CN114298861A - Power failure range analysis method and system based on power failure correlation clustering - Google Patents
Power failure range analysis method and system based on power failure correlation clustering Download PDFInfo
- Publication number
- CN114298861A CN114298861A CN202111619974.4A CN202111619974A CN114298861A CN 114298861 A CN114298861 A CN 114298861A CN 202111619974 A CN202111619974 A CN 202111619974A CN 114298861 A CN114298861 A CN 114298861A
- Authority
- CN
- China
- Prior art keywords
- power failure
- user
- power
- correlation coefficient
- data
- 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
Images
Abstract
The invention relates to the technical field of power distribution network fault analysis, and discloses a power failure range analysis method and a power failure range analysis system based on power failure correlation clustering.
Description
Technical Field
The invention relates to the technical field of power distribution network fault analysis, in particular to a power failure range analysis method and system based on power failure correlation clustering.
Background
At present, the informatization level of a distribution network area distribution user side is low, the management is disordered, the service level of a power user still has a short board, and the problems that the power failure information of the user is inaccurate, the active alarm cannot be realized, the power supply recovery time is long and the like generally exist. When the low-voltage side breaks down, because the low-voltage wiring information of the transformer area is lost or inaccurate, the power failure fault of the low-voltage transformer area is mostly determined by the fault report of a user telephone and the troubleshooting of rush-repair personnel one by one, the active rush-repair of the low-voltage side is difficult to realize, and the power utilization service experience of a user is influenced. Meanwhile, the current low-voltage side has the problems of large investment in the early stage and high later maintenance cost in the technical mode of realizing power failure alarm by additionally installing a large number of monitoring devices, and is difficult to popularize.
For a long time, the power failure emergency repair treatment is passive, information of power failure places, power failure scales and the like is not accurately acquired, even the situation that a user does not find power failure at home for a long time to cause power failure fault continuation and cause greater damage to user property and equipment exists, and then the problems that the user complains or requires power supply company compensation and the like are caused.
In the field of power operation and maintenance, a common analysis unit for minimum power failure is divided to carry out power failure analysis, and power quantity of equipment is called to judge whether the equipment is misinformed or not by receiving events of a terminal and a meter; judging whether the power failure of the platform area level terminal exists or not; after the station area level power failure is eliminated, judging a minimum power failure common analysis unit according to power failure events of all terminals or meters under the station area, and carrying out power failure research and judgment on the minimum power failure event common unit; and if the non-minimum power failure common unit fails, judging from the next level of the minimum power failure common unit. And finally, confirming the hierarchy power failure and the power failure state related to the power failure event, and finally obtaining the minimum power failure common unit for confirming the power failure.
For the research on the power failure correlation of a user, the power failure range of a power distribution network is analyzed through the connection relation in the prior art, the power failure range is rapidly located through the system according to the connection relation of distribution network equipment, and then the power failure influence range is obtained according to the query of a marketing system database.
However, in the analysis and judgment of the power failure range, the positioning needs to be performed through the connection relationship or the topological structure of the distribution network, but the informatization level of the distribution network station distribution user side is low, the management is disordered, and the low-voltage wiring information is missing or inaccurate, so that the power failure range is positioned through the connection relationship of the distribution network equipment, and a large error exists.
Disclosure of Invention
The invention provides a power failure range analysis method and system based on power failure correlation clustering, and solves the technical problem that the power failure range positioning error is large in the prior art.
In view of this, the first aspect of the present invention provides a method for analyzing a power outage range based on a power outage correlation cluster, including the following steps:
s1, acquiring power consumption data of each user in the distribution room in a preset time period based on the user side meter;
s2, performing clustering analysis on the power consumption data of each user by adopting a K-means clustering algorithm, so as to classify the power consumption data with the highest similarity into the same cluster group and obtain a plurality of cluster groups;
s3, calculating a power failure correlation coefficient between every two pieces of power consumption data in the same cluster group, generating a correlation coefficient matrix according to the power failure correlation coefficient, and selecting a user with the highest power failure correlation coefficient as a characteristic user according to the correlation coefficient matrix;
and S4, based on the power failure active alarm signal sent by the feature user in advance, polling the power failure state of the station area where the feature user is located, the feeder layer connected with the feature user and other users of the cluster group where the feature user is located by adopting a polling technology, thereby determining the power failure range.
Preferably, the method further comprises:
and freezing the electricity consumption data of each user in the distribution room according to the preset freezing time, and sending the frozen electricity consumption data to a metering center.
Preferably, step S2 specifically includes:
s201, forming a data set X (X) by using electricity consumption data1，x2,…,xn) In the data set X ═ X (X)1，x2,…,xn) Selecting K samples as initial clustering center x ═ (x) at random1，x2,…,xk) Defining a set of cluster groups corresponding to each of the initial cluster centers as S ═ S1,S2,…,Sk}；
S202, calculating the shortest Euclidean distance between each sample in the data set and the initial clustering center according to the following formula 1,
in formula 1, D (x) represents the shortest Euclidean distance, xiRepresents the ith sample, μ in the datasetiRepresents a cluster center, wherein the cluster center μiAs indicated by the general representation of the,
s203, calculating the selection probability P (x) of each sample being selected as the next cluster center according to the shortest Euclidean distance between each sample and the initial cluster center by the following formula 3,
s204, selecting a sample as a next clustering center according to the selection probability of each sample by a wheel disc method, and repeating the step until k next clustering centers are selected;
and S205, calculating the shortest Euclidean distance between each sample in the data set and k next clustering centers by using the formula 1, and distributing each sample to the cluster group corresponding to the clustering center with the smallest distance.
Preferably, step S3 specifically includes:
s301, generating an electric quantity data matrix based on two electric quantity data in the same cluster group, and recording the electric quantity data matrix as M, wherein M is an M multiplied by b dimensional matrix, M is the number of types of electric quantity data, and b is the number of users in the same cluster group;
s302, calculating the power failure correlation coefficient between every two pieces of power consumption data in the same cluster group by using the correlation coefficient calculation formula of the following formula 4,
in equation 4, ρ represents a power outage correlation coefficient, atA data matrix representing the electricity usage of meter a during a period t,
represents the average value of the electric meter a in the whole time period, T represents the total time period, gtA data matrix representing the electricity consumption of the electricity meter g during the period t,
represents the average value of the electricity meter g in the whole time period, LaaMean square sum, L, of the data matrix representing the power consumption of meter a during a period tggRepresents the sum of the squares of the mean differences of the electricity usage data matrix of the electricity meter g over the period t, cov (a, g) represents the covariance of the electricity meter a and the electricity meter g, wherein,
s303, generating a corresponding correlation coefficient matrix according to the power failure correlation coefficients of the same cluster group;
and S304, selecting the user with the highest power failure correlation coefficient as a characteristic user according to the correlation coefficient matrix.
Preferably, step S4 specifically includes:
s401, when a power failure signal is detected through an alarm ammeter pre-installed by the feature user, sending a power failure active alarm signal to the concentrator to which the feature user belongs, and reporting to a master station through the concentrator;
s402, acquiring the number of power failure active alarm signals received by the master station from the same station area within the same time margin, and executing the step S405 if the number of the power failure active alarm signals is equal to 1; if the number of the power failure active alarm signals is more than 1, executing the steps S403-S405;
s403, polling alarm electric meters at all loads directly accessed by the outlet end of the low-voltage side of the transformer area where the characteristic user is located to obtain power consumption data of all the loads, and determining the power-off state of the whole area, wherein the power-off state comprises a normal power consumption state and a power failure state;
s404, determining all feeder nodes connected with the characteristic user according to the topological connection relation of the feeders, and polling the alarm electric meters of all the feeder nodes to obtain the electricity consumption data of all the feeder nodes so as to determine the electricity stop state of each feeder node;
s405, polling the rest other users in the cluster group where the feature user is located according to the power failure correlation coefficient of the feature user in sequence to obtain power consumption data of each rest other user, so as to determine the power failure stop state of each user in the cluster group corresponding to the same moment, and executing the step S406;
s406, determining a power failure range according to the power failure state of the distribution area corresponding to the characteristic user, the power failure state of each feeder line node and the power failure state of each user in the cluster group.
In a second aspect, the present invention further provides a system for analyzing a power outage range based on a power outage correlation clustering, including:
the power consumption acquisition module is used for acquiring power consumption data of each user in the distribution area in a preset time period based on the user side meter;
the clustering module is used for clustering and analyzing the power consumption data of each user by adopting a K-means clustering algorithm, so that the power consumption data with the highest similarity is classified into the same cluster group to obtain a plurality of cluster groups;
the correlation calculation module is used for calculating a power failure correlation coefficient between every two pieces of power consumption data in the same cluster group, generating a correlation coefficient matrix according to the power failure correlation coefficient, and selecting a user with the highest power failure correlation coefficient as a characteristic user according to the correlation coefficient matrix;
and the power failure analysis module is used for polling the power failure state of the distribution area where the characteristic user is located, the feeder layer connected with the characteristic user and other users of the cluster group where the characteristic user is located by adopting a polling technology based on the power failure active alarm signal sent by the characteristic user in advance, so that the power failure range is determined.
Preferably, the system further comprises:
and the freezing module is used for freezing the electricity consumption data of each user in the distribution room according to the preset freezing time and sending the frozen electricity consumption data to the metering center.
Preferably, the clustering module specifically includes:
an initial module, configured to form a data set X ═ X from the electricity consumption data1，x2,…,xn) In the data set X ═ X (X)1，x2,…,xn) Selecting K samples as initial clustering center x ═ (x) at random1，x2,…,xk) Defining a set of cluster groups corresponding to each of the initial cluster centers as S ═ S1,S2,…,Sk}；
A distance calculation module for calculating the shortest Euclidean distance between each sample in the data set and the initial clustering center by the following formula 1,
in formula 1, D (x) represents the shortest Euclidean distance, xiRepresents the ith sample, μ in the datasetiRepresents a cluster center, wherein the cluster center μiAs indicated by the general representation of the,
a probability calculation module for calculating a selection probability P (x) of each sample being selected as a next cluster center by the following formula 3 according to each sample and a shortest Euclidean distance from an initial cluster center,
the cluster updating module is used for selecting a sample as a next cluster center according to the selection probability of each sample and a roulette method, so that k next cluster centers are selected;
and the clustering module is used for calculating the shortest Euclidean distance between each sample in the data set and k next clustering centers through formula 1, and distributing each sample to the cluster group corresponding to the clustering center with the smallest distance.
Preferably, the correlation calculation module specifically includes:
the matrix generation module is used for generating an electric quantity data matrix based on two electric quantity data in the same cluster group, and the electric quantity data matrix is marked as M, wherein M is an M multiplied by b dimensional matrix, M is the number of types of electric quantity data, and b is the number of users in the same cluster group;
a correlation coefficient calculation module for calculating the power failure correlation coefficient between two power consumption data in the same cluster group by the correlation coefficient calculation formula of the following formula 4,
in equation 4, ρ represents a power outage correlation coefficient, atA data matrix representing the electricity usage of meter a during a period t,
represents the average value of the electric meter a in the whole time period, T represents the total time period, gtA data matrix representing the electricity consumption of the electricity meter g during the period t,
represents the average value of the electricity meter g in the whole time period, LaaMean square sum, L, of the data matrix representing the power consumption of meter a during a period tggRepresents the sum of the squares of the mean differences of the electricity usage data matrix of the electricity meter g over the period t, cov (a, g) represents the covariance of the electricity meter a and the electricity meter g, wherein,
the matrix module is used for generating a corresponding correlation coefficient matrix according to the power failure correlation coefficients of the same cluster group;
and the characteristic user determining module is used for selecting the user with the highest power failure correlation coefficient as the characteristic user according to the correlation coefficient matrix.
Preferably, the power outage analysis module specifically includes:
the alarm module is used for sending a power failure active alarm signal to the concentrator when an alarm ammeter pre-installed by the feature user detects a power failure signal, and reporting the power failure active alarm signal to the main station through the concentrator;
the judging module is used for acquiring the number of power failure active warning signals received by the master station from the same station area within the same time margin;
the transformer area polling module is used for polling alarm electric meters at all loads directly accessed by the outlet end of the low-voltage side of the transformer area where the characteristic user is located so as to obtain power consumption data of all the loads, and accordingly determining the power-off state of the whole transformer area, wherein the power-off state comprises a normal power-on state and a power-off state;
the feeder polling module is used for determining all feeder nodes connected with the characteristic users according to the feeder topological connection relation, and polling the alarm electric meters of all the feeder nodes to obtain the electricity consumption data of all the feeder nodes so as to determine the electricity-out state of each feeder node;
the cluster group polling module is used for sequentially polling the rest other users in the cluster group where the characteristic user is located according to the power failure correlation coefficient of the characteristic user so as to obtain the power consumption data of each rest other user, and therefore the power failure state of each user in the cluster group corresponding to the same moment is determined;
and the power failure range determining module is used for determining the power failure range according to the power failure state of the distribution area corresponding to the characteristic user, the power failure state of each feeder node and the power failure state of each user in the cluster group.
According to the technical scheme, the invention has the following advantages:
the method comprises the steps of carrying out clustering analysis on power consumption data of each user through a K-means clustering algorithm, classifying the power consumption data with the highest similarity into the same cluster group, calculating a power failure correlation coefficient between every two power consumption data in the same cluster group, generating a correlation coefficient matrix, determining corresponding characteristic users, and carrying out active alarm on power failure of the characteristic users.
Drawings
Fig. 1 is a flowchart of a method for analyzing a power outage range based on a power outage correlation clustering provided in an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a system for analyzing a power outage range based on a power outage correlation cluster according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, 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.
For convenience of understanding, please refer to fig. 1, the method for analyzing the power outage range based on the power outage correlation clustering includes the following steps:
and S1, acquiring the electricity consumption data of each user in the distribution room in a preset time period based on the user side meter.
And S2, performing clustering analysis on the power consumption data of each user by adopting a K-means clustering algorithm, and classifying the power consumption data with the highest similarity into the same cluster group to obtain a plurality of cluster groups.
S3, calculating a power failure correlation coefficient between every two power consumption data in the same cluster group, generating a correlation coefficient matrix according to the power failure correlation coefficient, and selecting a user with the highest power failure correlation coefficient as a characteristic user according to the correlation coefficient matrix.
And S4, polling the power-off state of the station area where the characteristic user is located, the feeder layer connected with the characteristic user and other users of the cluster group where the characteristic user is located by adopting a polling technology based on the power-off active alarm signal sent by the characteristic user in advance, so as to determine the power-off range.
The embodiment provides a power failure range analysis method based on power failure correlation clustering, which comprises the steps of carrying out clustering analysis on power consumption data of all users through a K-means clustering algorithm, classifying the power consumption data with the highest similarity into the same cluster group, calculating a power failure correlation coefficient between every two power consumption data in the same cluster group, generating a correlation coefficient matrix, determining corresponding characteristic users, and carrying out active power failure alarm based on the characteristic users.
In one embodiment, the method further comprises:
and freezing the electricity consumption data of each user in the distribution room according to the preset freezing time, and sending the frozen electricity consumption data to a metering center.
In an example, the electricity consumption data can be frozen in a fixed point by taking days as a unit, and all the frozen electricity consumption data in the day are uploaded to a metering center, and errors in communication, statistics and the like may occur in the long-term data transmission process, which may cause deviation of original data, so that when data is extracted through a 'marketing-distribution' service data platform, the original data needs to be preprocessed, and the preprocessing mainly includes cleaning of error data, repairing of abnormal data and data denoising.
When similarity comparison is performed between users, data selected by two users need to be in the same time period, if electric quantity data of one user is not collected successfully at a certain event point, the data of the event point of the two users should be deleted, and correlation coefficient calculation is performed on the rest data.
In a specific embodiment, step S2 specifically includes:
s201, forming a data set X (X) by using electricity consumption data1，x2,…,xn) In the data set X ═ X (X)1，x2,…,xn) Selecting K samples as initial clustering center x ═ (x) at random1，x2,…,xk) Defining a set of cluster groups corresponding to each of the initial cluster centers as S ═ S1,S2,…,Sk}；
S202, calculating the shortest Euclidean distance between each sample in the data set and the initial clustering center according to the following formula 1,
in formula 1, D (x) represents the shortest Euclidean distance, xiRepresents the ith sample, μ in the datasetiRepresents a cluster center, wherein the cluster center μiAs indicated by the general representation of the,
s203, calculating the selection probability P (x) of each sample being selected as the next cluster center according to the shortest Euclidean distance between each sample and the initial cluster center by the following formula 3,
s204, selecting a sample as a next clustering center according to the selection probability of each sample by a wheel disc method, and repeating the step until k next clustering centers are selected;
and S205, calculating the shortest Euclidean distance between each sample in the data set and k next clustering centers by using the formula 1, and distributing each sample to the cluster group corresponding to the clustering center with the smallest distance.
It can be understood that, because the number of the users in the distribution room is large, the electric quantity data is continuously updated along with the time, the increase is rapid, the calculation of the power failure correlation coefficient for all the users is inconvenient, the users are divided into a plurality of clustering clusters, and the analysis and the judgment in the clustering clusters are more convenient.
In a specific embodiment, step S3 specifically includes:
s301, generating an electric quantity data matrix based on two electric quantity data in the same cluster group, and recording the electric quantity data matrix as M, wherein M is an M multiplied by b dimensional matrix, M is the number of types of electric quantity data, and b is the number of users in the same cluster group;
s302, calculating the power failure correlation coefficient between every two pieces of power consumption data in the same cluster group by using the correlation coefficient calculation formula of the following formula 4,
in equation 4, ρ represents a power outage correlation coefficient, atA data matrix representing the electricity usage of meter a during a period t,
represents the average value of the electric meter a in the whole time period, T represents the total time period, gtA data matrix representing the electricity consumption of the electricity meter g during the period t,
represents the average value of the electricity meter g in the whole time period, LaaMean square sum, L, of the data matrix representing the power consumption of meter a during a period tggRepresents the sum of the squares of the mean differences of the electricity usage data matrix of the electricity meter g over the period t, cov (a, g) represents the covariance of the electricity meter a and the electricity meter g, wherein,
s303, generating a corresponding correlation coefficient matrix according to the power failure correlation coefficients of the same cluster group;
and S304, selecting the user with the highest power failure correlation coefficient as a characteristic user according to the correlation coefficient matrix.
In one example, based on the research of the power outage correlation coefficient among users in the same cluster, the correlation degree grade of the corresponding attribute and the power outage correlation can be given. The correlation coefficient is between negative 1 and positive 1, and the larger the absolute value of the correlation coefficient is, the higher the power failure correlation degree between the represented users is, so that the interval is averagely divided into 8 relation grades, namely 4 negative correlation grades and 4 positive correlation grades.
In a specific embodiment, step S4 specifically includes:
s401, when a power failure signal is detected through an alarm ammeter pre-installed by a feature user, sending a power failure active alarm signal to the corresponding concentrator, and reporting to the master station through the concentrator.
The alarm ammeter is an intelligent ammeter with a super capacitor and has a function of detecting a power failure state. When a power failure event occurs, the ammeter can send an alarm signal to the concentrator, and the alarm signal is reported to the master station through the concentrator, so that active alarm based on the characteristic users is realized.
S402, acquiring the number of power failure active alarm signals received by the master station from the same station area within the same time margin, and executing the step S405 if the number of the power failure active alarm signals is equal to 1; if the number of the active warning signals is greater than 1, executing steps S403 to S405.
In the embodiment, the alarm conditions are classified into single-user alarm and multi-user alarm, and if the number of active alarm signals in power failure is equal to 1, the alarm signals can be classified as single-user alarm; if a plurality of alarm signals occur simultaneously, the alarm signals are classified as multi-user alarms. When multi-user alarm is judged, the calculated power failure correlation coefficient and the position relation of clustering distribution can be combined to form the polling priority, and polling is carried out in a layered mode in sequence.
And S403, polling alarm electric meters at all loads directly accessed by the outlet end of the low-voltage side of the transformer area where the characteristic user is located to obtain the power consumption data of all the loads, so as to determine the power-off state of the whole transformer area, wherein the power-off state comprises a normal power-on state and a power-off state.
The intelligent ammeter with the super capacitor is additionally arranged at all loads directly connected to the low-voltage side outlet end of the transformer, and the node is located at the head end of the transformer area, so that whether the node is in power failure in the transformer area level can be judged if the node gives an alarm, and the distribution of the active power failure alarm device under the transformer area level is realized.
S404, determining all feeder nodes connected by the characteristic users according to the topological connection relation of the feeders, and polling the alarm electric meters of all the feeder nodes to obtain the electricity consumption data of all the feeder nodes, so as to determine the electricity-out state of all the feeder nodes.
In the power failure range analysis process, the consideration of a shunt layer can be considered downwards, the front-back connection relation of each branch in each feeder line is fully considered in the shunt layer, and the intelligent electric meter with the super capacitor is additionally arranged on each feeder line head end node and the shunt head end node, so that the active power failure alarm device distribution of the shunt layer is realized.
S405, sequentially polling the rest other users in the cluster group where the feature user is located according to the power failure correlation coefficient of the feature user to obtain power consumption data of each rest other user, so as to determine the power failure stop state of each user in the cluster group corresponding to the same moment, and executing the step S406;
after the transformer area and the feeder line connection node are considered, the intelligent electric meters with the super capacitors are additionally arranged on other feature users in the cluster group, so that active warning of the power failure state under the local user level is realized. And polling the other users in the same cluster, wherein the polling sequence depends on the power failure correlation coefficient between the user and the characteristic user, and the users with high correlation coefficient value are polled preferentially to determine the power failure state of each user in the cluster at the moment.
S406, determining a power failure range according to the power failure state of the distribution area corresponding to the characteristic user, the power failure state of each feeder line node and the power failure state of each user in the cluster group.
It should be noted that, in this embodiment, in addition to eliminating the possibility of planned power outage, a polling plan under different alarm conditions is set based on a polling technique, and the power outage state of the user in the alarm cluster is analyzed by using a multi-level polling sampling scheme and combining the power outage states of the feature users, so that the power outage range is narrowed step by step and finally determined, and the accuracy of positioning the power outage range is improved.
The above is a detailed description of an embodiment of the method for analyzing the power outage range based on the power outage correlation clustering provided by the present invention, and the following is a detailed description of an embodiment of the system for analyzing the power outage range based on the power outage correlation clustering provided by the present invention.
For convenience of understanding, please refer to fig. 2, the present invention provides a system for analyzing a power outage range based on a power outage correlation cluster, comprising:
the power consumption obtaining module 100 is configured to obtain power consumption data of each user in the distribution room in a preset time period based on the user side meter;
the clustering module 200 is used for clustering and analyzing the power consumption data of each user by adopting a K-means clustering algorithm, so that the power consumption data with the highest similarity is classified into the same cluster group to obtain a plurality of cluster groups;
the correlation calculation module 300 is configured to calculate a power outage correlation coefficient between every two pieces of power consumption data in the same cluster group, generate a correlation coefficient matrix according to the power outage correlation coefficient, and select a user with the highest power outage correlation coefficient as a feature user according to the correlation coefficient matrix;
and the power failure analysis module 400 is configured to perform polling on the off-state power of the distribution area where the feature user is located, the feeder layer connected to the feature user, and other users of the cluster group where the feature user is located by using a polling technique based on a power failure active alarm signal sent by the feature user in advance, so as to determine a power failure range.
In one embodiment, the system further comprises:
and the freezing module is used for freezing the electricity consumption data of each user in the distribution room according to the preset freezing time and sending the frozen electricity consumption data to the metering center.
In a specific embodiment, the clustering module specifically includes:
an initial module, configured to form a data set X ═ X from the electricity consumption data1，x2,…,xn) In the data set X ═ X (X)1，x2,…,xn) Selecting K samples as initial clustering center x ═ (x) at random1，x2,…,xk) Defining a set of cluster groups corresponding to each of the initial cluster centers as S ═ S1,S2,…,Sk}；
A distance calculation module for calculating the shortest Euclidean distance between each sample in the data set and the initial clustering center by the following formula 1,
in formula 1, D (x) represents the shortest Euclidean distance, xiRepresents the ith sample, μ in the datasetiRepresents a cluster center, wherein the cluster center μiAs indicated by the general representation of the,
a probability calculation module for calculating a selection probability P (x) of each sample being selected as a next cluster center by the following formula 3 according to each sample and a shortest Euclidean distance from an initial cluster center,
the cluster updating module is used for selecting a sample as a next cluster center according to the selection probability of each sample and a roulette method, so that k next cluster centers are selected;
and the clustering module is used for calculating the shortest Euclidean distance between each sample in the data set and k next clustering centers through formula 1, and distributing each sample to the cluster group corresponding to the clustering center with the smallest distance.
In a specific embodiment, the correlation calculation module specifically includes:
the matrix generation module is used for generating an electric quantity data matrix based on two electric quantity data in the same cluster group, and the electric quantity data matrix is marked as M, wherein M is an M multiplied by b dimensional matrix, M is the number of types of electric quantity data, and b is the number of users in the same cluster group;
a correlation coefficient calculation module for calculating the power failure correlation coefficient between two power consumption data in the same cluster group by the correlation coefficient calculation formula of the following formula 4,
in equation 4, ρ represents a power outage correlation coefficient, atA data matrix representing the electricity usage of meter a during a period t,
represents the average value of the electric meter a in the whole time period, T represents the total time period, gtA data matrix representing the electricity consumption of the electricity meter g during the period t,
represents the average value of the electricity meter g in the whole time period, LaaMean square sum, L, of the data matrix representing the power consumption of meter a during a period tggRepresents the sum of the squares of the mean differences of the electricity usage data matrix of the electricity meter g over the period t, cov (a, g) represents the covariance of the electricity meter a and the electricity meter g, wherein,
the matrix module is used for generating a corresponding correlation coefficient matrix according to the power failure correlation coefficients of the same cluster group;
and the characteristic user determining module is used for selecting the user with the highest power failure correlation coefficient as the characteristic user according to the correlation coefficient matrix.
In one embodiment, the power outage analysis module specifically includes:
the alarm module is used for sending a power failure active alarm signal to the concentrator when a power failure signal is detected by an alarm ammeter pre-installed by a feature user, and reporting the power failure active alarm signal to the master station through the concentrator;
the judging module is used for acquiring the number of power failure active warning signals received by the master station from the same station area within the same time margin;
the transformer area polling module is used for polling alarm electric meters at all loads directly connected to the outlet end of the low-voltage side of the transformer area where the characteristic user is located so as to obtain power consumption data of all the loads, and therefore the power-off state of the whole transformer area is determined, wherein the power-off state comprises a normal power-on state and a power-off state;
the feeder polling module is used for determining all feeder nodes connected by the characteristic users according to the feeder topological connection relation, and polling the alarm electric meters of all the feeder nodes to obtain the electricity consumption data of all the feeder nodes so as to determine the electricity stop state of each feeder node;
the cluster group polling module is used for sequentially polling the rest other users in the cluster group where the characteristic user is located according to the power failure correlation coefficient of the characteristic user so as to obtain the power consumption data of each rest other user, and thus the power failure state of each user in the cluster group corresponding to the same moment is determined;
and the power failure range determining module is used for determining the power failure range according to the power failure state of the distribution area corresponding to the characteristic user, the power failure state of each feeder node and the power failure state of each user in the cluster group.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. The power outage range analysis method based on the power outage correlation clustering is characterized by comprising the following steps:
s1, acquiring power consumption data of each user in the distribution room in a preset time period based on the user side meter;
s2, performing clustering analysis on the power consumption data of each user by adopting a K-means clustering algorithm, so as to classify the power consumption data with the highest similarity into the same cluster group and obtain a plurality of cluster groups;
s3, calculating a power failure correlation coefficient between every two pieces of power consumption data in the same cluster group, generating a correlation coefficient matrix according to the power failure correlation coefficient, and selecting a user with the highest power failure correlation coefficient as a characteristic user according to the correlation coefficient matrix;
and S4, based on the power failure active alarm signal sent by the feature user in advance, polling the power failure state of the station area where the feature user is located, the feeder layer connected with the feature user and other users of the cluster group where the feature user is located by adopting a polling technology, thereby determining the power failure range.
2. The method of claim 1, further comprising:
and freezing the electricity consumption data of each user in the distribution room according to the preset freezing time, and sending the frozen electricity consumption data to a metering center.
3. The method for analyzing power outage range based on power outage correlation clustering according to claim 1, wherein step S2 specifically includes:
s201, forming a data set X (X) by using electricity consumption data1，x2,…,xn) In the data set X ═ X (X)1，x2,…,xn) Selecting K samples as initial clustering center x ═ (x) at random1，x2,…,xk) Defining a set of cluster groups corresponding to each of the initial cluster centers as S ═ S1,S2,…,Sk}；
S202, calculating the shortest Euclidean distance between each sample in the data set and the initial clustering center according to the following formula 1,
in formula 1, D (x) represents the shortest Euclidean distance, xiRepresents the ith sample, μ in the datasetiRepresents a cluster center, wherein the cluster center μiAs indicated by the general representation of the,
s203, calculating the selection probability P (x) of each sample being selected as the next cluster center according to the shortest Euclidean distance between each sample and the initial cluster center by the following formula 3,
s204, selecting a sample as a next clustering center according to the selection probability of each sample by a wheel disc method, and repeating the step until k next clustering centers are selected;
and S205, calculating the shortest Euclidean distance between each sample in the data set and k next clustering centers by using the formula 1, and distributing each sample to the cluster group corresponding to the clustering center with the smallest distance.
4. The method for analyzing power outage range based on power outage correlation clustering according to claim 1, wherein step S3 specifically includes:
s301, generating an electric quantity data matrix based on two electric quantity data in the same cluster group, and recording the electric quantity data matrix as M, wherein M is an M multiplied by b dimensional matrix, M is the number of types of electric quantity data, and b is the number of users in the same cluster group;
s302, calculating the power failure correlation coefficient between every two pieces of power consumption data in the same cluster group by using the correlation coefficient calculation formula of the following formula 4,
in equation 4, ρ represents a power outage correlation coefficient, atA data matrix representing the electricity usage of meter a during a period t,
represents the average value of the electric meter a in the whole time period, T represents the total time period, gtA data matrix representing the electricity consumption of the electricity meter g during the period t,
represents the average value of the electricity meter g in the whole time period, LaaMean square sum, L, of the data matrix representing the power consumption of meter a during a period tggRepresents the sum of the squares of the mean differences of the electricity usage data matrix of the electricity meter g over the period t, cov (a, g) represents the covariance of the electricity meter a and the electricity meter g, wherein,
s303, generating a corresponding correlation coefficient matrix according to the power failure correlation coefficients of the same cluster group;
and S304, selecting the user with the highest power failure correlation coefficient as a characteristic user according to the correlation coefficient matrix.
5. The method for analyzing power outage range based on power outage correlation clustering according to claim 1, wherein step S4 specifically includes:
s401, when a power failure signal is detected through an alarm ammeter pre-installed by the feature user, sending a power failure active alarm signal to the concentrator to which the feature user belongs, and reporting to a master station through the concentrator;
s402, acquiring the number of power failure active alarm signals received by the master station from the same station area within the same time margin, and executing the step S405 if the number of the power failure active alarm signals is equal to 1; if the number of the power failure active alarm signals is more than 1, executing the steps S403-S405;
s403, polling alarm electric meters at all loads directly accessed by the outlet end of the low-voltage side of the transformer area where the characteristic user is located to obtain power consumption data of all the loads, and determining the power-off state of the whole area, wherein the power-off state comprises a normal power consumption state and a power failure state;
s404, determining all feeder nodes connected with the characteristic user according to the topological connection relation of the feeders, and polling the alarm electric meters of all the feeder nodes to obtain the electricity consumption data of all the feeder nodes so as to determine the electricity stop state of each feeder node;
s405, polling the rest other users in the cluster group where the feature user is located according to the power failure correlation coefficient of the feature user in sequence to obtain power consumption data of each rest other user, so as to determine the power failure stop state of each user in the cluster group corresponding to the same moment, and executing the step S406;
s406, determining a power failure range according to the power failure state of the distribution area corresponding to the characteristic user, the power failure state of each feeder line node and the power failure state of each user in the cluster group.
6. Power failure range analysis system based on power failure correlation clustering is characterized by comprising:
the power consumption acquisition module is used for acquiring power consumption data of each user in the distribution area in a preset time period based on the user side meter;
the clustering module is used for clustering and analyzing the power consumption data of each user by adopting a K-means clustering algorithm, so that the power consumption data with the highest similarity is classified into the same cluster group to obtain a plurality of cluster groups;
the correlation calculation module is used for calculating a power failure correlation coefficient between every two pieces of power consumption data in the same cluster group, generating a correlation coefficient matrix according to the power failure correlation coefficient, and selecting a user with the highest power failure correlation coefficient as a characteristic user according to the correlation coefficient matrix;
and the power failure analysis module is used for polling the power failure state of the distribution area where the characteristic user is located, the feeder layer connected with the characteristic user and other users of the cluster group where the characteristic user is located by adopting a polling technology based on the power failure active alarm signal sent by the characteristic user in advance, so that the power failure range is determined.
7. The system of claim 6, further comprising:
and the freezing module is used for freezing the electricity consumption data of each user in the distribution room according to the preset freezing time and sending the frozen electricity consumption data to the metering center.
8. The system according to claim 6, wherein the clustering module specifically comprises:
an initial module, configured to form a data set X ═ X from the electricity consumption data1，x2,…,xn) In the data set X ═ X (X)1，x2,…,xn) Selecting K samples as initial clustering center x = (x) at random1，x2,…,xk) Defining a set of cluster groups corresponding to each of the initial cluster centers as S ═ S1,S2,…,Sk}；
A distance calculation module for calculating the shortest Euclidean distance between each sample in the data set and the initial clustering center by the following formula 1,
in formula 1, D (x) represents the shortest Euclidean distance, xiRepresents the ith sample, μ in the datasetiRepresents a cluster center, wherein the cluster center μiAs indicated by the general representation of the,
a probability calculation module for calculating a selection probability P (x) of each sample being selected as a next cluster center by the following formula 3 according to each sample and a shortest Euclidean distance from an initial cluster center,
the cluster updating module is used for selecting a sample as a next cluster center according to the selection probability of each sample and a roulette method, so that k next cluster centers are selected;
and the clustering module is used for calculating the shortest Euclidean distance between each sample in the data set and k next clustering centers through formula 1, and distributing each sample to the cluster group corresponding to the clustering center with the smallest distance.
9. The system according to claim 6, wherein the correlation calculation module specifically comprises:
the matrix generation module is used for generating an electric quantity data matrix based on two electric quantity data in the same cluster group, and the electric quantity data matrix is marked as M, wherein M is an M multiplied by b dimensional matrix, M is the number of types of electric quantity data, and b is the number of users in the same cluster group;
a correlation coefficient calculation module for calculating the power failure correlation coefficient between two power consumption data in the same cluster group by the correlation coefficient calculation formula of the following formula 4,
in equation 4, ρ represents a power outage correlation coefficient, atA data matrix representing the electricity usage of meter a during a period t,
represents the average value of the electric meter a in the whole time period, T represents the total time period, gtA data matrix representing the electricity consumption of the electricity meter g during the period t,
represents the average value of the electricity meter g in the whole time period, LaaMean square sum, L, of the data matrix representing the power consumption of meter a during a period tggRepresents the sum of the squares of the mean differences of the electricity usage data matrix of the electricity meter g over the period t, cov (a, g) represents the covariance of the electricity meter a and the electricity meter g, wherein,
the matrix module is used for generating a corresponding correlation coefficient matrix according to the power failure correlation coefficients of the same cluster group;
and the characteristic user determining module is used for selecting the user with the highest power failure correlation coefficient as the characteristic user according to the correlation coefficient matrix.
10. The system according to claim 6, wherein the power outage analysis module specifically comprises:
the alarm module is used for sending a power failure active alarm signal to the concentrator when an alarm ammeter pre-installed by the feature user detects a power failure signal, and reporting the power failure active alarm signal to the main station through the concentrator;
the judging module is used for acquiring the number of power failure active warning signals received by the master station from the same station area within the same time margin;
the transformer area polling module is used for polling alarm electric meters at all loads directly accessed by the outlet end of the low-voltage side of the transformer area where the characteristic user is located so as to obtain power consumption data of all the loads, and accordingly determining the power-off state of the whole transformer area, wherein the power-off state comprises a normal power-on state and a power-off state;
the feeder polling module is used for determining all feeder nodes connected with the characteristic users according to the feeder topological connection relation, and polling the alarm electric meters of all the feeder nodes to obtain the electricity consumption data of all the feeder nodes so as to determine the electricity-out state of each feeder node;
the cluster group polling module is used for sequentially polling the rest other users in the cluster group where the characteristic user is located according to the power failure correlation coefficient of the characteristic user so as to obtain the power consumption data of each rest other user, and therefore the power failure state of each user in the cluster group corresponding to the same moment is determined;
and the power failure range determining module is used for determining the power failure range according to the power failure state of the distribution area corresponding to the characteristic user, the power failure state of each feeder node and the power failure state of each user in the cluster group.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111619974.4A CN114298861A (en) | 2021-12-27 | 2021-12-27 | Power failure range analysis method and system based on power failure correlation clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111619974.4A CN114298861A (en) | 2021-12-27 | 2021-12-27 | Power failure range analysis method and system based on power failure correlation clustering |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114298861A true CN114298861A (en) | 2022-04-08 |
Family
ID=80968878
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111619974.4A Pending CN114298861A (en) | 2021-12-27 | 2021-12-27 | Power failure range analysis method and system based on power failure correlation clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114298861A (en) |
-
2021
- 2021-12-27 CN CN202111619974.4A patent/CN114298861A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105915398B (en) | Rural power grid fault based rapid detection and power restoration system and concentrator detection method thereof | |
| CN106199276B (en) | The intelligent diagnosis system and method for exception information in a kind of power information acquisition system | |
| US8635495B2 (en) | Systems and methods for classifying power network failures | |
| CN101111981B (en) | Mapping power system components | |
| Liu et al. | Knowledge-based system for distribution system outage locating using comprehensive information | |
| CN112187518B (en) | Intelligent fusion terminal area topology identification method and system | |
| CN109687590B (en) | Grading outage and restoration reporting method and system | |
| CN111160791A (en) | Abnormal user identification method based on GBDT algorithm and factor fusion | |
| CN103199628A (en) | Real-time power equipment monitoring system based on OSGA (Open Grid Service Architecture) grid technology | |
| CN112763846A (en) | Multi-data source information fusion-based intelligent power failure judgment method for distribution line | |
| CN108919044A (en) | A kind of unit style distribution network failure active identification method based on mutual verification scheme | |
| CN107944661B (en) | Integrated regulation and control system for main network and distribution network | |
| CN108696371B (en) | Network fault determination method and system | |
| CN112583119A (en) | Low-voltage intelligent platform area simulation platform system based on intelligent fusion terminal | |
| CN114298861A (en) | Power failure range analysis method and system based on power failure correlation clustering | |
| CN112036712A (en) | Power distribution terminal state evaluation index weight distribution method | |
| CN107294205B (en) | Substation state monitoring method based on information protection master station data | |
| CN115940134A (en) | Distribution network data analysis processing method based on fault-tolerant data | |
| CN113468472A (en) | Power distribution network fault positioning method and system | |
| CN111382392B (en) | Method for calculating total time of fault handling of power distribution network and method for calculating reliability of total time | |
| CN114062801A (en) | Power failure event quasi-real-time monitoring method and system based on edge calculation | |
| CN112001588A (en) | Accident event online pre-judging method and device based on N-1 state | |
| CN117559447B (en) | Power failure studying and judging data analysis method and system based on power grid model | |
| CN113837899B (en) | Optimal configuration method for feeder remote terminal unit | |
| TWI748672B (en) | Automatic meter reading abnormality analysis system and method thereof |
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 |