On-site Real-time Detection System for Electric Energy Meter and Its Detection Method

Abstract

The present invention discloses a real-time on-site detection system for electric energy meters and its detection method, which relates to the technical field of electric energy metering detection. The method includes the following steps: regularly and real-time collect relevant data of the electric energy meter through sensors, and at the same time read the electric energy reading of the electric energy meter based on a unit time interval; based on the current and voltage data within a unit time, calculate the theoretical electric energy value using the electric energy calculation formula, and combine it with the measured electric energy value real-time collected within a unit time to obtain the error value data of the electric energy meter, and judge and mark the electric energy meter based on the error threshold; for the electric energy meters with errors within the allowable range, establish an error database and separately record the error data. By combining the offline fault report and the error data analysis method, the present invention can not only improve the accuracy and efficiency of fault diagnosis, but also optimize the operation and maintenance strategy, reduce the maintenance cost, realize more efficient real-time detection and maintenance of electric energy meters, and enhance the practicability and functionality.

Description

Technical Field

[0001] The present invention relates to the technical field of electric energy metering detection. Specifically, it relates to a real-time on-site detection system for electric energy meters and its detection method. Background Art

[0002] A three-phase electric energy meter is an instrument used to measure the electric energy consumption in a three-phase power system. The three-phase power system is a common power supply method in industrial and large commercial buildings, providing power through three phase lines. Since the electric energy meter is related to the calculation of electricity charges, it is necessary to regularly detect the corresponding electric energy meter to ensure the accuracy of power data.

[0003] Chinese Patent with the publication number "CN109557497A" discloses "a new type of digital electric energy meter real-time on-site detection system, including a merging unit, a substation operation digital electric energy meter, a multi-channel digital standard electric energy meter, and a substation monitoring system; data streams conforming to the communication protocol are respectively transmitted from the merging unit to the corresponding channels of the substation operation digital electric energy meter and the multi-channel digital standard electric energy meter". Although it can complete the detection of the electric energy meter, it cannot determine the error change rate of the electric energy meter within a reasonable error range, cannot detect potential faulty electric energy meters, and cannot determine the fault type of abnormal electric energy meters, resulting in problems of low practicability and functionality.

[0004] In view of the above, this case proposes a real-time on-site detection system for electric energy meters and its detection method to solve the above technical problems. Summary of the Invention

[0005] In view of the problems in the related art, the present invention proposes a real-time on-site detection system for electric energy meters and its detection method to overcome the above technical problems existing in the existing related technologies.

[0006] For this purpose, the specific technical solutions adopted by the present invention are as follows:

[0007] A real-time on-site detection method for electric energy meters, the method includes the following steps:

[0008] S1. Regularly and real-time collect relevant data of the electric energy meter through sensors, and at the same time read the electric energy reading of the electric energy meter based on a unit time interval.

[0009] S2. Based on the current and voltage data within a unit time, calculate the theoretical electric energy value using the electric energy calculation formula, and combine it with the measured electric energy value collected in real-time within a unit time to obtain the error value data of the electric energy meter. Based on the error threshold, determine and mark the electric energy meter.

[0010] S3. For electric energy meters with errors within the allowable range, establish an error database, separately record the error data, analyze the error change trend, and mark and observe abnormal electric energy meters in advance.

[0011] S4. Based on the online and offline electricity meter detection data, extract the error value data of the offline electricity meter before and after the actual failure, further analyze the obtained error values, and pre-determine the failure.

[0012] As a preferred embodiment, the S1 includes the following sub-steps:

[0013] S11. Through the current sensor and voltage sensor, regularly collect the relevant data of the on-site electricity meter in real time based on T. The voltage sensor is used to measure the three-phase voltage value at the incoming line of the electricity meter, and the current sensor is used to measure the three-phase current value, and a time stamp is attached.

[0014] S12. Read the electricity meter in real time through the Modbus communication protocol and attach a time stamp. Upload the three-phase voltage value, three-phase current value and the real-time electricity meter reading to the electricity meter database through Wi-Fi.

[0015] As a preferred embodiment, the S2 includes the following sub-steps:

[0016] S21. Through the electricity meter database, extract the three-phase current value and three-phase voltage value of the electricity meter to be detected per unit time based on the time stamp, and obtain the theoretical electricity value of the electricity meter to be detected within the unit time t through the calculation formula. The algorithm formula is:

[0017]

[0018] where W0 represents the theoretical electricity value of the electricity meter to be detected within the unit time t, represents the average three-phase voltage value within the unit time t, represents the average three-phase current value within the unit time t;

[0019] S22. Through the electricity meter database, calculate the measured electricity value of the electricity meter to be detected within the unit time t. The specific algorithm formula is:

[0020] ΔW = W 最终 - W 起始 ;

[0021] where ΔW represents the measured electricity value of the electricity meter to be detected, W 最终 represents the actual reading of the electricity meter at the end of the unit time t, and W 起始 represents the actual reading of the electricity meter at the start of the unit time t;

[0022] S23. Based on the theoretical electricity value and the measured electricity value of the electricity meter to be detected, calculate the error value of the electricity meter to be detected. The specific algorithm formula is:

[0023] W1 = |ΔW - W0|;

[0024] Among them, W1 represents the error value of the electricity meter. By comparing with the error threshold W θ , the operating state of the electricity meter to be measured is determined:

[0025] When W1 ≥ W θ , it means that there is an obvious error in the electricity meter to be measured, and the file of the electricity meter to be measured in the electricity meter database is marked in red;

[0026] When W1 < W θ , it means that there is no obvious error in the electricity meter to be measured, and the error value of this time is recorded in the file.

[0027] As a preferred implementation manner, the S3 includes the following sub-steps:

[0028] S31. In the electricity meter files under the electricity meter database, an error database is established, and the error values within the normal range recorded each time are included in the database;

[0029] S32. Analyze the change trend of the error values in the error database, calculate the error change rate, and mark the files of the electricity meters with abnormal change rates.

[0030] As a preferred implementation manner, the S32 includes the following sub-steps:

[0031] S321. Based on the error data in the error database, calculate the error rate between two adjacent consecutive errors. The specific algorithm formula is:

[0032]

[0033] Among them, e i represents the error rate, E i+1 , E i represent two adjacent error data in the error database. The error rate is statistically calculated and the error change rate Δe i is obtained based on the calculation between two adjacent error rates. The algorithm formula is:

[0034] Δe i = e i+1 - e i ;

[0035] Among them, e i+1 , e i respectively represent two adjacent error rate data;

[0036] S322. Based on the error change rate Δe i , analyze the change tendency of the current file electricity meter. The specific steps are:

[0037] When any one of Δe iand Δe i+1 are both greater than 0, and when Δe i+1 > Δe θ it means that the current error of the watt-hour meter is gradually increasing, and the increasing speed exceeds the change rate threshold Δe θ , mark the current watt-hour meter file with blue.

[0038] As a preferred implementation, S322 includes the following sub-steps:

[0039] S3221. For the watt-hour meter database file marked with blue, based on the change rate threshold Δe θ , perform continuous error detection on the current watt-hour meter. The specific steps are as follows: perform secondary determination on the Δe i+1 data. When 1.25Δe θ ≤Δe i+1 <1.5Δe θ at this time, based on the cycle, obtain the three-phase current data and three-phase voltage data of the relevant watt-hour meter through the sensor, obtain the short-term error change rate, and perform short-term continuous determination on the error change rate. When there are times of short-term error change rates that are all greater than or equal to 1.25Δe θ , mark the relevant watt-hour meter file as red;

[0040] When 1.5Δe θ ≤Δe i+1 <1.75Δe θ at this time, based on the cycle, obtain the three-phase current data and three-phase voltage data of the relevant watt-hour meter through the sensor, obtain the short-term error change rate, and perform short-term continuous determination on the error change rate. When there are times of short-term error change rates that are all greater than or equal to 1.5Δe θ , mark the relevant watt-hour meter file as red;

[0041] When Δe i+1 ≥1.75Δe θ at this time, based on the cycle, obtain the three-phase current data and three-phase voltage data of the relevant watt-hour meter through the sensor, obtain the short-term error change rate, and perform short-term continuous determination on the error change rate. When there are times of short-term error change rates that are all greater than or equal to 1.75Δe θ , mark the relevant watt-hour meter file as red.

[0042] As a preferred implementation, S4 includes the following steps:

[0043] S41. Based on the fault report during offline electricity meter fault detection, retrieve the relevant electricity meter files in the electricity meter database, collect and organize the error data of the electricity meter before and after the fault occurs, and generate an error change curve model;

[0044] S42. Based on the error change curve model, generate curves for the error data in the existing red - marked files, and match the error change curve model through cosine similarity to pre - judge the fixed type of the electricity meter and record it in the relevant electricity meter files.

[0045] As a preferred implementation, S41 includes the following sub - steps:

[0046] S411. Based on the fault report during offline electricity meter fault detection, determine the fault types of the relevant electricity meters, including phase misalignment, frequency mismatch, open - circuit fault, and current transformer fault, and collect the error data groups in the relevant electricity meter files under D groups of the same fault;

[0047] S412. Smooth the data through the moving average method, and draw the error curve through Python. The specific steps are as follows:

[0048]

[0049] Among them, S t is the smoothed value at the t - th moment, E t-i represents the value of the original error data at the i - th position forward from the t - th moment, and n represents the size of the moving average window;

[0050] Visualize the smoothed error data through Python;

[0051] S413. Based on the Euclidean distance method, select similar curve segments for the error curves under D groups of the same fault respectively, cluster the similar segments through hierarchical clustering, select the segment of the clustering center as the representative segment, and synthesize the final fault curve in chronological order of the time stamp.

[0052] An on - site real - time detection system for electricity meters includes a data acquisition module, a real - time detection module, an error trend analysis module, and a fault matching module:

[0053] The data acquisition module includes a current sensor and a voltage sensor, which are used to regularly and real - time collect the relevant data of the electricity meter. At the same time, based on the unit time interval, establish an electricity meter database for the electricity meter readings and transmit the collected data to the subsequent modules;

[0054] The real-time detection module calculates the theoretical power value based on the current and voltage data within a unit time according to the power calculation formula, combines it with the measured power value collected in real time within the unit time to obtain the power meter error value data, and determines and marks the power meter based on the error threshold;

[0055] The error trend analysis module establishes an error database for power meters with errors within the allowable range, separately records the error data, analyzes the error change trend, and marks and observes abnormal power meters in advance;

[0056] The fault matching module extracts the error value data before and after the actual fault of the offline power meter based on the online and offline power meter detection data, further analyzes the obtained error values, generates a fault type curve, and pre-determines the error data in the red file of the power meter database.

[0057] The beneficial effects of the present invention are as follows:

[0058] 1. By collecting the error data of the power meter, analyzing the trend of the error change rate of the power meter within the allowable error range, discovering potential faulty power meters, and marking the potential faulty power meters, the present invention enables maintenance before the actual occurrence of the fault, avoiding greater power losses or equipment damages;

[0059] 2. By setting different secondary detection threshold intervals, the present invention can dynamically adjust the detection period according to the risk interval of the error change rate, ensure targeted data acquisition under different risk levels, reduce the data upload frequency and increase the determination threshold for the risk interval with a smaller error change rate, reduce the occupation of communication bandwidth and the probability of misjudgment, and for the high-risk interval, ensure timely discovery of abnormalities and take measures by uploading high-frequency data and reducing the determination threshold, avoiding fault delays;

[0060] 3. By combining the offline power meter fault report, collecting the error data of relevant faulty power meters, generating curves for the error values, and combining multiple groups of the same fault error curves to determine the final fault determination curve, and matching the error curves of the abnormally marked power meters in the power meter database, the present invention pre-determines the power meter fault, improving the work efficiency of the operation and maintenance personnel;

[0061] 4. By combining the offline fault report and the error data analysis method, the present invention can not only improve the accuracy and efficiency of fault diagnosis, but also optimize the operation and maintenance strategy, reduce the maintenance cost, achieve more efficient real-time detection and maintenance of the power meter, and enhance the practicability and functionality. Description of the Drawings

[0062] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for use in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0063] Figure 1 is a flowchart of a method for real-time on-site detection of an electric energy meter according to an embodiment of the present invention;

[0064] Figure 2 is a block diagram of a system for real-time on-site detection of an electric energy meter according to an embodiment of the present invention. Detailed implementation manners

[0065] To further illustrate each embodiment, the present invention provides accompanying drawings. These accompanying drawings are part of the disclosure of the present invention, mainly used to illustrate the embodiments, and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementation manners and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0066] According to an embodiment of the present invention, a system for real-time on-site detection of an electric energy meter and its detection method are provided. Now, the present invention will be further described in conjunction with the accompanying drawings and detailed implementation manners:

[0067] Embodiment 1:

[0068] As Figure 1 shown, a method for real-time on-site detection of an electric energy meter according to an embodiment of the present invention includes the following steps:

[0069] S1. Regularly and real-time collect relevant data of the electric energy meter through sensors, and at the same time read the electric energy reading of the electric energy meter based on a unit time interval;

[0070] S11. Based on a period T, regularly collect relevant data of the on-site electric energy meter through a current sensor and a voltage sensor. The voltage sensor is used to measure the three-phase voltage value at the incoming line of the electric energy meter, and the current sensor is used to measure the three-phase current value, and a time stamp is attached;

[0071] S12. Real-time read the electric energy meter through the Modbus communication protocol and attach a time stamp. Upload the three-phase voltage value, three-phase current value, and real-time reading of the electric energy meter to the electric energy meter database through Wi-Fi.

[0072] It should be noted that the electric energy meter database includes the files of all subordinate electric energy meters. Each file includes the location, installation time, and usage time of the current electric energy meter. At the same time, the three-phase voltage values, three-phase current values, and real-time readings of the relevant electric energy meters are received in real time through Wi-Fi, attached with time stamps, and recorded inside the relevant electric energy meter files. Parameter collection is performed on the electric energy meters once every regular period T. Depending on the usage scenarios of the three-phase electric energy meters, the regular period T is generally 18 - 36 hours.

[0073] S2. Based on the current and voltage data within a unit time, calculate the theoretical electric energy value using the electric energy calculation formula, and combine it with the measured electric energy value collected in real time within the unit time to obtain the error value data of the electric energy meter. Determine and mark the electric energy meter based on the error threshold.

[0074] S21. Through the electric energy meter database, extract the three-phase current values and three-phase voltage values of the electric energy meter to be detected within a unit time based on the time stamp, and obtain the theoretical electric energy value of the electric energy meter to be detected within the unit time t through the calculation formula. The algorithm formula is as follows:

[0075]

[0076] Among them, W0 represents the theoretical electric energy value of the electric energy meter to be detected within the unit time t. represents the average three-phase voltage value within the unit time t. represents the average three-phase current value within the unit time t.

[0077] S22. Through the electric energy meter database, calculate the measured electric energy value of the electric energy meter to be detected within the unit time t. The specific algorithm formula is as follows:

[0078] ΔW = W 最终 - W 起始 ;

[0079] Among them, ΔW represents the measured electric energy value of the electric energy meter to be detected, W 最终 represents the actual reading of the electric energy meter at the end of the unit time t, and W 起始 represents the actual reading of the electric energy meter at the start of the unit time t.

[0080] S23. Based on the theoretical electric energy value and the measured electric energy value of the electric energy meter to be detected, calculate the error value of the electric energy meter to be detected. The specific algorithm formula is as follows:

[0081] W1 = |ΔW - W0|;

[0082] Among them, W1 represents the error value of the electric energy meter. By comparing with the error threshold W θ , determine the operating state of the electric energy meter to be detected:

[0083] When W1 ≥ W θWhen it does, it indicates that there are obvious errors in the electricity meter to be tested, and the file of the electricity meter to be tested in the electricity meter database is marked in red;

[0084] When W1 < W θ it indicates that there are no obvious errors in the electricity meter to be tested, and the error value of this test is recorded in the file.

[0085] It should be noted that the error threshold of the electricity meter can be set according to the relevant regulations in the "DL / T 645-2017 Technical Specification for Electric Energy Meters" or the standards of the International Electrotechnical Commission (IEC), or can be set based on the empirical method by consulting experts in related fields. At the same time, according to the usage scenarios and accuracy levels of different electricity meters, the threshold is flexibly adjusted to meet different usage scenarios. The red-marked file indicates that the electricity meter is abnormal and needs to be calibrated or repaired offline.

[0086] Embodiment 2:

[0087] S3. For the electricity meters with errors within the allowable range, establish an error database, record the error data separately, analyze the error change trend, and mark and observe abnormal electricity meters in advance;

[0088] S31. In the electricity meter files under the electricity meter database, establish an error database, which includes the error values within the normal range recorded each time;

[0089] S32. Analyze the change trend of the error values in the error database, calculate the error change rate, and mark the files of the electricity meters with abnormal change rates.

[0090] S321. Based on the error data in the error database, calculate the error rate of two adjacent consecutive errors. The specific algorithm formula is:

[0091]

[0092] where e i represents the error rate, E i+1 , E i represent two adjacent error data in the error database. The error rate is statistically calculated and the error change rate Δe i is obtained based on the calculation of two adjacent error rates. The algorithm formula is:

[0093] Δe i = e i+1 - e i ;

[0094] where e i+1 , e i respectively represent two adjacent error rate data;

[0095] S322. Analyze the change tendency of the current archive watt-hour meter based on the error change rate Δe i , and the specific steps are as follows:

[0096] When any one Δe i and Δe i+1 are both greater than 0, and Δe i+1 >Δe θ , it means that the error of the current watt-hour meter is gradually increasing, and the increasing speed exceeds the change rate threshold Δe θ , and mark the current watt-hour meter archive in blue.

[0097] It should be noted that when Δe i is greater than 0, it means that the error change speed is accelerating, that is, the error is gradually increasing. When Δe i is less than 0, it means that the error change speed is slowing down, that is, the error is gradually decreasing. By adjusting the change rate threshold Δe θ , according to the specific scenario requirements, the change speed of the error can be grasped, the possible potential faulty watt-hour meters can be marked, and by analyzing the past error change rate data, the statistical characteristics of the change rate are calculated, including the mean value, standard deviation, and maximum value, and the change rate threshold is set based on the statistical results.

[0098] S3221. Based on the change rate threshold Δe θ , conduct continuous error detection on the current watt-hour meter for the watt-hour meter database archive with blue marks, and the specific steps are as follows: Conduct secondary determination on the Δe i+1 data. When 1.25Δe θ ≤Δe i+1 <1.5Δe θ , continuously for N times based on cycles, obtain the three-phase current data and three-phase voltage data of the relevant watt-hour meter through the sensor, obtain the short-term error change rate, and conduct short-term continuous determination on the error change rate. When there are times of short-term error change rates that are all greater than or equal to 1.25Δe θ , mark the relevant watt-hour meter archive in red;

[0099] When 1.5Δe θ ≤Δe i+1 <1.75Δe θ , continuously for N times based on cycles, obtain the three-phase current data and three-phase voltage data of the relevant watt-hour meter through the sensor, obtain the short-term error change rate, and conduct short-term continuous determination on the error change rate. When there are times of short-term error change rates that are all greater than or equal to 1.5Δe θ , mark the relevant watt-hour meter archive in red;

[0100] When Δe i+1 ≥ 1.75Δe θ , for N consecutive times, based on cycles, obtain the three-phase current data and three-phase voltage data of the relevant electricity meters through sensors, obtain the short-term error change rate, and conduct short-term continuous determination on the error change rate. When there are times of short-term error change rates that are all greater than or equal to 1.75Δe θ , mark the relevant electricity meter files as red.

[0101] It should be noted that through secondary determination of the error change rate, based on different error change rate values, by adopting different continuous detection schemes, conduct multiple continuous detections on the relevant electricity meters. Here, T is the normal detection period T of the relevant electricity meters. By shortening the regular detection period, according to different situations, continuously obtain error data under different regularities to assist in secondary determination. The value of N is generally 20, and it can also be set according to the specific usage scenario requirements. Divide the interval for secondary determination, which can reduce resource waste during continuous data upload.

[0102] S4. Based on the online and offline electricity meter detection data, extract the error value data of the offline electricity meter before and after the actual failure, further analyze the obtained error values, and conduct preliminary determination of the failure;

[0103] S41. Based on the failure report during the offline electricity meter failure detection, retrieve the relevant electricity meter files in the electricity meter database, collect and organize the electricity meter error data before and after the failure, and generate an error change curve model;

[0104] S411. Based on the failure report during the offline electricity meter failure detection, determine the failure types of the relevant electricity meters, including phase misalignment, frequency mismatch, open circuit failure, and current transformer failure, and collect the error data groups in the relevant electricity meter files under D groups of the same failure;

[0105] S412. Smooth the data through the moving average method, and draw the error curve through Python. The specific steps are as follows:

[0106]

[0107] Among them, S t is the smoothed value at the t-th moment, E t-i represents the value of the original error data at the i-th position forward from the t-th moment, and n represents the size of the moving average window;

[0108] Visualize the error data after smoothing through Python;

[0109] S413. Based on the Euclidean distance method, for the error curves under D groups of the same faults, similar curve segments are selected respectively. Through hierarchical clustering, the similar line segments are clustered, and the line segment of the clustering center is selected as the representative line segment, and the final fault curves are synthesized in the order of time stamps.

[0110] It should be noted that when processing the error curves under multiple groups of the same faults based on the Euclidean distance method and synthesizing the final fault curves, not only the overall similarity of the curves is considered, but also a suitable representative curve is selected through clustering and the central segment, ensuring the representativeness and accuracy of the final fault curves. The number of groups D is generally 15 groups, and the number of groups can also be added or reduced according to actual needs.

[0111] S42. Based on the error change curve model, curve generation is performed on the error data in the existing red - marked archives, and the error change curve model is matched through cosine similarity to pre - judge the fixed type of the electric energy meter and record it in the relevant electric energy meter archives.

[0112] It should be noted that when no corresponding error curve is matched, no recording operation is performed. By pre - judging the fault type in advance, it can assist the offline maintenance personnel in preparing for the maintenance of the electric energy meter faults.

[0113] Embodiment 3:

[0114] As Figure 2 shown, a real - time on - site detection system for electric energy meters includes a data acquisition module, a real - time detection module, an error trend analysis module, and a fault matching module:

[0115] The data acquisition module includes a current sensor and a voltage sensor, which are used to regularly and real - time collect the relevant data of the electric energy meter. At the same time, based on the unit time interval, the electric energy readings of the electric energy meter are used to establish an electric energy meter database, and the collected data is transmitted to the subsequent modules.

[0116] The real - time detection module calculates the theoretical electric energy value based on the current and voltage data within the unit time through the electric energy calculation formula, and combines it with the measured electric energy value collected in real - time within the unit time to obtain the error value data of the electric energy meter, and judges and marks the electric energy meter based on the error threshold.

[0117] The error trend analysis module establishes an error database for the electric energy meters with errors within the allowable range, separately records the error data, analyzes the error change trend, and marks and observes the abnormal electric energy meters in advance.

[0118] The fault matching module extracts the error value data before and after the actual faults of the offline electric energy meters based on the online and offline electric energy meter detection data, further analyzes the obtained error values, generates a fault type curve, and pre - judges the error data in the red archives of the electric energy meter database.

[0119] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for real-time detection of electric energy meter field, characterized in that: The method comprises the following steps: S1. Regularly collect data related to the energy meter in real time through sensors, including current and voltage data. At the same time, read the energy meter readings based on unit time intervals, and upload the current, voltage, and energy meter readings to the energy meter database. S2. Based on the current and voltage data per unit time, the theoretical electric energy value is calculated using the electric energy calculation formula, and combined with the actual measured electric energy value collected in real time per unit time, the electric energy meter error value data is obtained. The electric energy meter is judged and marked based on the error threshold. The error value of the electric energy meter to be tested is calculated based on the theoretical electric energy value and the actual measured electric energy value of the electric energy meter to be tested. When the error value of the electric energy meter to be tested is greater than or equal to the error threshold, the file of the electric energy meter to be tested in the electric energy meter database is marked red. S3. For energy meters with errors within the allowable range, establish an error database, record the error data of energy meters with errors within the allowable range separately, analyze the error change trend, mark and observe abnormal energy meters in advance, and mark the relevant energy meter files in red when the error change trend meets the set conditions; S4. Pre-determine the fault type of the electric energy meter to be tested based on the offline electric energy meter detection data. S4 includes the following sub-steps: S41. Based on the fault report during offline electric energy meter fault detection, retrieve relevant electric energy meter files from the electric energy meter database, collect and organize the electric energy meter error data before and after the fault occurs, and generate an error change curve model; S42. Based on the error change curve model, a curve is generated for the error data in the existing red mark file, and the error change curve model is matched by cosine similarity to pre-determine the fault type of the electric energy meter and record it in the relevant electric energy meter file.

2. The method for real-time detection of electric energy performance according to claim 1, characterized in that: The S1 includes the following sub-steps: S11, through the current sensor and voltage sensor, based on the periodic Real-time collection of on-site electricity meter data, where the voltage sensor is used to measure the three-phase voltage value at the electricity meter input line, and the current sensor is used to measure the three-phase current value, and attach a time stamp; S12, pass The communication protocol reads the energy meter readings in real time and attaches a timestamp. The three-phase voltage values, three-phase current values and real-time readings of the electric energy meter are uploaded to the electric energy meter database.

3. The method for real-time detection of electric energy performance according to claim 2, characterized in that: The S2 includes the following sub-steps: S21, through the electric energy meter database, extract the three-phase current value and the three-phase voltage value of the electric energy meter to be tested per unit time based on the timestamp, and obtain the unit time of the electric energy meter to be tested through the calculation formula The theoretical electric energy value within is calculated using the following algorithm: ； in, Represents unit time Theoretical electric energy value of the electric energy meter to be tested, Represents unit time The average three-phase voltage value within Represents unit time Average three-phase current value within; S22. Calculate the unit time through the electricity meter database The specific algorithm formula for the measured electric energy value of the electric energy meter to be tested is: ； in, Represents the actual measured energy value of the energy meter to be tested, Represents unit time The actual reading of the energy meter at the end of the test. Represents unit time The actual reading of the energy meter at the beginning; S23. Based on the theoretical electric energy value and the actual electric energy value of the electric energy meter to be tested, the error value of the electric energy meter to be tested is calculated. The specific algorithm formula is: ； in, Represents the error value of the energy meter, by comparing the error threshold , determine the operating status of the electric energy meter to be tested: when ≥ When , it means that the electric energy meter to be tested has obvious error, and the electric energy meter file to be tested in the electric energy meter database is marked in red; when < When , it means that there is no obvious error in the electric energy meter to be tested, and the error value at that time is recorded in the file.

4. The method for real-time detection of electric energy performance according to claim 3, characterized in that: The S3 includes the following sub-steps: S31. Establish an error database in the electric energy meter files under the electric energy meter database, wherein the error database includes an error value within a normal range for each record; S32. Analyze the error value change trend in the error database, calculate the error change rate, and file mark the electric energy meter with abnormal change rate.

5. The method for real-time detection of electric energy performance according to claim 4, characterized in that: The S32 includes the following sub-steps: S321. Calculate the error rate of two consecutive adjacent error data based on the error data in the error database. The specific algorithm formula is: ； in, represents the error rate, Represents two adjacent error data in the error database, calculates the error rate and obtains the error change rate based on the two adjacent error rates , and its algorithm formula is: ； in, Represent two adjacent error rate data respectively; S322, based on the error change rate , analyze the error change trend of the current archive electricity meter, and the specific steps are: When any time and are greater than 0, and > When , it means that the current electric energy meter error is gradually increasing, and the rate of increase exceeds the change rate threshold , mark the current electricity meter file in blue.

6. The method for real-time detection of electric energy performance according to claim 5, characterized in that: The S322 includes the following sub-steps: S3221, for the blue-marked energy meter files, based on the change rate threshold , conduct continuous error detection on the current electric energy meter, the specific steps are: The data is judged twice. ≤ < When, continuous Based on Period, obtain the three-phase current data and three-phase voltage data of the relevant electric energy meter through the sensor, obtain the short-term error change rate, and make a short-term continuous judgment on the error change rate. The sub-short-term error change rate is greater than or equal to When the relevant electricity meter file is marked in red; when ≤ < When, continuous Based on Period, obtain the three-phase current data and three-phase voltage data of the relevant electric energy meter through the sensor, obtain the short-term error change rate, and make a short-term continuous judgment on the error change rate. The sub-short-term error change rate is greater than or equal to When the relevant electricity meter file is marked in red; when ≥ When, continuous Based on Period, obtain the three-phase current data and three-phase voltage data of the relevant electric energy meter through the sensor, obtain the short-term error change rate, and make a short-term continuous judgment on the error change rate. The sub-short-term error change rate is greater than or equal to When the relevant electricity meter file is marked in red.

7. The method for real-time detection of electric energy performance according to claim 1, characterized in that: The S41 includes the following sub-steps: S411, based on the fault report of the offline electric energy meter fault detection, determine the fault type of the relevant electric energy meter, including phase misalignment, frequency mismatch, open circuit fault and current transformer fault, and collect Error data group in related electric energy meter files under the same fault; S412, smoothing the data by moving average method, The specific steps for drawing the error curve are as follows: ； in, It is The smoothed value at time, Represents the original error data in Always moving forward The value of the position, Represents the size of the moving average window; pass Visualize the smoothed error data; S413, based on the Euclidean distance method, The error curves under the same fault are grouped together, and similar curve segments are selected respectively. Similar segments are clustered through hierarchical clustering, and the segment at the cluster center is selected as the representative segment. The final fault curve is synthesized in the order of timestamps.

8. A real-time detection system for electric energy meter, characterized in that: The method adopts the real-time detection method of electric energy meter field according to any one of claims 1 to 7, comprising a data acquisition module, a real-time detection module, an error trend analysis module, and a fault matching module: The data acquisition module includes a current sensor and a voltage sensor, which are used to regularly collect data related to the electric energy meter in real time, including current and voltage data, and read the electric energy reading of the electric energy meter based on a unit time interval, upload the current, voltage and electric energy reading data of the electric energy meter to the electric energy meter database, and transmit the collected data to subsequent modules; The real-time detection module calculates the theoretical electric energy value based on the electric energy calculation formula through the current and voltage data in the unit time, and obtains the electric energy meter error value data in combination with the measured electric energy value collected in real time in the unit time, judges and marks the electric energy meter based on the error threshold, calculates the error value of the electric energy meter to be tested based on the theoretical electric energy value of the electric energy meter to be tested and the measured electric energy value of the electric energy meter to be tested, and when the error value of the electric energy meter to be tested is greater than or equal to the error threshold, marks the electric energy meter file to be tested in the electric energy meter database in red; The error trend analysis module establishes an error database for electric energy meters with errors within the allowable range, records the error data separately, analyzes the error change trend, marks and observes abnormal electric energy meters in advance, and marks the relevant electric energy meter files in red when the error change trend meets the set conditions; The fault matching module pre-determines the fault type of the electric energy meter to be tested based on the offline electric energy meter detection data, specifically as follows: The fault matching module retrieves relevant electric energy meter files from the electric energy meter database based on the fault report during offline electric energy meter fault detection, collects and organizes the error data of the electric energy meter before and after the fault occurs, and generates an error change curve model; The fault matching module generates a curve for the error data in the existing red mark file based on the error change curve model, and pre-determines the fault type of the electric energy meter by matching the error change curve model through cosine similarity, and records it in the relevant electric energy meter file.

Images