electric energy meter, and send the warning information according to the warning classification.

8. A storage medium, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the digital electric energy meter intelligent early warning method according to any one of claims 1 to 6.

9. A terminal, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the digital electric energy meter intelligent early warning method according to any one of claims 1-6.

Images