CN113466578A - Rural power grid distribution area box table topological relation identification method and user electricity utilization monitoring method - Google Patents
Rural power grid distribution area box table topological relation identification method and user electricity utilization monitoring method Download PDFInfo
- Publication number
- CN113466578A CN113466578A CN202110586960.0A CN202110586960A CN113466578A CN 113466578 A CN113466578 A CN 113466578A CN 202110586960 A CN202110586960 A CN 202110586960A CN 113466578 A CN113466578 A CN 113466578A
- Authority
- CN
- China
- Prior art keywords
- box
- meter
- data
- tree
- voltage
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000012544 monitoring process Methods 0.000 title claims abstract description 38
- 230000005611 electricity Effects 0.000 title claims abstract description 28
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims description 14
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 13
- 238000012706 support-vector machine Methods 0.000 claims description 13
- 238000006073 displacement reaction Methods 0.000 claims description 11
- 238000010219 correlation analysis Methods 0.000 claims description 5
- 238000000926 separation method Methods 0.000 claims description 5
- 238000007670 refining Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 235000018185 Betula X alpestris Nutrition 0.000 claims 1
- 235000018212 Betula X uliginosa Nutrition 0.000 claims 1
- 238000012423 maintenance Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000013480 data collection Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention discloses a rural power grid distribution area box table topological relation recognition method and a user electricity consumption monitoring method. According to the method for monitoring the power consumption of the user, the state of the electric meter can be further obtained by utilizing an SVM algorithm according to the identification result of the box-meter relation, and the power consumption of the user can be monitored through the state of the electric meter, so that data support is provided for fault location of a subsequent power consumption system and line loss lean management of a transformer area.
Description
Technical Field
The invention relates to the technical field of box table topological relation identification, in particular to a box table topological relation identification method for a rural power grid distribution area and a user power utilization monitoring method.
Background
At present, a distribution network in a transformer area bears the key function of the tail end of a power network for conveying domestic electricity, and with the deepening of an intelligent power grid in the field of the distribution network, a rural power grid is an important basis for ensuring the normal development of electric power work, however, as rural power grids are more in users and complex in structure and are influenced by power utilization characteristics such as seasonal power and time-interval power, the rural power grids are low in voltage quality, frequent in failure and high in line loss rate, cannot adapt to and meet the power utilization requirements of rural areas in new periods, and cannot further promote the intellectualization of the distribution network in the rural areas. In addition, after the meter box is replaced again on site, the meter is replaced on site and the electric energy meter is stopped, the information needs to be manually updated again in the marketing system by operation and maintenance personnel on site, the metering automation system cannot actively synchronize files with the marketing system, files related to a distribution area in the metering automation system need to be manually and repeatedly issued, and the workload is large. The method is especially important for identifying the relation of the rural power grid distribution area box table to help operation and maintenance personnel to know the relation state of the distribution area box table in time and monitor the power utilization state of users.
The existing box meter topological relation identification method is mainly divided into two types, one type is based on a station area instantaneous power-off method, and the method has the advantages that: the checking result is accurate and reliable, and the recognition rate is high. The disadvantages are that: the power failure data needs to be acquired at a specific moment or scene, and the timeliness is not high; if a user in the transformer area has the conditions of meter box replacement, phase change and the like, if information maintenance is not timely, the method needs to be used again for identification, and power supply reliability indexes and user experience are affected due to repeated power failure. And field personnel with skilled service are actively matched, the conditions of positions, subordination relations and the like between households and meter boxes in the jurisdiction are clear, meter boxes with abnormal states can be conveniently and quickly positioned, and a large amount of manpower and material resources are wasted by repeatedly troubleshooting. The other type is based on a voltage zero-crossing time sequence and an SNR data fusion technology, hardware is used for collecting the difference between an electric energy meter and a route, the difference between the electric energy meter and a zero crossing and an SNR value, the identification accuracy is high, but the data collection is influenced by starting, and the normalization development cannot be carried out.
Therefore, how to provide an efficient and reliable method for identifying the topological relation of the box table of the rural power grid area is an urgent problem to be solved by the technical personnel in the field.
Disclosure of Invention
In view of the above, the invention provides a rural power grid platform area box table topological relation identification method and a user power consumption monitoring method, and effectively solves the problems that the existing box table topological relation identification method is low in timeliness, consumes manpower and material resources, cannot be normally developed and the like.
In order to achieve the purpose, the invention adopts the following technical scheme:
on one hand, the invention provides a method for identifying box table topological relation of a rural power grid area, which comprises the following steps:
acquiring box table data: acquiring ammeter voltage and ammeter box data of a target area, and dividing ammeter boxes and ammeters by taking a transformer area as a unit to obtain box meter data of the target area;
identifying box table relationships: and according to the voltage characteristics that the electric meters of the same meter box have the same frequency, carrying out correlation analysis on the electric meter voltage in the meter box data, clustering the electric meters of the same meter box into clusters, and obtaining a box meter relation identification result.
Further, the process of identifying the box table relationship is realized by a BIRCH algorithm, which specifically includes:
construction of CF: performing primary cluster classification on the electric meters according to the electric meter voltage data in the box meter data, calculating the distance between the electric meter voltage data, comparing the calculated distance with a preset distance threshold value between the electric meter data, dividing the electric meter voltage data into dense data and sparse data, classifying the dense data into electric meter basic clusters, and removing the sparse data;
constructing a CF tree: setting a distance threshold between a branching factor and clusters, and constructing a CF tree based on the basic clusters of the electric meter;
global clustering: operating all leaf nodes on the CF tree through global clustering or semi-global clustering, calculating the mass center of each sub-cluster, and representing each sub-cluster by using the mass center;
refining clusters: and taking the mass center as a seed, redistributing the voltage data of the electric meter to the nearest seed to obtain a new cluster, and updating the CF tree until a final tree structure is formed to obtain a box-meter relationship identification result.
BIRCH (Balanced Iterative reduction and Clustering using hierarchical structures) can identify the imbalance of data distribution in a data set, cluster points distributed in dense areas, and remove outliers that will be distributed in sparse areas. In addition, BIRCH is a method of incremental clustering where the clustering decision for each point is based on the data points that have currently been processed, rather than global data points.
The BIRCH algorithm uses two concepts of CF (Clustering Feature) and Clustering Feature tree for summarizing cluster description.
Further, after the step of building the CF tree and before the step of global clustering, the method further comprises the following steps:
simplifying the CF tree: and traversing all leaf nodes on the initialized CF tree, removing the abnormal points, reducing the clustering range and grouping.
The process of simplifying the CF tree is selectable, the step is to connect and construct a bridge of the CF tree and the global cluster, and the method is mainly used for reducing the cluster range and improving the data processing efficiency of the clustering process.
According to the box-meter topological relation identification method provided by the invention, the voltage characteristics can be identified through a clustering algorithm and difference analysis is carried out, so that the real box-meter relation is cleared, the box-meter relation identification accuracy is higher, and the identification process is more efficient and convenient.
On the other hand, the invention also provides a user electricity consumption monitoring method, which uses the rural power grid distribution area box table topological relation identification method and comprises the following steps:
constructing a model: according to the box table relation identification result, constructing and training by using historical voltage data and an SVM algorithm to obtain a user power utilization model;
monitoring power utilization: and inputting the newly added voltage data into the user power utilization model, outputting the monitoring state of the corresponding ammeter, and monitoring the user power utilization according to the monitoring state of the ammeter.
The user electricity utilization monitoring method provided by the invention can discover the disorder relation between the hidden meter box and the user electricity meter, can discover abnormal electricity utilization users, and is convenient to take measures to deal with the abnormal electricity utilization in time, thereby ensuring the electricity utilization safety of the users.
Further, the process of constructing the model specifically includes:
based on the box meter identification result, the voltage data of the electric meter is taken as a characteristic value, the monitoring state of the electric meter is taken as a result value, and a support vector machine is selected as a classifier;
in a sample plate space, dividing to obtain a hyperplane, wherein the hyperplane is expressed by a linear equation as follows:
wTx+b＝0
wherein, w is a normal vector and determines the direction of the hyperplane, b is a displacement term and determines the distance between the hyperplane and the origin;
by expressing the degree of confidence in the classification prediction by the function interval, assuming that the hyperplane can correctly classify the training samples, the following relationship exists:
two equations are combined as:
y(wTx+b)≥1
wherein y is the monitoring state of the ammeter, and x is voltage data;
normalizing the normal vector w, making | | | w | | | | 1, and changing the function interval into a geometric distance;
maximizing the function interval to obtain an equivalent model loss function;
constructing according to the model loss function to obtain a Lagrangian function;
according to the Lagrange function, solving minimum values of a normal vector w and a displacement term b, solving a maximum value of a Lagrange vector to obtain an optimal separation hyperplane, and constructing to obtain a user power utilization model;
and training the user power utilization model by taking the historical voltage data as a training set to obtain the trained user power utilization model.
Further, the function interval is changed to a geometric distance by the formula:
wherein r is a geometric distance, x is voltage data, y is a monitoring state of the ammeter, w is a normal vector, and b is a displacement term.
Further, the equivalent model loss function is:
wherein w is a normal vector and b is a displacement term.
Further, the lagrangian function is:
wherein alpha is a Lagrangian vector, w is a normal vector, and b is a displacement term.
According to the technical scheme, compared with the prior art, the rural power grid distribution area box table topological relation identification method and the user electricity consumption monitoring method are disclosed and provided, the rural power grid distribution area box table topological relation identification method can perform cluster analysis on voltage data of users under a distribution area, and therefore box table topological relation is accurately identified; according to the method for monitoring the power consumption of the user, the state of the electric meter can be further obtained by utilizing an SVM algorithm according to the identification result of the box-meter relation, and the power consumption of the user can be monitored through the state of the electric meter, so that data support is provided for fault location of a subsequent power consumption system and line loss lean management of a transformer area.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic flow chart of an implementation of a box table topology relationship identification method for a rural power grid area provided by the invention;
fig. 2 is a schematic flow chart of an implementation of a user electricity consumption monitoring method provided by the present invention;
FIG. 3 is a schematic diagram illustrating the identification of box table topology relationship and the monitoring of power consumption of a user according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of an implementation flow of a BIRCH clustering algorithm;
FIG. 5 is a schematic view of a visual display chart of a box-table relationship identification result;
fig. 6 is a graph illustrating the user electricity consumption monitoring data.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
On one hand, referring to fig. 1 and fig. 3, the embodiment of the invention discloses a method for identifying a topological relation of a box table of a rural power grid area, which comprises the following steps:
s11: acquiring box table data: and acquiring the ammeter voltage and the ammeter box data of the target area, and dividing the ammeter box and the ammeter by taking the transformer area as a unit to obtain the box meter data of the target area.
This embodiment acquires ammeter voltage and table case data from power consumption information acquisition system, and at this moment, ammeter and table case relation are mutually independent to the platform district is the unit, carries out the division of table case and ammeter, obtains the case table data of minim.
S12: identifying box table relationships: according to the fact that the electric meters of the same meter box have voltage characteristics of the same frequency, correlation analysis is conducted on the electric meter voltage in the meter box data, the electric meters of the same meter box are clustered, and a meter box relation identification result is obtained.
According to the electricity utilization characteristics and rules, the electric energy meters in the same meter box have the voltage characteristics of the same frequency, correlation analysis is carried out on the voltage data of the electric energy meters, the electric energy meters in the same meter box are clustered, and then the box meter relation identification is realized.
According to the embodiment of the invention, all the electric meters in the same region are clustered through the BIRCH algorithm, and if the electric meters and the corresponding cluster numbers are consistent, the electric meters are considered to be in the same meter box. Referring to fig. 4, the specific process is as follows:
step 1: the method comprises the steps of constructing CF, wherein the data collection and screening process comprises the steps of firstly carrying out primary cluster classification on electric meters according to voltage data, calculating distances among the voltage data, dividing the electric meters into a group when the electric meters are close to each other, considering points with long distances as outliers, and rejecting the outliers after finding the points. The CF also records the characteristic data of the number of meters in each cluster, the linear sum and the square sum of the voltage of the meters, and the like. The present embodiment calculates the distance between the voltage data by the euclidean distance calculation method.
Step 2: combining into a CF tree, wherein the process can be understood as a process of initializing cluster characteristics, after obtaining the basic cluster class of the electric meter, starting to construct the CF tree, setting a branching factor and a threshold T, wherein the branching factor is used for controlling the number of the CFs in the tree node, the node is split into two nodes when the number of the branching factor exceeds the number of the branching factor, the threshold T is used for controlling the distance between the CFs on the nodes, the distance between the voltage cluster classes of the electric meter is smaller than the threshold T, the points with close distances, namely the points smaller than the threshold T, are dense data, and construct the CF, and the points with farther distances, namely the points larger than or equal to the threshold T, are sparse data and need to be eliminated.
And step 3: and (3) narrowing the range, simplifying the clustering feature tree, wherein the process can be understood as a process for refining the clustering feature tree, the step is optional, the step is a bridge connecting the step 2 and the step 3, and similar to the step 2, the step needs to traverse leaf nodes of the initialized clustering feature tree, remove more abnormal points, narrow the range and perform grouping.
And 4, step 4: global clustering, where all leaf nodes are operated using global or semi-global clustering, a clustering algorithm with data points is easily adapted to a set of sub-clusters, each sub-cluster being represented by its cluster feature vector. The centroid of the sub-clusters is calculated and then each sub-cluster is represented by the centroid, which can capture the main distribution law of the data.
And 5: and (4) refining the clusters, because the data is only roughly summarized in the step 4, the original data is only scanned once, and the clusters need to be continuously improved. This step uses the center of the cluster (i.e., centroid) generated in the previous stage as a seed and reassigns the data points to the nearest seed to obtain a new set of clusters. And when a new electric meter cluster needs to be added into the CF tree, performing tree insertion operation, and correspondingly updating the CF tree until a final tree structure is formed.
This not only allows the migration of points belonging to that sub-cluster, but also ensures that all copies of a given data point migrate into the same sub-cluster. At the same time, an option to discard outliers is also provided. That is, if the closest point is too far away, the seed may be treated as an outlier and not included in the result.
The BIRCH algorithm can learn characteristics by self without specifying the number of clusters and classify, and can realize accurate box-table topological relation identification.
As shown in fig. 5, each point in the graph represents an electricity meter, and different shapes represent different meter boxes, so as to visually display the box-table relationship.
On the other hand, referring to fig. 2 and fig. 3, the embodiment of the present invention further discloses a user electricity consumption monitoring method, where the method uses the above-mentioned rural power grid block area box table topology relationship identification method, and includes:
s21: constructing a model: according to the box table relation identification result, constructing and training by using historical voltage data and an SVM algorithm to obtain a user power utilization model;
s22: monitoring power utilization: and inputting the newly added voltage data into the power utilization model of the user, outputting the monitoring state of the corresponding ammeter, and monitoring the power utilization of the user according to the monitoring state of the ammeter.
Referring to the box-meter relationship identification result obtained by the box-meter topological relationship identification method, in the embodiment, voltage data is used as a characteristic value x, the state (abnormal and normal) of the electric meter is used as a result value y, a Support Vector Machine (SVM) suitable for training of a small sample is selected as a classifier, the electricity consumption of the user is monitored, and if the voltage characteristic of the user is inconsistent with the historical state, the electricity consumption of the user is considered to be abnormal.
As a typical binary classification problem, an SVM (Support vector machine) is a linear classifier defined with the largest interval on a feature space, and the goal is to find a hyperplane in an n-dimensional data space, so that samples are classified into two classes and the interval between the hyperplane and data on both sides is the largest.
In this embodiment, a user power consumption model is constructed and trained by an SVM algorithm, so as to implement a user power consumption monitoring process, as follows:
in the template space, the division hyperplane can be described by the following linear equation:
wTx+b＝0
where w is the normal vector, which determines the direction of the hyperplane, and b is the displacement term, which determines the distance between the hyperplane and the origin. The separating hyperplane divides the sample data into two types, one is calculated according to the volume, and the other is calculated according to the requirement.
The function interval can be used to represent the confidence level of the classification prediction, and assuming that the hyperplane can correctly classify the training samples, the following relationship exists:
two equations are combined as:
y(wTx+b)≥1
wherein y is the monitoring state of the electricity meter;
however, if w and b are changed proportionally, the hyperplane is not changed, and the function interval becomes twice of the original one, so some constraint, such as normalization, needs to be added to the normal vector w, i.e., | w | | | 1, so that the function interval becomes the geometric distance, and the formula is:
to find a dividing hyperplane with "maximum separation", the separation needs to be maximized, i.e.
To maximize the spacing, only | | w | | non-woven cells need to be maximized-1Equivalent to minimizing | | w | | non-calculation2The loss function of the model is simplified as:
solving the constrained optimization problem, solving the dual problem by using a Lagrange multiplier method, and constructing a Lagrange function to obtain:
where α is the Lagrangian vector.
According to Lagrange duality, the duality problem of the original problem is the extremely small problem, w and b need to be firstly minimized, and then alpha needs to be maximized, so that an optimal separation hyperplane is obtained, further construction of a user electricity utilization model is achieved, and the user electricity utilization model can judge whether voltage data of the current ammeter are abnormal or not.
The SVM adds a fault tolerance mechanism, the error of the support vector machine on some samples is allowed, and in addition, the kernel function is introduced to solve the problem that training samples are inseparable in consideration of the complexity of data.
Therefore, historical voltage data is selected as a training set to train the power utilization model of the user, after the model is released, the newly added voltage data can be used for real-time state monitoring, if abnormity is found, an alarm can be given in time, public assets are maintained, and property loss of the user is avoided.
In this embodiment, after the user electricity utilization model is established, the voltage data of the electricity meter in the previous day is predicted regularly every day, if the voltage data is normal, the voltage data is returned to be normal, and if the voltage data is abnormal, the voltage data is returned to be abnormal, and an alarm is given in time. As shown in fig. 6, for monitoring data of the power consumption of the user, whether the power consumption of the user is normal or not can be obviously seen through the fluctuation condition of the voltage data, so that the power consumption state of the user can be more intuitively monitored.
Therefore, the method can clearly and accurately identify the voltage characteristics and perform difference analysis, so that the real box-meter relationship is cleared, the disorder relationship between the hidden meter box and the user meter or the abnormal electricity consumption customer is found, the accuracy of the box-meter relationship of the transformer area is improved, the checking of the higher-level topological relationship between the transformer-line-box-user is improved, and the operation and maintenance personnel can be helped to know the relation state of the box-meter relationship of the transformer area in time.
In summary, compared with the prior art, the method disclosed by the embodiment of the invention has the following advantages:
1. the method comprises the steps of clustering voltage data of users in a distribution area, carrying out feature analysis, simplifying optimization, sample training and the like, using a BITTH incremental clustering method to cluster the users into groups with the same number as that of filing meter boxes, using a classifier SVM to monitor electricity consumption of the users, and if the voltage characteristics of the users are inconsistent with the historical state, considering that the electricity consumption of the users is abnormal, so that accurate identification of a user-meter box relation and monitoring of the electricity consumption of the users are formed.
2. The table area box meter relation topology identification is to perform correlation analysis on voltage data of electric meters, perform multi-dimensional grouping according to voltage characteristics that electric energy meters of the same meter box have the same frequency, cluster the electric meters of the same meter box, and further realize analog meter box grouping calculation.
3. The power utilization habits of users can be reasonably analyzed, the power utilization condition can be timely monitored, and then data support is provided for follow-up accurate fault location and lean management of line loss of the transformer area.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A rural power grid distribution area box table topological relation identification method is characterized by comprising the following steps:
acquiring box table data: acquiring ammeter voltage and ammeter box data of a target area, and dividing ammeter boxes and ammeters by taking a transformer area as a unit to obtain box meter data of the target area;
identifying box table relationships: and according to the voltage characteristics that the electric meters of the same meter box have the same frequency, carrying out correlation analysis on the electric meter voltage in the meter box data, clustering the electric meters of the same meter box into clusters, and obtaining a box meter relation identification result.
2. The rural power grid platform area box table topological relation recognition method according to claim 1, wherein the box table relation recognition process is realized through a BIRCH algorithm, and specifically comprises the following steps:
construction of CF: performing primary cluster classification on the electric meters according to the electric meter voltage data in the box meter data, calculating the distance between the electric meter voltage data, comparing the calculated distance with a preset distance threshold value between the electric meter data, dividing the electric meter voltage data into dense data and sparse data, classifying the dense data into electric meter basic clusters, and removing the sparse data;
constructing a CF tree: setting a distance threshold between a branching factor and clusters, and constructing a CF tree based on the basic clusters of the electric meter;
global clustering: operating all leaf nodes on the CF tree through global clustering or semi-global clustering, calculating the mass center of each sub-cluster, and representing each sub-cluster by using the mass center;
refining clusters: and taking the mass centers as seeds, redistributing the voltage data of the electric meter to the nearest seeds to obtain new cluster types, updating the CF tree until a final tree structure is formed, and obtaining a box-meter relationship identification result by using the final tree structure.
3. The rural power grid block area box table topological relation recognition method according to claim 2, wherein after the step of building the CF tree and before the step of global clustering, the method further comprises the following steps:
simplifying the CF tree: and traversing all leaf nodes on the initialized CF tree, removing the abnormal points, reducing the clustering range and grouping.
4. A method for monitoring electricity consumption of users, which is characterized by using the method for identifying topological relation of box tables of rural power grids of any one of claims 1 to 3, and comprises the following steps:
constructing a model: according to the box table relation identification result, constructing and training by using historical voltage data and an SVM algorithm to obtain a user power utilization model;
monitoring power utilization: and inputting the newly added voltage data into the user power utilization model, outputting the monitoring state of the corresponding ammeter, and monitoring the user power utilization according to the monitoring state of the ammeter.
5. The method according to claim 4, wherein the process of constructing the model specifically comprises:
based on the box meter identification result, the voltage data of the electric meter is taken as a characteristic value, the monitoring state of the electric meter is taken as a result value, and a support vector machine is selected as a classifier;
in a sample plate space, dividing to obtain a hyperplane, wherein the hyperplane is expressed by a linear equation as follows:
wTx+b＝0
wherein, w is a normal vector and determines the direction of the hyperplane, b is a displacement term and determines the distance between the hyperplane and the origin;
by expressing the degree of confidence in the classification prediction by the function interval, assuming that the hyperplane can correctly classify the training samples, the following relationship exists:
two equations are combined as:
y(wTx+b)≥1
wherein y is the monitoring state of the ammeter, and x is voltage data;
normalizing the normal vector w, making | | | w | | | | 1, and changing the function interval into a geometric distance;
maximizing the function interval to obtain an equivalent model loss function;
constructing according to the model loss function to obtain a Lagrangian function;
according to the Lagrange function, solving minimum values of a normal vector w and a displacement term b, solving a maximum value of a Lagrange vector to obtain an optimal separation hyperplane, and constructing to obtain a user power utilization model;
and training the user power utilization model by taking the historical voltage data as a training set to obtain the trained user power utilization model.
6. The method according to claim 5, wherein the function interval is changed into a geometric distance according to the formula:
wherein r is a geometric distance, x is voltage data, y is a monitoring state of the ammeter, w is a normal vector, and b is a displacement term.
7. The method according to claim 5, wherein the equivalent model loss function is:
wherein w is a normal vector and b is a displacement term.
8. The method according to claim 5, wherein the Lagrangian function is:
wherein alpha is a Lagrangian vector, w is a normal vector, and b is a displacement term.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110586960.0A CN113466578A (en) | 2021-05-27 | 2021-05-27 | Rural power grid distribution area box table topological relation identification method and user electricity utilization monitoring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110586960.0A CN113466578A (en) | 2021-05-27 | 2021-05-27 | Rural power grid distribution area box table topological relation identification method and user electricity utilization monitoring method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113466578A true CN113466578A (en) | 2021-10-01 |
Family
ID=77871585
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110586960.0A Pending CN113466578A (en) | 2021-05-27 | 2021-05-27 | Rural power grid distribution area box table topological relation identification method and user electricity utilization monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113466578A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114091608A (en) * | 2021-11-24 | 2022-02-25 | 国网河南省电力公司许昌供电公司 | Data mining-based user variable relationship identification method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105023092A (en) * | 2015-06-24 | 2015-11-04 | 国网山东省电力公司电力科学研究院 | Power load monitoring method based on electric quantity feature analysis |
| CN108628841A (en) * | 2017-03-22 | 2018-10-09 | 湖南本来文化发展有限公司 | The APP of Guangdong language accent and English is translated based on BIRCH clustering algorithms |
| CN109345013A (en) * | 2018-09-25 | 2019-02-15 | 广西电网有限责任公司电力科学研究院 | A kind of distribution user power utilization behavior prediction method |
| CN109685277A (en) * | 2018-12-28 | 2019-04-26 | 国网冀北电力有限公司经济技术研究院 | Electricity demand forecasting method and device |
| CN109712641A (en) * | 2018-12-24 | 2019-05-03 | 重庆第二师范学院 | A kind of processing method of audio classification and segmentation based on support vector machines |
| CN110569316A (en) * | 2019-08-07 | 2019-12-13 | 浙江大学 | low-voltage distribution area user topology identification method based on t-SNE dimension reduction technology and BIRCH clustering |
| CN111026927A (en) * | 2019-12-26 | 2020-04-17 | 国网福建省电力有限公司电力科学研究院 | Low-voltage transformer area running state intelligent monitoring system |
| CN111162608A (en) * | 2020-01-08 | 2020-05-15 | 国网湖北省电力有限公司电力科学研究院 | Distribution transformer area topology identification and verification method based on correlation analysis |
| CN112000917A (en) * | 2020-10-30 | 2020-11-27 | 杭州电子科技大学 | Low-voltage user meter box identification method integrating known phase difference and address information |
-
2021
- 2021-05-27 CN CN202110586960.0A patent/CN113466578A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105023092A (en) * | 2015-06-24 | 2015-11-04 | 国网山东省电力公司电力科学研究院 | Power load monitoring method based on electric quantity feature analysis |
| CN108628841A (en) * | 2017-03-22 | 2018-10-09 | 湖南本来文化发展有限公司 | The APP of Guangdong language accent and English is translated based on BIRCH clustering algorithms |
| CN109345013A (en) * | 2018-09-25 | 2019-02-15 | 广西电网有限责任公司电力科学研究院 | A kind of distribution user power utilization behavior prediction method |
| CN109712641A (en) * | 2018-12-24 | 2019-05-03 | 重庆第二师范学院 | A kind of processing method of audio classification and segmentation based on support vector machines |
| CN109685277A (en) * | 2018-12-28 | 2019-04-26 | 国网冀北电力有限公司经济技术研究院 | Electricity demand forecasting method and device |
| CN110569316A (en) * | 2019-08-07 | 2019-12-13 | 浙江大学 | low-voltage distribution area user topology identification method based on t-SNE dimension reduction technology and BIRCH clustering |
| CN111026927A (en) * | 2019-12-26 | 2020-04-17 | 国网福建省电力有限公司电力科学研究院 | Low-voltage transformer area running state intelligent monitoring system |
| CN111162608A (en) * | 2020-01-08 | 2020-05-15 | 国网湖北省电力有限公司电力科学研究院 | Distribution transformer area topology identification and verification method based on correlation analysis |
| CN112000917A (en) * | 2020-10-30 | 2020-11-27 | 杭州电子科技大学 | Low-voltage user meter box identification method integrating known phase difference and address information |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114091608A (en) * | 2021-11-24 | 2022-02-25 | 国网河南省电力公司许昌供电公司 | Data mining-based user variable relationship identification method |
| CN114091608B (en) * | 2021-11-24 | 2024-02-20 | 国网河南省电力公司许昌供电公司 | Household variable relation identification method based on data mining |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI440862B (en) | Electrical detection method and system based on user feedback information | |
| CN104809658B (en) | A kind of rapid analysis method of low-voltage distribution network taiwan area line loss | |
| CN113267692B (en) | Low-voltage transformer area line loss intelligent diagnosis and analysis method and system | |
| CN109904925B (en) | Power distribution station area user transformation relation checking method | |
| CN104834984A (en) | Electric power transaction supervision risk early warning system based on unified and interconnected electric power market | |
| CN104408667A (en) | Comprehensive assessment method and system of power quality | |
| CN114048870A (en) | Power system abnormity monitoring method based on log characteristic intelligent mining | |
| CN112366690B (en) | Low-voltage area transverse node relation identification method based on strong synchronous voltage characteristic matching | |
| Trindade et al. | Data analytics in smart distribution networks: Applications and challenges | |
| CN111612056A (en) | Low-pressure customer variation relation identification method based on fuzzy clustering and zero-crossing offset | |
| CN113466578A (en) | Rural power grid distribution area box table topological relation identification method and user electricity utilization monitoring method | |
| Li et al. | Identification of distribution network topology parameters based on multidimensional operation data | |
| CN110597792A (en) | Multistage redundant data fusion method and device based on synchronous line loss data fusion | |
| CN111191883B (en) | Intelligent distribution transformer terminal micro-application selection method and system | |
| CN115908082A (en) | Enterprise pollution discharge monitoring method and device based on electricity utilization characteristic indexes | |
| CN109242132A (en) | Subregion peak load prediction technique based on MapReduce frame | |
| Ya’An | Application of artificial intelligence in computer network technology in the era of big data | |
| CN113259380B (en) | Method and device for detecting network attack of micro-grid in mountainous area | |
| Zhao et al. | Research on Line-transformer-user Topological Anomaly Recognition Model Based on Multi-source Data Mining | |
| Hu et al. | A Bad Data Identification Method for Multiple Spatio-temporal Data in Power Distribution Network | |
| Yu et al. | Troubleshooting and Traceability Method Based on MapReduce Big Data Platform and Improved Genetic Reduction Algorithm for Smart Substation | |
| Li et al. | Study on the extraction method of high-risk operation scenarios under high proportion wind power grid-connected power system | |
| Bing et al. | Research on Power Quality Demand Analysis Model for Multi User by Integrated Neural Network | |
| CN117477531A (en) | Identification method and system for 220kV side bus topological structure of transformer substation | |
| Li et al. | Design of Intelligent Distribution Network Planning Assistant Decision Algorithm Based on Machine Learning Algorithm |
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 |