CN112485748A - Phase-to-phase judgment method for single-phase electric meter - Google Patents
Phase-to-phase judgment method for single-phase electric meter Download PDFInfo
- Publication number
- CN112485748A CN112485748A CN202011103028.XA CN202011103028A CN112485748A CN 112485748 A CN112485748 A CN 112485748A CN 202011103028 A CN202011103028 A CN 202011103028A CN 112485748 A CN112485748 A CN 112485748A
- Authority
- CN
- China
- Prior art keywords
- data
- phase
- electric meter
- meter
- gateway
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000004364 calculation method Methods 0.000 claims abstract description 34
- 238000011156 evaluation Methods 0.000 claims abstract description 4
- 238000004140 cleaning Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 16
- 238000012360 testing method Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 16
- 230000001419 dependent effect Effects 0.000 claims description 8
- 230000003203 everyday effect Effects 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 230000005611 electricity Effects 0.000 claims description 2
- 238000012217 deletion Methods 0.000 claims 2
- 230000037430 deletion Effects 0.000 claims 2
- 230000001502 supplementing effect Effects 0.000 claims 2
- 238000013480 data collection Methods 0.000 abstract description 4
- 238000007405 data analysis Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000003442 weekly effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/04—Testing or calibrating of apparatus covered by the other groups of this subclass of instruments for measuring time integral of power or current
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R25/00—Arrangements for measuring phase angle between a voltage and a current or between voltages or currents
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A single-phase ammeter phase discrimination judging method comprises the steps of establishing a model according to historical data of an ammeter and a gateway table, and judging the gateway table and the phase discrimination to which the ammeter belongs; in the real-time judgment, the data of each electric meter per day is judged according to the established model to obtain the corresponding gateway and phase, then the results of multiple days are summarized, and the result with the most days is selected as the gateway and phase of the user electric meter. According to the method for distinguishing the electric meter phase, the electric meter phase is determined by adopting a single-day distinguishing and summarizing evaluation mode during distinguishing, and compared with the method for distinguishing multi-day data at one time, the method reduces the calculation pressure and improves the accuracy. Meanwhile, a correction method using analog data is provided for the condition that the data collection is asynchronous so as to influence the phase judgment result of the ammeter.
Description
Technical Field
The invention belongs to the technical field of electric power, relates to electric power data analysis, is used for electric meter data analysis in a power distribution network scene, and is a single-phase electric meter phase judging method.
Background
The analysis and processing of the power grid operation data including power load modeling, power network loss calculation and the like have very key significance for improving the automation level of the power grid, and are beneficial to improving the power supply quality and improving the power supply reliability. In the process, the phase information of the single-phase intelligent electric meter has important value for analyzing and processing the operation data of the power grid, can enable load modeling, grid loss calculation and the like to be more accurate, and can be further used for load balance analysis, load unbalance management and the like. Based on the current huge number of electric meters, manual identification and confirmation are obviously unrealistic, so that an automatic phase-to-phase judgment method for single-phase electric meters is needed. The current methods for judging the phase mainly include the following methods:
1. and (4) a judgment method based on information clustering. The voltages of different phases are different in time sequence, and the voltage of the single-phase electric meter at the same moment is collected and subjected to cluster analysis, so that the phase of the electric meter is judged. The method is simple and common, but the accuracy is still required to be improved, in addition, because the data are clustered based on the data acquired simultaneously, the acquisition equipment and the higher requirements on the acquisition equipment are set, and the condition that the phase judgment result of the electric meter is influenced because the data acquisition is asynchronous exists.
2. A method for determining carrier-based communication. The method needs the concentrator and the intelligent electric meter to be provided with carrier communication modules, the concentrator sends carrier signals to the electric meter, and the measured phase position of the electric meter is judged through the response of the electric meter. However, the method has high requirements on equipment and is not easy to popularize.
Disclosure of Invention
The invention aims to solve the problems that: the electric power grid operation data analysis needs electric meter phase judgment, but the existing method is difficult to meet the requirements in the aspects of accuracy, data synchronism, practicality, easy popularization and the like.
The technical scheme of the invention is as follows: a single-phase ammeter phase discrimination judging method comprises the steps of firstly establishing a discrimination model according to history data of an ammeter and a gateway table, and using the discrimination model to discriminate the gateway table and the phase to which the ammeter belongs; when the electric meters are judged in real time, data are collected on each electric meter and the corresponding meter, the corresponding meter and the corresponding phase of the electric meter are judged every day according to the established judgment model, then results of multiple days are collected, and the result with the largest number of days is selected as the corresponding meter and the corresponding phase of the electric meter according to the judgment results of the electric meters every day.
Further, when real-time discrimination is carried out, the method for correcting the voltage and the current of the electric meter and the voltage and the current of the gateway meter by using analog data for the collected data of the electric meter and the gateway meter comprises the following steps:
1) randomly extracting voltage and current data with a set proportion for each electric meter, and extracting current and voltage data corresponding to all gateway meters corresponding to the extraction time of the electric meter data;
2) data processing: sorting all the extracted data from small to large according to time to obtain n groups of data sets [ electric meter voltage, electric meter current, gateway meter voltage and gateway meter current ], setting a variable X to represent a data set number, wherein the variable X belongs to [1, n ];
3) processing the data group by using a proximity fitting or linear interpolation method to obtain a new data set for inputting a discrimination model for judgment; or randomly extracting data with a set proportion from the n groups of data groups, and respectively processing the data by using a proximity fitting method and a linear interpolation method to extract data which are predicted not to be extracted, so as to obtain a new data set for inputting the discrimination model for judgment.
Preferably, the discriminant model of the present invention is: the method comprises the steps of constructing a model by taking the voltage of an ammeter as a dependent variable and the current of the ammeter, the voltage of a gateway meter and the current of the gateway meter as independent variables, constructing a training set and a testing set by historical data of the ammeter and the gateway meter, wherein the training set is used for training the model, and the testing set is used for judging the quality of the model; during testing, the voltage and current of the electric meter in the test set are respectively substituted into the model together with the A, B, C three-phase voltage and current of different gateway tables to obtain the discrimination coefficient of the model, the result with the maximum discrimination coefficient is selected, and the corresponding gateway table and phase position of the result are the gateway table and phase position of the electric meter.
Compared with the prior art, the method provided by the invention adjusts the application of the data set and the algorithm model result, the prior art adopts multi-day data to substitute the model for calculation, and generally adopts weekly data, because the problems of accuracy reduction and contingency increase exist due to the fact that the data volume is too small, and the accuracy of the discrimination result of single-day data is generally not more than 90%; the invention adopts single-day data substitution calculation, and then integrates a plurality of single-day results to carry out voting judgment to obtain the final phase result, and the accuracy of the obtained result is not continued to the simple superposition of the results of the single-day data, but is greatly improved and can reach more than 97%. Meanwhile, aiming at the condition that asynchronous interference phase judgment results exist in data collection, the invention also provides a method for combining proximity fitting and linear interpolation, interference factors are eliminated, and the accuracy of the judgment results is improved. According to the method, the single-day data calculation is performed, then the majority priority judgment method and the data asynchronous interference elimination method are performed, so that the calculation pressure on equipment can be reduced, the calculation efficiency is improved, and meanwhile, the accuracy is improved. The results of field and field tests show that the accuracy of the invention in judging different data sets is over 97%, and the accuracy in judging more than 50% of data sets can reach 100%.
Drawings
FIG. 1 is a flow chart of a phase determination method for a single-phase electric meter according to the present invention.
Detailed Description
The invention provides a phase discrimination judging method for a single-phase electric meter, which comprises the steps of firstly establishing a discrimination model according to historical data of the electric meter and a gate table, and discriminating the gate table and the phase discrimination to which the electric meter belongs; when the electric meters are judged in real time, data are collected on each electric meter and the corresponding meter, the corresponding meter and the corresponding phase of the electric meter are judged every day according to the established judgment model, then results of multiple days are collected, and the result with the largest number of days is selected as the corresponding meter and the corresponding phase of the electric meter according to the judgment results of the electric meters every day.
The specific implementation method of the invention comprises the aspects of data collection, data cleaning, algorithm model and result evaluation, and the specific flow is shown in figure 1.
1. And (6) collecting data. The method comprises the steps of collecting four parameters of a gateway table and a user electric meter in minute-level current, voltage, power and collecting time, and the meter address numbers of the gateway table and the user electric meter.
2. And (6) data cleaning. The data cleaning aims at four objects, namely a missing value, an abnormal value, a repeated value and a useless value, and different forms of different objects are processed by adopting corresponding methods so as to obtain expected data. Including non-empty, repeat value, outlier, useless value, missing value, etc.
And checking the field data when the non-empty cleaning requirement field is not empty. If the data is null, corresponding processing is required, namely clearing is carried out; repeated cleaning requires uniform storage of the same type of data in the multi-service system, and then the uniqueness of the main key is ensured; useless value cleaning requires clearing data fields in the service which are not needed to be used or are useless; cleaning abnormal values, including checking value errors, format errors, logic errors, data inconsistency and the like, and cleaning and correcting according to specific conditions; the missing value cleaning needs to determine the missing value range, strategies are respectively formulated according to the set missing proportion and field importance, missing values with high importance and low missing rate are filled through calculation or are estimated through experience or business knowledge, attempts with high importance and high missing rate are obtained through counting and completing from other channels or other fields are used for calculation, processing or simple filling is not conducted on the missing values with low importance and low missing rate, the field is deleted when the missing rate with low importance is high, and the importance and the missing rate are determined according to the set missing proportion and the field importance.
Although the data cleaning can reduce wrong or non-value data, when the data are collected by the electric meter and the gateway meter, the situation that the data collection is asynchronous to influence the phase judgment result of the electric meter still exists objectively, so that a method for correcting the voltage and the current of the electric meter and the voltage and the current of the gateway meter by using an analog data method is adopted in data processing, and the method is specifically divided into the following three steps:
1) randomly extracting voltage and current data with a set proportion for each electric meter, and extracting current and voltage data corresponding to all gateway meters corresponding to the extraction time of the electric meter data; the extraction is generally carried out according to the proportion of 75 percent, and the adjustment can be carried out according to the actually acquired data volume and the hardware calculation processing capacity;
2) data processing: sorting all the extracted data from small to large according to time to obtain n groups of data sets [ electric meter voltage, electric meter current, gateway meter voltage and gateway meter current ], setting a variable X to represent a data set number, wherein the variable X belongs to [1, n ];
3) processing the data group by using a proximity fitting or linear interpolation method to obtain a new data set for inputting a discrimination model for judgment; as an improvement, two processing methods can be combined, data with a set proportion is randomly extracted from n groups of data sets, and the data are processed by using a proximity fitting method and a linear interpolation method respectively to extract data prediction unextracted data to obtain a new data set for inputting a discrimination model for judgment.
The proximity fitting method comprises the following specific steps:
(1) setting variables: setting the variable X as independent variable, and setting the dependent variable as voltage and current of the electric meter and voltage and current of the gateway meter respectively;
(2) for the data sets subjected to the proximity fitting, sampling is carried out proportionally, for example, sampling is carried out 75%, the data sets are traversed, and for each non-sampled data set, points corresponding to K sampled data sets with the minimum distance are selected; taking the predicted ammeter current as an example, selecting the ammeter current of the data set with the minimum distance from the ammeter current of the non-sampled data set, wherein the distance calculation is to calculate the distance between two points by taking a variable X as an X axis and a dependent variable as a y axis;
(3) respectively predicting the voltage of an ammeter, the current of the ammeter, the voltage of a gateway table and the current of the gateway table to obtain a predicted value of the non-sampled data set;
(4) the predicted values for the sampled data set and the non-sampled data set are integrated to form a new data set.
The linear interpolation method comprises the following specific steps:
(1) sampling the data group subjected to linear interpolation according to a proportion, reserving the value of the sampled data group, and changing the value of the non-sampled data group into a null value;
(2) the data sets are sorted according to time, a variable X is used as an X axis, the data sets are used as a y axis, the data sets are traversed, null values are supplemented according to a linear interpolation formula, and the linear interpolation formula is as follows:
in the above formula, x0The value, X, of a variable X preceding a null position1The value, y, of a variable X which is a non-null position following the null position0In the data set of preceding non-empty positions as empty positionsValue of (a), y1Is the value in the data set of the non-null position following the null position, X is the variable X value of the null position;
3) the null values are supplemented by the above linear interpolation to obtain a new data set.
If the processing is carried out by adopting a combination mode of proximity fitting and linear interpolation, the proximity fitting is carried out on 75% of the extracted data, and the rest is processed by adopting a linear interpolation mode, then the new data sets of 75% of the data sets in the n groups of data sets are obtained by the proximity fitting processing, and the new data sets of 25% of the data sets in the n groups of data sets are obtained by the linear interpolation. And combining the two to obtain a complete new data set for inputting the judgment model.
By the processing method, the interference caused by asynchronous data acquisition is eliminated, and the accuracy of judgment is improved. Experimental verification shows that when a test is carried out on a certain data set, if the interference caused by asynchronous data acquisition is not eliminated, the accuracy rate is 95.52%. After the linear interpolation method is used, the accuracy is improved to 99.25%, the adjacent fitting has similar accuracy, and the accuracy is further improved to 99.63% after the adjacent fitting and the linear interpolation method are further combined.
3. And (4) evaluating an algorithm model and a result. The method comprises the following five steps:
(1) for each ammeter, selecting ammeter voltage as a dependent variable, selecting ammeter current, gateway meter voltage and gateway meter current as independent variables, obtaining a data set from data cleaned, analyzing the correlation among variables, selecting effective indexes to model, randomly dividing the data of the data set into a training set and a testing set according to a set proportion, wherein the training set is used for training a model, and the testing set is used for judging the quality of the model. And respectively introducing the current and the active power into the model by using SPSS software for calculating the result values of the test, the judgment coefficient delta and the like.
(2) Establishing a model: using training set data with a sample size of m, randomly taking m samples from which there is a put back to form a sub-sample set, with a put back to take out one again and then sampling from all samples. The method comprises the following steps that a sub-sample set has P characteristics, wherein the characteristics comprise electric meter current, gateway meter current and gateway meter voltage, and T times are extracted in total, so that T sub-sample sets are available in total, all characteristics are selected for each sub-sample set, namely P characteristics are used as split characteristic subsets, trees are respectively constructed for the T sub-sample sets, then the average value of the results of the T trees is calculated to obtain the prediction output of a final model, in the process of training the model, the model is continuously subjected to parameter adjustment, the optimal model is finally selected through evaluation indexes, finally, the electric meter voltage in a test set is predicted by using the trained optimal model, and the real voltage and the predicted voltage of the electric meter are subjected to calculation of a discrimination coefficient delta through the following formula;
in the above formula UiIn order to obtain the true voltage of the electric meter,
in order to predict the value of the model,
the average value of the real voltage of the ammeter is shown, and m is the sample amount.
(3) And (3) establishing a model of the step (2) by the voltage and the current of each electric meter and the voltage and the current of the A phase, the B phase and the C phase of different relation tables respectively, and obtaining a plurality of coefficient values according to the formula in the step (2).
(4) And (4) comparing the discrimination coefficients in the step (3), and selecting the model with the maximum discrimination coefficient, wherein the gateway table and the phase corresponding to the model are the gateway table and the phase to which the electric meter belongs.
(5) And judging the gate table and the phase position of each electric meter every day, finally summarizing the results of multiple days, and selecting the result with the most days as the gate table and the phase position of the electric meter.
In the prior art, the phase of the electric meter is generally judged once by adopting data of one week (7 days), namely, the voltage and current of the electric meter and the voltage and current data of the gateway meter are gathered for 7 days to calculate so as to ensure the accuracy of the data. In the data analysis aspect, although such a calculation method collects more data and has a sufficient sample size to ensure the accuracy of the determination, the calculation method also causes problems of increased calculation pressure and slower calculation speed due to increased data size. In the prior art, the larger the sample data size of the model is, the more the accuracy of the judgment result is improved. The invention is based on the principle that firstly, the data volume of a single algorithm is reduced, the calculation once in the previous 7 days is changed into the calculation once per day, 7 calculations are respectively carried out in the previous 7 days, and then the results of the 7 days are subjected to the majority of priority determination. Theoretically, model calculations for a single day data volume would reduce accuracy, but the present invention, after re-aggregating the 7 day results for majority priority, instead improves accuracy, as shown in table 1.
The scheme of the invention not only disperses the calculation task, reduces the calculation pressure and the data bearing pressure of the master station, improves the calculation efficiency, keeps the equal calculation time, but also improves the accuracy of the learning training result. The accuracy of three different sets of data sets given in table 1 is compared by using one-time calculation and fractional calculation, and the data sets are all obtained from a national power grid electricity utilization information acquisition platform.
TABLE 1 comparison of Once-and-fractional calculation accuracy
As can be seen from the table, the accuracy of multiple calculations is improved compared with that of one-time calculation, and the accuracy of 100% is achieved. For the operation of the power grid, the method has high requirements on the phase-to-phase judgment accuracy, and because one-time misjudgment can cause serious consequences, the calculation accuracy is improved to have great significance, and the superiority of the method in multiple calculations compared with a one-time calculation method is also proved.
Claims (8)
1. A single-phase ammeter phase discrimination judging method is characterized in that a discrimination model is established according to history data of an ammeter and a gateway table and is used for discriminating the gateway table and the phase to which the ammeter belongs; when the electric meters are judged in real time, data are collected on each electric meter and the corresponding meter, the corresponding meter and the corresponding phase of the electric meter are judged every day according to the established judgment model, then results of multiple days are collected, and the result with the largest number of days is selected as the corresponding meter and the corresponding phase of the electric meter according to the judgment results of the electric meters every day.
2. The method as claimed in claim 1, wherein the method for correcting the voltage and current of the electric meter and the voltage and current of the gateway meter using the analog data is used for the collected data of the electric meter and the gateway meter during the establishment of the discriminant model and during the real-time discriminant, comprising the following steps:
1) randomly extracting voltage and current data with a set proportion for each electric meter, and extracting current and voltage data corresponding to all gateway meters corresponding to the extraction time of the electric meter data;
2) data processing: sorting all the extracted data from small to large according to time to obtain n groups of data sets [ electric meter voltage, electric meter current, gateway meter voltage and gateway meter current ], setting a variable X to represent a data set number, wherein the variable X belongs to [1, n ];
3) processing the data group by using a proximity fitting or linear interpolation method to obtain a new data set for inputting a discrimination model for judgment; or randomly extracting data with a set proportion from the n groups of data groups, and respectively processing the data by using a proximity fitting method and a linear interpolation method to extract data which are predicted not to be extracted, so as to obtain a new data set for inputting the discrimination model for judgment.
3. The method for judging the phase of the single-phase electric meter according to claim 2, wherein the proximity fitting method comprises the following steps:
(1) setting variables: setting the variable X as independent variable, and setting the dependent variable as voltage and current of the electric meter and voltage and current of the gateway meter respectively;
(2) sampling data subjected to proximity fitting in proportion, traversing data groups, and selecting points corresponding to K sampled data groups with the minimum distance from each non-sampled data group;
(3) taking the average value of the values of the dependent variables corresponding to the K points as a predicted value of the dependent variable of the non-sampled data group;
(4) the predicted values for the sampled data set and the non-sampled data set are integrated to form a new data set.
4. The phase judging method of the single-phase electric meter according to claim 2, wherein the linear interpolation method comprises the following steps:
(1) sampling the data subjected to linear interpolation in proportion, reserving the value of a sampled data group, and changing the value of a non-sampled data group into a null value;
(2) traversing the data set, and supplementing null values according to a linear interpolation formula, wherein the linear interpolation formula is as follows:
in the above formula, x0The value, X, of a variable X preceding a null position1The value, y, of a variable X which is a non-null position following the null position0The value, y, in the data set preceding the non-null position of the null position1Is the value in the data set of the non-null position following the null position, X is the variable X value of the null position;
3) and supplementing null values through linear interpolation to obtain a new data set.
5. The method as claimed in any one of claims 1 to 4, wherein the discriminant model is: the method comprises the steps of constructing a model by taking the voltage of an ammeter as a dependent variable and the current of the ammeter, the voltage of a gateway meter and the current of the gateway meter as independent variables, constructing a training set and a testing set by historical data of the ammeter and the gateway meter, wherein the training set is used for training the model, and the testing set is used for judging the quality of the model; during testing, the voltage and current of the electric meter in the test set are respectively substituted into the model together with the A, B, C three-phase voltage and current of different gateway tables to obtain the discrimination coefficient of the model, the result with the maximum discrimination coefficient is selected, and the corresponding gateway table and phase position of the result are the gateway table and phase position of the electric meter.
6. The method as claimed in claim 5, wherein the discriminant model is characterized in that a training set and a test set are constructed from historical data of the electric meter and a gateway table, the training set data sample size is m, m samples are randomly extracted from the training set data sample size to form a sub-sample set, the sub-sample set has P features, the features include electric meter current, gateway table current and gateway table voltage, T times of extraction are totally performed, T sub-sample sets are total, all features, namely P features, are selected for each sub-sample set as split feature subsets, trees are respectively constructed for the T sub-sample sets, then an average value is calculated for results of the T trees to obtain a predicted output of a final model, the model is continuously subjected to parameter adjustment in the process of training the model, an optimal model is finally selected through evaluation of indexes, and then the electric meter voltage in the test set is predicted by using the trained optimal model, and calculating a discrimination coefficient delta from the real voltage and the predicted voltage of the electric meter by the following formula:
in the above formula UiIn order to obtain the true voltage of the electric meter,
in order to predict the value of the model,
the average value of the real voltage of the electricity meter is shown.
7. The method as claimed in any one of claims 1 to 4, wherein the data of the electric meter and the gateway meter includes four parameters of minute-level current, voltage, power and collection time of the electric meter and the gateway meter, and a meter address number of the gateway meter and the electric meter.
8. The method for judging the phase of the single-phase electric meter based on the regression algorithm as claimed in any one of claims 1 to 4, wherein data cleaning is firstly carried out on collected electric meter and gateway meter data, non-empty cleaning, repeated value cleaning, abnormal value cleaning, unused value cleaning and missing value cleaning are carried out on missing values, abnormal values, repeated values and unused values, and when a field is required to be non-empty, the field data is checked, and if the data is empty, the field data needs to be cleaned; repeated cleaning requires uniform storage of the same type of data in the multi-service system, and then the uniqueness of the main key is ensured; useless value cleaning requires clearing data fields in the service which are not needed to be used or are useless; cleaning abnormal values, including checking value errors, format errors, logic errors, data inconsistency and the like, and cleaning and correcting according to specific conditions; the missing value cleaning needs to determine the missing value range, strategies are respectively formulated according to the preset missing proportion and the field importance, and the missing values with high importance and low missing rate are filled through calculation or are estimated through experience or business knowledge; the attempt with high importance and high loss rate is obtained by taking data from other channels and completing or using other fields through calculation; the importance is low, the deletion rate is low, and no treatment or simple filling is performed; and deleting the field if the deletion rate is high and the importance is low.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011103028.XA CN112485748B (en) | 2020-10-15 | 2020-10-15 | Single-phase ammeter phase judging method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011103028.XA CN112485748B (en) | 2020-10-15 | 2020-10-15 | Single-phase ammeter phase judging method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112485748A true CN112485748A (en) | 2021-03-12 |
| CN112485748B CN112485748B (en) | 2023-10-24 |
Family
ID=74926128
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011103028.XA Active CN112485748B (en) | 2020-10-15 | 2020-10-15 | Single-phase ammeter phase judging method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112485748B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113075475A (en) * | 2021-03-24 | 2021-07-06 | 南方电网科学研究院有限责任公司 | Phase-splitting identification method, device, terminal and storage medium for electric energy meters in transformer area |
| CN115049874A (en) * | 2022-06-10 | 2022-09-13 | 广东电网有限责任公司 | User phase analysis method, system, device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170207818A1 (en) * | 2014-07-17 | 2017-07-20 | Jiangsu Linyang Energy Co., Ltd. | Method for differentiating power distribution areas and phases by using voltage characteristics |
| CN109738723A (en) * | 2018-12-29 | 2019-05-10 | 重庆邮电大学 | A kind of electric energy meter three-phase automatic identification method |
| CN110389267A (en) * | 2019-07-17 | 2019-10-29 | 国网陕西省电力公司电力科学研究院 | A kind of low-voltage platform area platform family relation recognition method based on intelligent electric energy meter acquisition data |
| CN110727662A (en) * | 2019-09-10 | 2020-01-24 | 国网浙江省电力有限公司电力科学研究院 | Low-voltage transformer area user phase identification method and system based on correlation analysis |
| CN111710150A (en) * | 2020-05-14 | 2020-09-25 | 国网江苏省电力有限公司南京供电分公司 | Abnormal electricity consumption data detection method based on countermeasure self-coding network |
-
2020
- 2020-10-15 CN CN202011103028.XA patent/CN112485748B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170207818A1 (en) * | 2014-07-17 | 2017-07-20 | Jiangsu Linyang Energy Co., Ltd. | Method for differentiating power distribution areas and phases by using voltage characteristics |
| CN109738723A (en) * | 2018-12-29 | 2019-05-10 | 重庆邮电大学 | A kind of electric energy meter three-phase automatic identification method |
| CN110389267A (en) * | 2019-07-17 | 2019-10-29 | 国网陕西省电力公司电力科学研究院 | A kind of low-voltage platform area platform family relation recognition method based on intelligent electric energy meter acquisition data |
| CN110727662A (en) * | 2019-09-10 | 2020-01-24 | 国网浙江省电力有限公司电力科学研究院 | Low-voltage transformer area user phase identification method and system based on correlation analysis |
| CN111710150A (en) * | 2020-05-14 | 2020-09-25 | 国网江苏省电力有限公司南京供电分公司 | Abnormal electricity consumption data detection method based on countermeasure self-coding network |
Non-Patent Citations (2)
| Title |
|---|
| 张丽强等: "基于多元线性回归的单相电表相别判断方法", 《电力自动化设备》 * |
| 程维杰等: "智能变电站稳定控制装置数字采样的设计与实现", 《电力工程技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113075475A (en) * | 2021-03-24 | 2021-07-06 | 南方电网科学研究院有限责任公司 | Phase-splitting identification method, device, terminal and storage medium for electric energy meters in transformer area |
| CN115049874A (en) * | 2022-06-10 | 2022-09-13 | 广东电网有限责任公司 | User phase analysis method, system, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112485748B (en) | 2023-10-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110350528B (en) | Low-voltage distribution area topology automatic identification method | |
| CN107340492A (en) | Electric power meter failure analysis methods with scene anticipation are excavated based on big data | |
| CN110389269B (en) | Low-voltage distribution area topological relation identification method and device based on current optimization matching | |
| CN107909208A (en) | Damage method drops in a kind of taiwan area distribution | |
| CN110321919A (en) | It is a kind of for the single-phase clustering method with the identification of phase relation in every family of low pressure | |
| CN112485748A (en) | Phase-to-phase judgment method for single-phase electric meter | |
| CN112329806B (en) | Ammeter clustering method for topology identification of low-voltage transformer area | |
| CN102033185A (en) | Method for judging wrong wiring of electric energy meter based on apparent power estimation | |
| CN113723844A (en) | Low-voltage transformer area theoretical line loss calculation method based on ensemble learning | |
| CN111654392A (en) | Low-voltage distribution network topology identification method and system based on mutual information | |
| CN112288015A (en) | Distribution network electrical topology identification method and system based on edge calculation improved KNN | |
| CN110555619A (en) | Power supply capacity evaluation method based on intelligent power distribution network | |
| CN111027886A (en) | Evaluation method of low-voltage treatment scheme considering unit cost effectiveness | |
| CN111178679A (en) | Phase identification method based on clustering algorithm and network search | |
| CN112182499B (en) | Low-voltage distribution network topological structure identification method based on time sequence electric quantity data | |
| CN112667567A (en) | Operation cost archiving method and device combining electric power data and power grid topology | |
| CN111612054A (en) | User electricity stealing behavior identification method based on non-negative matrix factorization and density clustering | |
| CN107240957B (en) | Method for monitoring and early warning of power grid based on large data flow type calculation | |
| CN113285440B (en) | Low-voltage transformer area loss reduction optimization method | |
| CN113627452A (en) | Electricity stealing behavior detection method based on machine learning random forest algorithm | |
| CN112000917B (en) | Low-voltage user meter box identification method integrating known phase difference and address information | |
| Zhong et al. | Low-voltage distribution network topology identification method based on segmented current features and cnn-lstm deep learning | |
| Li | Abnormal energy consumption analysis based on big data mining technology | |
| Xie et al. | Energy System Time Series Data Quality Maintenance System Based on Data Mining Technology | |
| CN114709814A (en) | Household transformation relation and topological connection relation identification method for distribution network area |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |