A system for identifying power consumption topology of household-to-substation relationships

Abstract

The present invention relates to a system for identifying power consumption topology of household-to-substation relationships in a transformer substation area, comprising a household-to-substation relationship database, a data acquisition module, an information scheduling module, a household-to-substation analysis module, a household-to-substation configuration module, a household-to-substation monitoring module, and a line loss adjustment module. Through such an arrangement, a line loss regression model is constructed through historical sample training without power outages, and the power consumption end is classified based on the line loss regression model, thereby achieving more accurate abnormal situation screening and completing the identification of abnormal situations of household-to-substation relationships.

Description

Technical Field

[0001] The present invention relates to the field of power consumption topology identification, and more specifically, to a system for identifying power consumption topology of a transformer substation and a household transformer. Background Art

[0002] With the in-depth development of refined marketing management of power systems, the correct substation topology information in the marketing and distribution archives is of great significance to line loss management, abnormality monitoring, fault repair and other applications in low-voltage distribution networks.

[0003] Due to incomplete records of basic files in the substation, load changes of the substation, and no records of changes in the electric meter files during maintenance, the substation topology file relationship is wrong. The traditional low-voltage distribution network topology analysis mainly relies on manual investigation. Due to the large number of mesh distribution lines in the low-voltage distribution network, cable trenches or overhead lines are laid in parallel. Inspections are not only inefficient and costly, but also unable to obtain low-voltage substation topology information in real time and accurately. At the same time, the identification of household-to-substation relationships through power outages has an impact on residents' lives. Although the current popularity of smart grids allows some areas to identify household-to-substation relationships through low-voltage carriers, this method is difficult to apply in some substations, and some old substations do not have the possibility of a unified carrier. If the household-to-substation relationship is recorded abnormally, it will be even less likely to be discovered. Summary of the invention

[0004] In view of this, the object of the present invention is to provide a system for identifying power consumption topology of household transformer relationships in an area.

[0005] In order to solve the above technical problems, the technical solution of the present invention is:

[0006] A system for identifying power consumption topology of household transformer relationships in a transformer substation area, comprising a household transformer relationship database, a data acquisition module, an information dispatching module, a household transformer analysis module, a household transformer configuration module, a household transformer monitoring module, and a line loss adjustment module;

[0007] The household-to-household change relationship database stores household-to-household change relationship information, and the household-to-household change relationship information includes transformer data and power consumption end data;

[0008] The data acquisition module includes a total power acquisition unit and a branch power acquisition unit, wherein the total power acquisition unit is used to acquire the total power information of the transformer, and the branch power unit is used to acquire the branch power information of each power consumption end;

[0009] The information scheduling module groups the received total power information and branch power information according to the household change relationship information in the household change relationship database;

[0010] The household change analysis module includes a model building unit, a deviation analysis unit, and a deviation marking unit. The model building unit is configured with a model building strategy. The model building strategy establishes a line loss regression model according to the total power information and branch power information of the same group. The deviation analysis unit determines the degree of fit of the line loss regression model corresponding to each branch power information of the group. The deviation marking unit is configured with a first fitting threshold and a second fitting threshold. If the degree of fit corresponding to a branch power information is lower than the first fitting threshold, an adjustment fitting mark is established for the power consumption end data corresponding to the branch power information. If the degree of fit of the branch power information is between the first fitting threshold and the second fitting threshold, a monitoring fitting mark is established for the power consumption end data corresponding to the branch power information. If the degree of fit of the branch power information is higher than the second fitting threshold, a configuration fitting mark is established for the power consumption end corresponding to the branch power information.

[0011] The household change configuration module includes a configuration trigger unit, a horizontal fitting unit and a household change configuration unit. The configuration trigger unit is configured with a first trigger condition. When the power end with a configuration fitting mark meets the first trigger condition, a configuration trigger signal is generated. The horizontal fitting unit responds to the configuration trigger signal. The horizontal fitting unit fits the line loss regression model of other groups according to the branch power information of the power end to calculate the corresponding fitting degree. If there is a line loss regression model with a fitting degree lower than the first fitting threshold, the transformer data corresponding to the group is obtained. The household change configuration unit deletes the power end data corresponding to the power end from the original household change relationship information, and adds it to the new household change relationship information corresponding to the transformer data.

[0012] The household change monitoring module includes a monitoring trigger unit and an information monitoring unit. The monitoring trigger unit is configured with a second trigger condition. When the power consumption terminal with the monitoring fitting mark meets the second trigger condition, a monitoring trigger signal is generated. The information monitoring unit responds to the monitoring trigger signal to re-acquire the corresponding branch power information and total power information, and re-inputs the obtained branch power information and total power information into the household change analysis module;

[0013] The line loss adjustment module includes an adjustment trigger unit and a line loss monitoring unit. The adjustment trigger unit is configured with a third trigger condition. When the power consumption end with an adjustment fitting mark meets the third trigger condition, a line loss trigger signal is generated. The line loss monitoring unit responds to the adjustment trigger signal to obtain the corresponding branch power information in real time, and calculates the branch line loss reference value based on the obtained branch power information, and updates the corresponding power consumption end data with the obtained branch line loss reference value.

[0014] Furthermore, the model building strategy includes a data acquisition step, a sample generation step, a model building step and a model training step. The data acquisition step includes acquiring historical power consumption data from branch power information and total power information respectively. The sample generation step includes generating power consumption samples according to historical power consumption data at the same time. The model building step includes building a line loss model equation:

[0015] y=β0+β1x1+β2x2+β3x3……+β p x p , where y is the total meter power value, β0 is the fixed loss error number of the substation, β1, β2, β3, ... β p is the loss error number of the power application end, x1, x2, x3, ... x p is the power value of the corresponding power application terminal;

[0016] The model training step includes bringing the power consumption samples into the training line loss model equation to obtain a line loss regression model.

[0017] Furthermore, the third trigger condition is that the power consuming end has a third benchmark number of continuous adjustment fitting marks, and the power consumption of the power consuming end at a corresponding time exceeds a third power consumption threshold.

[0018] Furthermore, the fixed loss error number of the station area is generated according to a fixed error calculation formula, and the fixed error calculation formula is β0=δ0+a1δ1+a2δ2......+a n δ n , where δ1, δ2, ... δ n is the branch line loss reference value corresponding to the power consumption end that meets the third trigger condition, where a1, a2, ... a n is a correction weight, and the correction weight is positively correlated with the number of adjustment trigger signals of the application terminal, and the δ0 is a preset reference station error number.

[0019] Furthermore, the household change monitoring module is configured with a monitoring information database, which stores electricity user information, the electricity user information including user type data and actual electricity usage data, the user type data reflects the type of the electricity user end, and the actual electricity usage data reflects the electricity usage of the electricity user end. The second trigger condition includes that the electricity user end has electricity user information stored in the monitoring information database, and the electricity consumption corresponding to the actual electricity usage data exceeds the second benchmark threshold.

[0020] Furthermore, the monitoring trigger signal includes monitoring period data, and the monitoring period data is generated according to the user type data and the power usage period corresponding to the second trigger condition.

[0021] Furthermore, it also includes a user management module, which obtains electricity user information by connecting to the user terminal network, and the user terminal is configured as a mobile terminal or a smart meter.

[0022] Furthermore, the first trigger condition is a first reference number of consecutive configuration fitting marks.

[0023] Furthermore, the horizontal fitting unit is configured with a horizontal correlation database, which stores horizontal correlation data. The horizontal correlation data reflects the correlation relationship between transformers, and an association priority value is configured for each power consumption end and transformer. The horizontal fitting unit fits the group to which the transformer belongs according to the order of the association priority values.

[0024] Furthermore, the first fitting threshold value ranges from 2% to 5%, and the second fitting threshold value ranges from 10% to 15%.

[0025] The technical effects of the present invention are mainly reflected in the following aspects: under the premise of no power outage, the line loss regression model is trained by using power samples, and then different power consumption ends are divided according to the actual fitting situation, and then the abnormal household-to-household relationship is screened, and at the same time, the actual power consumption information of the user that cannot be distinguished can be monitored, and then the determined household-to-household relationship can be used as a parameter for model training and adjustment, so that the accuracy of model fitting is improved, thereby identifying abnormal household-to-household relationship recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 : A topological principle diagram of a system for identifying the relationship between household transformers and power consumption in a transformer area according to the present invention;

[0027] Figure 2 : Schematic diagram of a data acquisition module of a system for identifying the topology of power consumption between households and transformers in a substation area according to the present invention;

[0028] Figure 3 : The abnormal correction comparison waveform of the household transformer relationship of the present invention;

[0029] Figure 4 : Schematic diagram of a household transformer analysis module of a system for identifying the relationship between household transformers in a transformer substation area;

[0030] Figure 5 : Schematic diagram of household transformer configuration module of a system for identifying power consumption topology of household transformer relationships in a transformer area of ​​the present invention;

[0031] Figure 6 : Schematic diagram of a household transformer monitoring module of a system for identifying the relationship between household transformers in a transformer substation area;

[0032] Figure 7: Schematic diagram of the line loss adjustment module of a power consumption topology identification system for a transformer-to-substation relationship in an area according to the present invention.

[0033] Attached figure numbers: 100, household change relationship database; 200, data acquisition module; 210, total power acquisition unit; 220, branch power acquisition unit; 300, information scheduling module; 400, household change analysis module; 410, model building unit; 420, deviation analysis unit; 430, deviation marking unit; 500, household change configuration module; 510, configuration trigger unit; 520, lateral fitting unit; 530, household change configuration unit; 600, household change monitoring module; 610, monitoring trigger unit; 620, information monitoring unit; 700, line loss adjustment module; 710, adjustment trigger unit; 720, line loss monitoring unit; 800, user management module. DETAILED DESCRIPTION

[0034] The specific implementation modes of the present invention are further described below in conjunction with the accompanying drawings to make the technical solutions of the present invention easier to understand and grasp.

[0035] Reference Figure 1 As shown, a system for identifying power consumption topology of household transformer relationships in a transformer substation area includes a household transformer relationship database 100, a data acquisition module 200, an information scheduling module 300, a household transformer analysis module 400, a household transformer configuration module 500, a household transformer monitoring module 600, and a line loss adjustment module 700;

[0036] The household-change relationship database 100 stores household-change relationship information, and the household-change relationship information includes transformer data and power-using end data; the household-change relationship information reflects the relationship between the transformer and the corresponding power-using end. For old residential areas, the household-change relationship is stored in the form of text files. Later, according to some changes in usage methods and line complexity, manual adjustments to the lines but no timely archiving, etc., errors may occur in the household-change relationship. This error may cause errors and impacts in the power grid big data analysis results. However, if the power outage method is used, it will inevitably cause a greater impact, and the carrier measurement method has certain requirements on the line quality, so both have disadvantages. The present invention preferentially formats and stores all known household-change relationship information, so that the original household-change relationship is constructed, and then the transformer data such as hardware parameters, location, power consumption, maintenance status, etc. are formatted and processed, and the power-using end data is processed in the same way, so that a data basis can be provided for subsequent big data corrections, and errors and mistakes in the data results will not be caused by format mismatches.

[0037] Reference Figure 2As shown, the data acquisition module 200 includes a total power acquisition unit 210 and a branch power acquisition unit 220. The total power acquisition unit 210 is used to collect the total power information of the transformer, and the branch power unit is used to collect the branch power information of each power consumption end. Since the entire power grid architecture has total transformer power collection and actual power collection of each household from the perspective of the substation, these two power collection modules are currently available. For traditional power collection, the branch power collection unit 220 or the total power collection unit 210 only needs to establish a corresponding formatting interface for real-time information collection and then format the information to obtain the daily substation power consumption value and the power consumption value of each power consumption end, so that the total power consumption at each time and the power consumption of each power consumption end can be obtained.

[0038] The information scheduling module 300 groups the received total power information and branch power information according to the household-to-household relationship information of the household-to-household relationship database 100; the information scheduling module 300 groups the received total power information and branch power information according to the household-to-household relationship database 100. Because the transmission logic and topology corresponding to different household meters and transformer meters may be different, it is impossible to analyze based on the transmission logic and topology. Therefore, the grouping relationship is matched according to the information in the database, and then the corresponding mathematical model is constructed for each group through this sub-tenancy relationship.

[0039] Reference Figure 4 As shown, the household change analysis module 400 includes a model building unit 410, a deviation analysis unit 420, and a deviation marking unit 430. The model building unit 410 is configured with a model building strategy. The model building strategy establishes a line loss regression model according to the total power information and branch power information of the same group. The deviation analysis unit 420 is based on the fitting degree of the line loss regression model corresponding to each branch power information of the group. The deviation marking unit 430 is configured with a first fitting threshold and a second fitting threshold. If the fitting degree corresponding to a branch power information is lower than the first fitting threshold, an adjustment fitting mark is established for the power consumption end data corresponding to the branch power information. If the fitting degree of the branch power information is between the first fitting threshold and the second fitting threshold, the adjustment fitting mark is established for the power consumption end data corresponding to the branch power information. If the fitting degree of the branch power information is between the fitting threshold and the second fitting threshold, a monitoring fitting mark is established for the power consumption end data corresponding to the branch power information; if the fitting degree of the branch power information is higher than the second fitting threshold, a configuration fitting mark is established for the power consumption end corresponding to the branch power information; the model building strategy includes a data acquisition step, a sample generation step, a model building step and a model training step, the data acquisition step includes acquiring historical power consumption data from the branch power information and the total power information respectively, the sample generation step includes generating power consumption samples according to the historical power consumption data at the same time, and the model building step includes building a line loss model equation: y = β0 + β1x1 + β2x2 + β3x3......+β p xp , where y is the total meter power value, β0 is the fixed loss error number of the substation, β1, β2, β3, ... β p is the loss error number of the power application end, x1, x2, x3, ... x p For the power value of the power application end; the model training step includes bringing the power consumption sample into the training line loss model equation to obtain the line loss regression model. The first fitting threshold value range is between 2%-5%, and the second fitting threshold value range is between 10%-15%. The household transformer anomalies between the substations can be solved by solving how to replace the household meter to achieve the stability of the household meter line loss, thereby realizing the identification function of the abnormal relationship between the household transformers in the substation. Two substations with household transformer relationship problems are generally reflected in the line loss. When the line loss of one substation increases, the line loss of the other substation decreases, showing a more drastic negative correlation fluctuation. If the line losses of the two substations are bundled and calculated, their total line loss will be relatively stable. Therefore, solving the household transformer relationship as a mathematical problem, that is, solving how to adjust the household transformer relationship, can make the line losses of the two substations relatively stable. Therefore, the first step is to establish a line loss regression model for each group corresponding to each substation. To establish a regression model, we first need to obtain enough samples. Since the power supply is continuously running, the power consumption information is continuously collected. By intercepting the power consumption in different time periods, we can obtain training samples. The power of all household meters in the substation is used as the independent variable of the regression equation, and the power of the total meter in the substation is used as the dependent variable for each linear regression. Based on the power supply relationship, we can derive the general relationship of the model as follows: y = β0 + β1x1 + β2x2 + β3x3...... + β p x p , and by bringing in samples, we can get β1, β2, β3, ... β pThe equation can be restored and the model training step is completed. If the electricity of this household meter belongs to a certain total meter, its corresponding regression coefficient will be close to 1, and its statistics will be relatively large. If the electricity of this household meter does not belong to a certain total meter, its corresponding regression coefficient will be close to 0, and its statistics will be relatively small. In this way, the statistical inaccuracy problems caused by statistical errors and line loss rates can be known, and then the fitting deviation value of each data can be obtained by bringing in actual data. For example, if the first power consumption end is to be fitted, the power consumption of other power consumption ends in the same period and the total power consumption can be obtained. The theoretical power consumption of the power consumption end can be calculated according to the model, and then the power consumption end of the power consumption end obtained according to the actual situation, the proportion of the difference to the theoretical power consumption is the fitting deviation value, so that each power consumption end can be fitted and verified in turn, and then marked according to the power consumption results. It should be noted that during the verified power consumption period, the corresponding power consumption end should be in a working state and the power consumption should generally exceed a certain threshold to avoid excessive deviation values. Then, by comparing the data in real time, each deviation value is marked. Generally, when the model is accurate, the deviation value will not be too large. If the deviation value exceeds 10%-15%, it means that there may be problems with the household change relationship. However, if the deviation value is less than 2%-5%, it means that the household change relationship should be accurate after verification. Therefore, the above situation is divided according to the deviation marking unit 430 and processed by different unit modules.

[0040] Reference Figure 5 As shown, the household change configuration module 500 includes a configuration trigger unit 510, a horizontal fitting unit 520 and a household change configuration unit 530. The configuration trigger unit 510 is configured with a first trigger condition. When the power consumption end with a configuration fitting mark meets the first trigger condition, a configuration trigger signal is generated. The horizontal fitting unit 520 responds to the configuration trigger signal. The horizontal fitting unit 520 fits the line loss regression model of other groups according to the branch power information of the power consumption end to calculate the corresponding fitting degree. If there is a line loss regression model with a fitting degree lower than the first fitting threshold, the line loss regression model is obtained. The transformer data corresponding to the group, the household transformer configuration unit 530 deletes the power consumption end data corresponding to the power consumption end from the original household transformer relationship information, and adds it to the new household transformer relationship information corresponding to the transformer data; first, if the deviation value is large, it means that there is a problem with the household transformer relationship, so a horizontal comparison should be performed to determine the correct household transformer relationship, so when it is marked and meets the first trigger condition, through horizontal comparison, this data is brought into other line loss regression models. If the fit is high, it means that the household transformer relationship is wrong, and the household transformer model can be adjusted to realize the correction of the household transformer relationship, as shown in the specific example. Figure 3As shown, if the corresponding situation does not appear, it means that there is a problem with the line loss regression model (it may be the original group or a new group), then it is necessary to wait for the line loss regression model to be corrected in real time. The first trigger condition is a configuration fitting mark with a continuous first reference number. The horizontal fitting unit 520 is configured with a horizontal association database, which stores horizontal association data. The horizontal association data reflects the association relationship between transformers. Each power consumption end and transformer are configured with an association priority value. The horizontal fitting unit 520 fits the group where the transformer is located according to the order of the association priority value. First, if continuous marks appear, it means that there is indeed an abnormal household-to-transformer relationship at this power consumption end, and then according to the horizontal fitting method, and preferably, the horizontal fitting is based on the priority relationship, and the association priority value is set according to the position relationship, physical details, hardware parameters, etc. between the corresponding transformers, that is, the closer the physical position and the closer the hardware parameter situation, the easier it is to have a household-to-transformer relationship error, so according to the association priority value, the amount of calculation can be effectively reduced and the efficiency of household-to-transformer relationship verification can be improved.

[0041] Reference Figure 6As shown, the household change monitoring module 600 includes a monitoring trigger unit 610 and an information monitoring unit 620. The monitoring trigger unit 610 is configured with a second trigger condition. When the power end with the monitoring fitting mark meets the second trigger condition, a monitoring trigger signal is generated. The information monitoring unit 620 responds to the monitoring trigger signal to re-acquire the corresponding branch power information and total power information, and re-inputs the obtained branch power information and total power information into the household change analysis module 400; the household change monitoring module 600 is configured with a monitoring information database, the monitoring information database stores power user information, the power user information includes user type data and actual power consumption data, the user type data reflects the type of the power end, and the actual power consumption data reflects the power consumption of the power end. The second trigger condition includes that the power end stores the power user information in the monitoring information database, and the power consumption corresponding to the actual power consumption data exceeds the second reference threshold. The monitoring trigger signal includes monitoring period data, and the monitoring period data is generated according to the user type data and the power consumption period corresponding to the second trigger condition. The function of the household change monitoring module 600 is to monitor the uncertain electricity users. The specific monitoring method is user feedback from the electricity user. For example, the user actually generates a request or monitors the power consumption provided by a household smart meter or a smart home monitoring system. According to the generated user power consumption information, the next monitoring time period is determined based on whether the power consumption information meets the conditions. That is, during the next monitoring time period, most users are in a power consumption state. For example, for residential users, the monitoring time period may be a peak time period of multiple power consumption. In this way, when the actual power consumption at the electricity user end is large, fitting can be performed to judge the correctness of the household change relationship of this electricity user end.

[0042] The user management module 800 is connected to the user end network to obtain the information of the electricity user, and the user end is configured as a mobile terminal or a smart meter. The user management module 800 can obtain the user's electricity usage information to ensure the user's electricity safety.

[0043] Reference Figure 7As shown, the line loss adjustment module 700 includes an adjustment trigger unit 710 and a line loss monitoring unit 720. The adjustment trigger unit 710 is configured with a third trigger condition. When the power end with an adjustment fitting mark meets the third trigger condition, a line loss trigger signal is generated. The line loss monitoring unit 720 responds to the adjustment trigger signal to obtain the corresponding branch power information in real time, and calculates the branch line loss reference value based on the obtained branch power information, and updates the corresponding power end data with the obtained branch line loss reference value. The third trigger condition is that the power end has a third reference number of continuous adjustment fitting marks, and the power consumption of the power end at the corresponding time exceeds the third power consumption threshold. The fixed loss error number of the substation is generated according to the fixed error calculation formula, and the fixed error calculation formula is β0=δ0+a1δ1+a2δ2......+a n δ n , where δ1, δ2, ... δ n is the branch line loss reference value corresponding to the power consumption end that meets the third trigger condition, where a1, a2, ... a n is a correction weight, and the correction weight is positively correlated with the number of adjustment trigger signals of the application terminal, and the δ0 is a preset reference station error number. As another key module of the present invention, the line loss adjustment unit achieves the convergence of the model under the control of big data by adjusting the fixed loss error number of the substation. Since the line loss regression model itself may contain abnormal samples and the abnormal samples are not distinguished, the abnormal samples will cause the line loss regression model to not converge to the accurate value. When the number of abnormal samples is large or the abnormal amount of the abnormal samples is large, the line loss regression samples are easily deviated. Therefore, the model adjusts the variable of the substation fixed loss, and obtains the branch line loss reference value through the known normal samples configured as the household-to-user relationship, and then corrects the substation line loss reference value. In this way, the line loss rate obtained by the normal sample can affect the model, so that the model can converge to the ideal model under the training of big data. In fact, although different power consumption ends in a substation are different, the physical correlation is large, for example, the construction time is similar, the construction distance is equivalent, and the power transmission materials and equipment used are roughly the same, so the line loss rate and the rate of change of the line loss rate will also be roughly the same, so the above correction strategy can converge the line loss regression model.

[0044] Of course, the above are only typical examples of the present invention. In addition, the present invention may also have many other specific implementations. All technical solutions formed by equivalent replacement or equivalent transformation fall within the scope of protection required by the present invention.

Claims

1. A system for identifying the power consumption topology of household transformers in a transformer area, characterized by: It includes household change relationship database, data collection module, information dispatch module, household change analysis module, household change configuration module, household change monitoring module and line loss adjustment module; The household-to-household change relationship database stores household-to-household change relationship information, and the household-to-household change relationship information includes transformer data and power consumption end data; The data collection module includes a total power collection unit and a branch power collection unit, wherein the total power collection unit is used to collect the total power information of the transformer, and the branch power collection unit is used to collect the branch power information of each power consumption end; The information scheduling module groups the received total power information and branch power information according to the household change relationship information in the household change relationship database; The household change analysis module includes a model building unit, a deviation analysis unit, and a deviation marking unit. The model building unit is configured with a model building strategy. The model building strategy establishes a line loss regression model according to the total power information and branch power information of the same group. The deviation analysis unit determines the degree of fit of the line loss regression model corresponding to each branch power information of the group. The deviation marking unit is configured with a first fitting threshold and a second fitting threshold. If the degree of fit corresponding to a branch power information is lower than the first fitting threshold, an adjustment fitting mark is established for the power consumption end data corresponding to the branch power information. If the degree of fit of the branch power information is between the first fitting threshold and the second fitting threshold, a monitoring fitting mark is established for the power consumption end data corresponding to the branch power information. If the degree of fit of the branch power information is higher than the second fitting threshold, a configuration fitting mark is established for the power consumption end corresponding to the branch power information. The household change configuration module includes a configuration trigger unit, a horizontal fitting unit and a household change configuration unit. The configuration trigger unit is configured with a first trigger condition. When the power end with a configuration fitting mark meets the first trigger condition, a configuration trigger signal is generated. The horizontal fitting unit responds to the configuration trigger signal. The horizontal fitting unit fits the line loss regression model of other groups according to the branch power information of the power end to calculate the corresponding fitting degree. If there is a line loss regression model with a fitting degree lower than the first fitting threshold, the transformer data corresponding to the group is obtained. The household change configuration unit deletes the power end data corresponding to the power end from the original household change relationship information, and adds it to the new household change relationship information corresponding to the transformer data. The household change monitoring module includes a monitoring trigger unit and an information monitoring unit. The monitoring trigger unit is configured with a second trigger condition. When the power consumption terminal with the monitoring fitting mark meets the second trigger condition, a monitoring trigger signal is generated. The information monitoring unit responds to the monitoring trigger signal to re-acquire the corresponding branch power information and total power information, and re-inputs the obtained branch power information and total power information into the household change analysis module; The line loss adjustment module includes an adjustment trigger unit and a line loss monitoring unit. The adjustment trigger unit is configured with a third trigger condition. When the power consumption end with an adjustment fitting mark meets the third trigger condition, a line loss trigger signal is generated. The line loss monitoring unit responds to the adjustment trigger signal to obtain the corresponding branch power information in real time, and calculates the branch line loss reference value based on the obtained branch power information, and updates the corresponding power consumption end data with the obtained branch line loss reference value.

2. A system for identifying power consumption topology of a transformer substation according to claim 1, characterized in that: The model building strategy includes a data acquisition step, a sample generation step, a model building step and a model training step. The data acquisition step includes acquiring historical power consumption data from branch power information and total power information respectively. The sample generation step includes generating power consumption samples according to historical power consumption data at the same time. The model building step includes building a line loss model equation: y=β0+β1x1+β2x2+β3x3……+β p x p , where y is the total meter power value, β0 is the fixed loss error number of the substation, β1, β2, β3, ... β p is the loss error number of the power application end, x1, x2, x3, ... x p is the power value of the corresponding power application terminal; The model training step includes bringing the power consumption samples into the training line loss model equation to obtain the line loss regression model.

3. A system for identifying power consumption topology of a transformer substation according to claim 2, characterized in that: The third trigger condition is that the power consuming end has a third reference number of continuous adjustment fitting marks, and the power consumption of the power consuming end at a corresponding time exceeds a third power consumption threshold.

4. A system for identifying power topology of household transformer relationships in an area as claimed in claim 3, characterized in that: The fixed loss error number of the station area is generated according to the fixed error calculation formula, and the fixed error calculation formula is β0=δ0+a1δ1+a2δ2......+a n δ n , where δ1, δ2, ... δ n is the branch line loss reference value corresponding to the power consumption end that meets the third trigger condition, where a1, a2, ... a n is a correction weight, and the correction weight is positively correlated with the number of adjustment trigger signals of the application terminal, and the δ0 is a preset reference station error number.

5. The system for identifying the power consumption topology of the household transformer relationship in the substation area according to claim 1, characterized in that: The household change monitoring module is configured with a monitoring information database, which stores electricity user information. The electricity user information includes user type data and actual electricity usage data. The user type data reflects the type of the electricity user end, and the actual electricity usage data reflects the electricity usage of the electricity user end. The second trigger condition includes that the electricity user end has electricity user information stored in the monitoring information database, and the electricity consumption corresponding to the actual electricity usage data exceeds the second benchmark threshold.

6. A system for identifying power consumption topology of a transformer substation and a household transformer as claimed in claim 5, characterized in that: The monitoring trigger signal includes monitoring period data, and the monitoring period data is generated according to user type data and a power usage period corresponding to the second trigger condition.

7. A system for identifying power consumption topology of a transformer substation and a household transformer as claimed in claim 6, characterized in that: It also includes a user management module, which obtains electricity user information by connecting to the user terminal network. The user terminal is configured as a mobile terminal or a smart meter.

8. The system for identifying the power consumption topology of the household transformer relationship in the substation area according to claim 1, characterized in that: The first trigger condition is a first reference number of consecutive configuration fit marks.

9. The system for identifying the power consumption topology of the household transformer relationship in the substation area according to claim 1, characterized in that: The horizontal fitting unit is configured with a horizontal association database, which stores horizontal association data. The horizontal association data reflects the association relationship between transformers, and an association priority value is configured for each power consumption end and transformer. The horizontal fitting unit fits the group to which the transformer belongs according to the order of the association priority values.

10. The system for identifying the power consumption topology of the household transformer relationship in the substation area according to claim 1, characterized in that: The first fitting threshold value ranges from 2% to 5%, and the second fitting threshold value ranges from 10% to 15%.

Images