WO2020252785A1 - 用电异常识别方法、装置和计算机可读存储介质 - Google Patents

用电异常识别方法、装置和计算机可读存储介质 Download PDF

Info

Publication number
WO2020252785A1
WO2020252785A1 PCT/CN2019/092387 CN2019092387W WO2020252785A1 WO 2020252785 A1 WO2020252785 A1 WO 2020252785A1 CN 2019092387 W CN2019092387 W CN 2019092387W WO 2020252785 A1 WO2020252785 A1 WO 2020252785A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
abnormal
smart meter
transformer
meter reading
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.)
Ceased
Application number
PCT/CN2019/092387
Other languages
English (en)
French (fr)
Inventor
李昂
李晶
刘浩
王丹
华文韬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siemens Ltd China
Siemens AG
Siemens Corp
Original Assignee
Siemens Ltd China
Siemens AG
Siemens Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Siemens Ltd China, Siemens AG, Siemens Corp filed Critical Siemens Ltd China
Priority to CN201980096669.6A priority Critical patent/CN113853528A/zh
Priority to PCT/CN2019/092387 priority patent/WO2020252785A1/zh
Priority to EP19933291.7A priority patent/EP3968040A4/en
Priority to US17/618,933 priority patent/US20220260619A1/en
Publication of WO2020252785A1 publication Critical patent/WO2020252785A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R22/00—Arrangements for measuring time integral of electric power or current, e.g. electricity meters
    • G01R22/06—Arrangements for measuring time integral of electric power or current, e.g. electricity meters by electronic methods
    • G01R22/061—Details of electronic electricity meters
    • G01R22/066—Arrangements for avoiding or indicating fraudulent use
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R11/00—Electromechanical arrangements for measuring time integral of electric power or current, e.g. of consumption
    • G01R11/02—Constructional details
    • G01R11/24—Arrangements for avoiding or indicating fraudulent use
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R22/00—Arrangements for measuring time integral of electric power or current, e.g. electricity meters
    • G01R22/06—Arrangements for measuring time integral of electric power or current, e.g. electricity meters by electronic methods
    • G01R22/10—Arrangements for measuring time integral of electric power or current, e.g. electricity meters by electronic methods using digital techniques
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—ELECTRIC POWER NETWORKS; CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J13/00—Circuit arrangements for providing remote monitoring or remote control of equipment in a power distribution network
    • H02J13/12—Monitoring network conditions, e.g. electrical magnitudes or operational status

Definitions

  • the present invention relates to the field of electric power technology, and in particular to a method, a device and a computer-readable storage medium for identifying electric power abnormalities.
  • the smart meter is the smart terminal of the smart grid.
  • smart meters In addition to the basic power metering function of traditional electric energy meters, in order to adapt to the use of smart grids and new energy, smart meters also have two-way multi-rate metering functions, user-side control functions, and two-way data communication with multiple data transmission modes Functions and other intelligent functions. Smart meters can prevent some low-level electricity theft. At present, there have been acts of stealing electricity against smart meters.
  • transformers are not distinguished, but the same abnormal identification strategy is adopted for each user of all transformers.
  • the number of users served by the transformer and the industry classification may be different. This indiscriminate identification strategy for stealing electricity results in low recognition accuracy and slow recognition speed.
  • the embodiments of the present invention provide a method, a device, and a computer-readable storage medium for identifying abnormal electricity usage.
  • a method for identifying abnormal electricity usage including:
  • Classify transformers where the classified categories include public transformers or dedicated transformers, where the public transformer corresponds to multiple first users, and the dedicated transformer corresponds to one second user;
  • a second abnormal user is identified from the second user.
  • the identification of the first abnormal user from the first user based on the smart meter reading of the public transformer and the smart meter reading of the first user includes:
  • the first abnormal user is identified from the first users.
  • the first abnormal user can be quickly determined based on the correlation detection between the electricity consumption statistics data of the public transformer and the electricity consumption statistics data of the first user.
  • the identifying the first abnormal user from the first user based on the smart meter reading of the first user includes:
  • the first user in each user cluster whose distance from the cluster center is greater than the first threshold is determined as the first abnormal user.
  • the clustering algorithm is used to determine the first abnormal user to ensure the accuracy of identification.
  • the identification of the first abnormal user from the first user based on the smart meter reading of the public transformer and the smart meter reading of the first user includes:
  • the two dimensions of comprehensive correlation detection and clustering algorithm determine the first abnormal user, so as to avoid missing abnormal users.
  • the identifying the second abnormal user from the second user based on the smart meter reading of the second user includes:
  • a second user in each user cluster whose distance from the cluster center is greater than a second threshold is determined as the second abnormal user.
  • the clustering algorithm is used to determine the second abnormal user to ensure the accuracy of identification.
  • the identification of the second abnormal user from the second user based on the smart meter reading of the dedicated transformer includes:
  • the load factor of the special transformer is determined based on the smart meter reading of the special transformer, the second user corresponding to the special transformer whose load factor is greater than the predetermined load factor first threshold value is determined as the second abnormal user, and The second user corresponding to the dedicated transformer whose load factor is less than the predetermined second threshold of the load factor is determined to be the second abnormal user; or
  • the power factor of the dedicated transformer is determined based on the smart meter reading of the dedicated transformer, the second user corresponding to the dedicated transformer whose power factor is greater than the predetermined first threshold value of the power factor is determined as the second abnormal user, and The second user corresponding to the dedicated transformer whose power factor is less than the predetermined second threshold value of the power factor is determined to be the second abnormal user.
  • the second abnormal user can be quickly determined based on the threshold judgment of the power factor or load rate.
  • the method further includes:
  • the load factor of the special transformer is determined based on the smart meter reading of the special transformer, the second user corresponding to the special transformer whose load factor is greater than the predetermined load factor first threshold value is determined as the second abnormal user, and The second user corresponding to the dedicated transformer whose load factor is less than a predetermined second threshold value of the load factor is determined to be the second abnormal user;
  • the power factor of the dedicated transformer is determined based on the smart meter reading of the dedicated transformer, the second user corresponding to the dedicated transformer whose power factor is greater than the predetermined first threshold value of the power factor is determined as the second abnormal user, and The second user corresponding to the dedicated transformer whose power factor is less than the predetermined second threshold value of the power factor is determined to be the second abnormal user.
  • the three dimensions of power factor, load factor, and clustering algorithm are combined to determine the second abnormal user to avoid missing abnormal users.
  • a device for identifying abnormal electricity consumption including:
  • the classification module is configured to classify transformers, the classification categories include public transformers or dedicated transformers, wherein the public transformers correspond to multiple first users, and the dedicated transformers correspond to one second user;
  • the first identification module is used to identify the first abnormal user from the first user based on the smart meter reading of the public transformer and the smart meter reading of the first user; or, based on the smart meter reading of the first user, The first abnormal user is identified among the first users;
  • the second identification module is used to identify the second abnormal user from the second user based on the smart meter reading of the second user or the smart meter reading of the dedicated transformer.
  • the first identification module is configured to determine the first electricity consumption statistical data in the first time interval based on the smart meter readings of the public transformer, and determine the location based on the smart meter readings of the first user.
  • the first abnormal user can be quickly determined based on the correlation detection between the power consumption pattern of the public transformer and the power consumption pattern of the first user.
  • the first identification module is configured to determine the first electricity consumption statistical data in the first time interval based on the smart meter reading of the first user;
  • the first users are clustered to form user clusters, and the first users that cannot be clustered into any user cluster are determined as the first abnormal users; the distance between each user cluster and the cluster center is greater than a first threshold
  • the first user of is determined to be the first abnormal user.
  • a clustering algorithm is used to determine the first abnormal user to ensure the accuracy of identification.
  • the first identification module is configured to determine the first electricity consumption statistical data in the first time interval based on the smart meter readings of the public transformer, and determine the location based on the smart meter readings of the first user.
  • the second electricity consumption statistical data in the first time interval is detected based on the correlation between the first electricity consumption statistical data and the second electricity consumption statistical data, and the first user is identified from the first user.
  • Abnormal users clustering the first users based on the first electricity consumption statistical data to form user clusters, and determining the first users that cannot be clustered into any user cluster as the first abnormal users;
  • the first user in the user cluster whose distance from the cluster center is greater than the first threshold is determined to be the first abnormal user.
  • the two dimensions of comprehensive correlation detection and clustering algorithm determine the first abnormal user, so as to avoid missing abnormal users.
  • the second identification module is configured to determine the third electricity consumption statistical data in the second time interval based on the smart meter reading of the second user, and pair the third electricity consumption statistical data based on the third electricity consumption statistical data.
  • the second users are clustered to form user clusters, the second users that cannot be clustered into any user cluster are determined as the second abnormal users, and the distance between each user cluster and the cluster center is greater than the second threshold The second user is determined to be the second abnormal user.
  • the clustering algorithm is used to determine the second abnormal user to ensure the accuracy of the identification.
  • the second identification module is configured to determine the load factor of the special transformer based on the smart meter reading of the special transformer, and determine the load factor of the special transformer whose load factor is greater than a predetermined first threshold value of the load factor.
  • the corresponding second user is determined to be the second abnormal user, and the second user corresponding to the dedicated transformer whose load rate is less than the predetermined second threshold of the load rate is determined to be the second abnormal user; or, based on The smart meter reading of the dedicated transformer determines the power factor of the dedicated transformer, the second user corresponding to the dedicated transformer whose power factor is greater than the predetermined first threshold value of the power factor is determined as the second abnormal user, and all The second user corresponding to the dedicated transformer whose power factor is less than the predetermined second threshold value of the power factor is determined to be the second abnormal user.
  • the second abnormal user can be quickly determined based on the threshold determination of the power factor or load rate.
  • the second identification module is further configured to determine the load factor of the special transformer based on the smart meter reading of the special transformer, and determine the load factor of the special transformer whose load factor is greater than a predetermined first threshold value of the load factor
  • the corresponding second user is determined to be the second abnormal user, and the second user corresponding to the dedicated transformer whose load rate is less than the predetermined second threshold of the load rate is determined as the second abnormal user;
  • the smart meter reading of the transformer determines the power factor of the dedicated transformer, the second user corresponding to the dedicated transformer whose power factor is greater than the predetermined first threshold value of the power factor is determined as the second abnormal user, and the The second user corresponding to the dedicated transformer whose power factor is less than the predetermined second threshold value of the power factor is determined to be the second abnormal user.
  • the three dimensions of power factor, load factor and clustering algorithm are combined to determine the second abnormal user to avoid missing abnormal users.
  • a device for identifying abnormal electricity consumption including a processor and a memory;
  • An application program that can be executed by the processor is stored in the memory, and is used to make the processor execute the method for identifying an abnormal power usage as described in any one of the above items.
  • the embodiment of the present invention also implements an abnormal power consumption identification device with a processor-memory structure.
  • a computer-readable storage medium in which computer-readable instructions are stored, and the computer-readable instructions are used to execute the method for identifying abnormal electricity usage as described in any of the above items.
  • the embodiment of the present invention also implements a computer-readable storage medium containing computer-readable instructions for executing the method for identifying abnormal electricity usage.
  • Fig. 1 is a flow chart of a method for identifying abnormal electricity usage according to an embodiment of the present invention.
  • Fig. 2 is a schematic diagram of an exemplary topology of a distribution network according to an embodiment of the present invention.
  • Fig. 3 is a schematic diagram of an exemplary processing process of identifying an abnormal electricity usage according to the present invention.
  • Fig. 4 is a structural diagram of a device for identifying an abnormality in electricity usage according to an embodiment of the present invention.
  • Fig. 5 is a structural diagram of a device for recognizing abnormal electricity usage with a processor-memory structure according to an embodiment of the present invention.
  • the embodiment of the present invention classifies transformers as public transformers or special transformers, and adopts respective abnormal identification strategies for public transformers or special transformers. , Thereby improving the recognition accuracy.
  • Fig. 1 is a flow chart of a method for identifying abnormal electricity usage according to an embodiment of the present invention.
  • the method includes:
  • Step 101 Classify the transformers.
  • the classified categories include public transformers or dedicated transformers, where the public transformer corresponds to multiple first users, and the dedicated transformer corresponds to one second user.
  • the transformer is preferably implemented as a distribution transformer in a distribution network.
  • a distribution transformer in a distribution network.
  • it can be implemented as a 6 kilovolt (kV)-20kV medium-voltage distribution transformer or a 220-volt (V) low-voltage distribution transformer.
  • the transformer is classified as a public transformer or a dedicated transformer.
  • the public transformer is used to increase power for multiple users, and each user who obtains power from the public transformer is a first user.
  • the dedicated transformer is used to provide power to one user, who is the second user.
  • the second user is a group user with a large power consumption, such as a shopping mall, a large factory or a group company;
  • the first user is an individual user with a small power consumption, such as residential households, convenience stores, individual vendors, etc. .
  • the number of classified public transformers can be one or more.
  • the number of dedicated transformers can also be one or more.
  • the number of second users is correspondingly multiple.
  • Step 102 Identify the first abnormal user from the first user based on the smart meter reading of the public transformer and the smart meter reading of the first user; or, identify the first user from the first user based on the smart meter reading of the first user An abnormal user.
  • the first abnormal user is identified from the first users to speed up the identification speed.
  • the identification of the first abnormal user from the first user based on the smart meter reading of the public transformer and the smart meter reading of the first user includes: determining the first abnormal user in the first time interval based on the smart meter reading of the public transformer The first electricity consumption statistics data, the second electricity consumption statistics data in the first time interval are determined based on the smart meter reading of the first user; the correlation detection based on the first electricity consumption statistics data and the second electricity consumption statistics data is obtained from The first abnormal user is identified among the first users.
  • the first time interval may be days, weeks, months, quarters, or natural years.
  • the first electricity consumption statistics data is the electricity consumption statistics data of the public transformer in the first time interval, reflecting the electricity consumption patterns of the public transformer in the first time interval;
  • the second electricity consumption statistics data is the first user in the first time interval
  • the electricity consumption statistics data reflects the electricity consumption pattern of the first user in the first time interval.
  • the first electricity consumption statistics data and the second electricity consumption statistics data can be specifically implemented as: a day's electricity curve; a week's electricity curve; a natural year's average daily electricity curve, and so on.
  • the first abnormal user can be determined. For example: superimpose the power curve of all first users in a week into a comprehensive power curve; compare the power curve of the public transformer during the week with the comprehensive power curve (for example, use autocorrelation method or modified covariance method to compare Shape or amplitude difference) to determine a non-technical electrical loss curve (for example, a line loss curve). Then, the non-technical power loss curve and the power curve of each first user are respectively correlated (for example, using autocorrelation method or modified covariance method to compare power spectrum differences) to determine the first abnormal user.
  • the first abnormal user is quickly determined based on the correlation detection between the electricity consumption statistics of the public transformer and the electricity consumption statistics of the first user.
  • identifying the first abnormal user from the first user based on the smart meter reading of the first user includes: determining the first electricity consumption statistical data in the first time interval based on the smart meter reading of the first user ; Cluster the first users to form a user cluster based on the first electricity consumption statistical data in the first time interval, and determine the first user that cannot be clustered into any user cluster as the first abnormal user; each user The first user in the cluster whose distance from the cluster center is greater than the first threshold is determined to be the first abnormal user.
  • the process of clustering the first users based on the first electricity consumption statistics data to form a user cluster specifically includes: taking the first electricity consumption statistics data of the first users as the clustering object, assuming that there are M first users, then there are M corresponding first electricity consumption statistics data, correspondingly, the clustering objects are also M; randomly select K (K is less than M) clustering objects as the initial clustering centers, and then calculate each clustering object and each The distance between the cluster centers, each cluster object is assigned to the nearest cluster center, and the cluster center and the cluster object assigned to the cluster center represent a user cluster. Once all cluster objects are allocated, the cluster center of each cluster is recalculated based on the existing cluster objects in the user cluster. This process will continue to repeat until a certain termination condition is met.
  • the termination condition can be any of the following: (1) No (or minimum number) of cluster objects are reassigned to different user clusters; (2) No (or minimum number) of cluster centers change again; (3) Error square And local minimum.
  • k-means clustering algorithm, K-MEDOIDS algorithm or CLARANS algorithm can be used to perform the above clustering process.
  • a clustering algorithm can also be used to determine the first abnormal user to ensure the accuracy of identification.
  • the identification of the first abnormal user from the first user based on the smart meter reading of the public transformer and the smart meter reading of the first user includes: determining the first abnormal user in the first time interval based on the smart meter reading of the public transformer The first electricity consumption statistics data, the second electricity consumption statistics data in the first time interval are determined based on the smart meter reading of the first user, and the correlation detection is based on the first electricity consumption statistics data and the second electricity consumption statistics data, from Identify the first abnormal user among the first users; cluster the first users based on the first electricity consumption statistics data to form a user cluster, determine the first user that cannot be clustered into any user cluster as the first abnormal user, and The first user in each user cluster whose distance from the cluster center is greater than the first threshold is determined as the first abnormal user.
  • correlation detection and clustering algorithm can also be integrated to determine the first abnormal user to avoid missing abnormal users.
  • Step 103 Identify a second abnormal user from the second user based on the smart meter reading of the second user or the smart meter reading of the dedicated transformer.
  • the second abnormal user is identified from the second users based on the abnormal identification strategy applicable to the dedicated transformer.
  • identifying the second abnormal user from the second user based on the smart meter reading of the second user includes: determining the third electricity consumption statistical data in the second time interval based on the smart meter reading of the second user; Cluster the second users based on the third electricity consumption statistics data to form user clusters, and determine the second users who cannot be clustered into any user cluster as the second abnormal users; each user cluster is more than the cluster center distance The second user with the second threshold is determined to be the second abnormal user.
  • the second time interval may be days, weeks, months, quarters, or natural years.
  • the third electricity consumption statistical data is the electricity consumption statistical data of the second user in the second time interval, and reflects the electricity consumption pattern of the second user in the second time interval.
  • the third electricity consumption statistics can be specifically implemented as: a day's electricity curve; a week's electricity curve; a natural year's average daily electricity curve, and so on.
  • the process of clustering the second users based on the third electricity consumption statistics data to form a user cluster specifically includes: taking the third electricity consumption statistics data of the second users as a clustering object, and assuming there are J second users, then There are corresponding J third electricity consumption statistics data, and correspondingly, there are J cluster objects; randomly select T (T less than J) cluster objects as the initial cluster centers, and then calculate each cluster object and For the distance between each cluster center, each cluster object is assigned to the nearest cluster center.
  • the cluster center and the cluster object assigned to the cluster center represent a user cluster. Once all cluster objects are allocated, the cluster center of each user cluster is recalculated based on the existing cluster objects in the cluster. This process will continue to repeat until a certain termination condition is met.
  • the termination condition can be any of the following: (1) No (or minimum number) of cluster objects are reassigned to different user clusters; (2) No (or minimum number) of cluster centers change again; (3) Error square And local minimum.
  • k-means clustering algorithm, K-MEDOIDS algorithm or CLARANS algorithm can be used to perform the above clustering process.
  • the clustering algorithm can be directly used to determine the second abnormal user to ensure the accuracy of identification.
  • identifying the second abnormal user from the second user based on the smart meter reading of the dedicated transformer includes: determining the load rate of the dedicated transformer based on the smart meter reading of the dedicated transformer, and determining the load rate to be greater than the predetermined load rate.
  • the second user corresponding to the dedicated transformer with a threshold value is determined to be the second abnormal user, and the second user corresponding to the dedicated transformer whose load factor is less than the predetermined second threshold of the load rate is determined as the second abnormal user; or , Determine the power factor of the special transformer based on the smart meter reading of the special transformer, determine the second user corresponding to the special transformer whose power factor is greater than the predetermined first threshold value of the power factor as the second abnormal user, and set the power factor less than the predetermined value
  • the second user corresponding to the dedicated transformer with the second threshold value of the power factor is determined to be the second abnormal user.
  • the second abnormal user can be quickly determined based on the threshold value determination of the power factor or load rate.
  • identifying the second abnormal user from the second user based on the smart meter reading of the dedicated transformer includes: determining the third electricity consumption statistical data in the second time interval based on the smart meter reading of the second user; Cluster the second users based on the third electricity consumption statistics data to form user clusters, and determine the second users who cannot be clustered into any user cluster as the second abnormal users; each user cluster is more than the cluster center distance
  • the second user with the second threshold value is determined to be the second abnormal user; based on the smart meter reading of the dedicated transformer, the power consumption pattern of the second user is determined based on the smart meter reading of the second user, based on the power consumption pattern of the second user
  • the second users are clustered to form user clusters, the second users who cannot be clustered into any user cluster are determined as the second abnormal users, and the second users with abnormal electricity consumption patterns in each user cluster are determined as the second Abnormal user: Determine the load factor of the special transformer based on the smart meter reading of the special transformer, and determine the second user corresponding to the
  • the three dimensions of power factor, load factor, and clustering algorithm can also be integrated to determine the second abnormal user to avoid missing abnormal users.
  • the first abnormal user determined in step 102 and the second abnormal user determined in step 103 can be summarized, and finally a list of all abnormal users is presented to the inspector through a display interface or the like.
  • Fig. 2 is a schematic diagram of an exemplary topology of a distribution network according to an embodiment of the present invention.
  • the distribution cable 200 is connected with a public transformer 201, a first dedicated transformer 202, a second dedicated transformer 203 and a third dedicated transformer 215.
  • the distribution cable 200 provides power for the public transformer 201, the first dedicated transformer 202, the second dedicated transformer 203, and the third dedicated transformer 215.
  • the public transformer 201 serves three users, namely the first household user 207, the second household user 208 and the convenience store user 209.
  • a master meter 204 is connected to the common transformer 201.
  • the first household user 207 has its own smart meter 212
  • the second household user 208 has its own smart meter 213, and the convenience store user 209 has its own smart meter 214.
  • the reading of the smart meter 212 is used to indicate the power consumption of the first household user 207;
  • the reading of the smart meter 213 is used to indicate the power consumption of the second household user 208;
  • the reading of the smart meter 214 is used to indicate the convenience store user 209 electricity consumption.
  • the reading of the smart meter 214 is used to indicate the power consumption of the public transformer 201.
  • the dedicated transformer 202 serves one user, which is the first group user 205.
  • the first group user 205 has its own smart meter 210.
  • the reading of the smart meter 210 is used to indicate the power consumption of the first group user 205.
  • the dedicated transformer 203 serves one user and is the second group user 206.
  • the second group user 206 has its own smart meter 211.
  • the reading of the smart meter 211 is used to indicate the power consumption of the user 206 of the second group.
  • the dedicated transformer 215 serves one user and is the third group user 216.
  • the third group user 216 has its own smart meter 217. The reading of the smart meter 217 is used to indicate the power consumption of the third group user 216.
  • the anomaly identification strategy applicable to the public transformer 201 includes at least one of the following strategies:
  • Strategy 1 First, determine the statistical data of the public transformer 201 (for example, a day's power curve) based on the readings of the general table 204 of the public transformer. Then, the statistical data of the first household user 207 is determined based on the reading of the smart meter 212 of the first household user 207 (similarly, the power curve of a day), and the second household user 208 is determined based on the reading of the smart meter 213 of the second household user 208 Based on the reading of the smart meter 214 of the convenience store user 209, the statistical data of the convenience store user 209 (similarly, the one day power curve) is determined.
  • the statistical data of the public transformer 201 for example, a day's power curve
  • the statistical data of the first household user 207, the statistical data of the second household user 208, and the statistical data of the convenience store user 209 are superimposed into comprehensive electricity statistics; the statistical data of the public transformer 201 and the comprehensive electricity statistics are correlated Compare (for example, compare the difference in shape or amplitude) to determine the non-technical electricity loss curve; then compare the statistical data of the first household user 207, the statistical data of the second household user 208 and the convenience store user 209 with the non-technical electricity loss curve.
  • the loss curves are compared for correlation (for example, comparing power spectrum differences) to identify abnormal users.
  • Strategy 2 Determine the statistical data of the first household user 207 based on the reading of the smart meter 212 of the first household user 207 (for example, a curve of the average amount of electricity in a natural year), based on the reading of the smart meter 213 of the second household user 208 Determine the statistical data of the second home user 208 (similarly, for example, a curve of the average daily electricity amount in a natural year), and determine the statistical data of the convenience store user 209 based on the reading of the smart meter 214 of the convenience store user 209 (also, for example, a Average curve of the amount of electricity in a natural year).
  • the clustering result shows that the first family user 207 and the second family user 208 can be clustered into the same user cluster (ie, family user cluster), but the convenience store user 209 cannot cluster into this user cluster.
  • the user 209 is determined to be an abnormal user.
  • users in each user cluster whose distance from the cluster center is greater than a predetermined threshold are determined as abnormal users. For example, in the user cluster where the first household user 207 and the second household user 208 are located, abnormal users with abnormal power consumption patterns are determined.
  • Strategy 3 Execute the above strategy 1 and strategy 2 at the same time. Specifically, it includes: determining the statistical data of the public transformer 201 (for example, the electric quantity curve of a day) based on the reading of the general table 204 of the public transformer. Then, determine the statistical data of the first household user 207 based on the reading of the smart meter 212 of the first household user 207 (similarly, the power curve of a day), and determine the second household user 208 based on the reading of the smart meter 213 of the second household user 208 Based on the reading of the smart meter 214 of the convenience store user 209, the statistical data of the convenience store user 209 (similarly, the one day power curve) is determined.
  • the statistical data of the public transformer 201 for example, the electric quantity curve of a day
  • the statistical data of the first household user 207, the statistical data of the second household user 208, and the statistical data of the convenience store user 209 are superimposed into comprehensive electricity statistics; the statistical data of the public transformer 201 and the comprehensive electricity statistics are correlated Compare (for example, compare the difference in shape or amplitude) to determine the non-technical electricity loss curve; then compare the statistical data of the first household user 207, the second household user 208, and the convenience store user 209 with the non-technical data.
  • Technical electrical loss curves are compared for correlation (for example, comparing power spectrum differences) to identify abnormal users.
  • the statistical data of the first household user 207 is determined based on the reading of the smart meter 212 of the first household user 207 (for example, a curve of the average amount of electricity in a natural year), based on the reading of the smart meter 213 of the second household user 208 Determine the statistical data of the second home user 208 (similarly, for example, a curve of the average daily electricity amount in a natural year), and determine the statistical data of the convenience store user 209 based on the reading of the smart meter 214 of the convenience store user 209 (also, for example, a Average curve of the amount of electricity in a natural year).
  • the clustering result shows that the first family user 207 and the second family user 208 can be clustered into the same user cluster (ie, family user cluster), but the convenience store user 209 cannot cluster into this user cluster.
  • the user 209 is determined to be an abnormal user.
  • users in each user cluster whose distance from the cluster center is greater than a predetermined threshold are determined as abnormal users. For example, in the user cluster where the first household user 207 and the second household user 208 are located, abnormal users with abnormal power consumption patterns are determined.
  • the abnormal identification strategies applicable to the dedicated transformer 202, the dedicated transformer 203, and the dedicated transformer 215 include at least one of the following:
  • Strategy 1 First, determine the statistical data of the first group user 205 based on the readings of the smart meter 210 of the first group user 205 (for example, a day’s power curve), and determine the second group based on the reading of the smart meter 211 of the second group user 206
  • the statistical data of the group user 206 (similarly, the electric power curve for a day) is determined based on the reading of the smart meter 217 of the third group user 216 (similarly, the electric power curve for a day).
  • the first group user 205, the second group user 206, and the third group user 216 are clustered based on the statistical data to form a user cluster, and the second user who cannot be clustered into any user cluster is determined as an abnormal user.
  • the third group user 216 cannot be clustered into the user cluster corresponding to a large factory.
  • the group user 216 is determined to be an abnormal user.
  • the users in the user cluster whose distance from the cluster center is greater than a predetermined threshold are determined as abnormal users. For example, in the user cluster where the first group user 205 and the second group user 206 are located, abnormal users with abnormal power consumption patterns are determined.
  • Strategy 2 Determine the load factor of the first dedicated transformer 202 based on the reading of the smart meter 210 of the first group user 205, and determine the load factor of the second dedicated transformer 203 based on the reading of the smart meter 211 of the second group user 206.
  • the reading of the smart meter 216 of the group user 217 determines the load factor of the third dedicated transformer 215.
  • the load factor of the first dedicated transformer 202, the load factor of the second dedicated transformer 203, and the load factor of the third dedicated transformer 215 are respectively compared with predetermined load factor thresholds to determine abnormal users.
  • the first group user 205 is determined as an abnormal user; when the load factor of the first dedicated transformer 202 is found to be less than the lower limit threshold of the load factor When value is set, the first group user 205 is determined as an abnormal user.
  • Strategy 3 Determine the power factor of the first dedicated transformer 202 based on the reading of the smart meter 210 of the first group user 205, and determine the power factor of the second dedicated transformer 203 based on the reading of the smart meter 211 of the second group user 206.
  • the reading of the smart meter 216 of the group user 217 determines the power factor of the third dedicated transformer 215.
  • the power factor of the first dedicated transformer 202, the power factor of the second dedicated transformer 203, and the power factor of the third dedicated transformer 215 are respectively compared with predetermined power factor thresholds to determine abnormal users.
  • the second group user 206 is determined as an abnormal user; when the power factor of the second dedicated transformer 203 is found to be less than the power factor lower limit threshold , The second group user 206 is determined as an abnormal user.
  • Strategy 4 execute the above strategy 1, strategy 2 and strategy 3 at the same time. Specifically, it includes: determining the statistical data of the first group user 205 based on the reading of the smart meter 210 of the first group user 205 (for example, a day's power curve), and determining the second group user based on the reading of the smart meter 211 of the second group user 206 The statistical data of 206 (similarly, the electric power curve for a day) is determined based on the reading of the smart meter 217 of the third group user 216 (similarly, the electric power curve for one day).
  • the first group user 205, the second group user 206, and the third group user 216 are clustered based on the statistical data to form a user cluster, and the second user that cannot be clustered into any user cluster is determined as an abnormal user.
  • the first group user 205 and the second group user 206 can be clustered into a user cluster corresponding to a large factory, but the third group user 216 cannot be clustered into the user cluster corresponding to a large factory.
  • the group user 216 is determined to be an abnormal user.
  • the users in the user cluster whose distance from the cluster center is greater than a predetermined threshold are determined as abnormal users. For example, in the user cluster where the first group user 205 and the second group user 206 are located, abnormal users with abnormal power consumption patterns are determined.
  • the load factor of the first dedicated transformer 202 is determined based on the reading of the smart meter 210 of the first group user 205
  • the load factor of the second dedicated transformer 203 is determined based on the reading of the smart meter 211 of the second group user 206.
  • the reading of the smart meter 216 of the user 217 determines the load factor of the third dedicated transformer 215.
  • the load factor of the first dedicated transformer 202, the load factor of the second dedicated transformer 203, and the load factor of the third dedicated transformer 215 are respectively compared with a predetermined load factor threshold, and the load factor will be greater than or less than the load factor upper limit threshold.
  • the user corresponding to the special transformer with the lower limit threshold of the rate is determined to be an abnormal user.
  • the power factor of the first dedicated transformer 202 is determined based on the reading of the smart meter 210 of the first group user 205
  • the power factor of the second dedicated transformer 203 is determined based on the reading of the smart meter 211 of the second group user 206.
  • the reading of the smart meter 216 of the user 217 determines the power factor of the third dedicated transformer 215.
  • the power factor of the first dedicated transformer 202, the power factor of the second dedicated transformer 203, and the power factor of the third dedicated transformer 215 are respectively compared with a predetermined power factor threshold, and the power factor is greater than or less than the power factor upper limit threshold.
  • the user corresponding to the special transformer with the lower limit threshold of the factor is determined to be an abnormal user.
  • FIG. 3 is an exemplary processing schematic diagram of the identification processing process of abnormal electricity consumption according to the present invention.
  • a classification operation 302 is performed, so that the transformer 301 is classified as a public transformer 303 or a dedicated transformer 304.
  • the total meter reading 305 of the public transformer and the smart meter reading 306 of the public transformer user are collected respectively.
  • the correlation detection processing 307 is performed on the electricity consumption pattern reflected in the general meter reading 305 of the public transformer and the electricity consumption pattern reflected in the smart meter reading 306 of the public transformer user to determine the abnormal user.
  • the clustering process 308 is performed on the user pattern reflected by the smart meter reading of the public transformer user to determine the abnormal user.
  • the user's electricity consumption statistical data is determined based on the user's smart meter reading, and the user performs clustering processing 309 based on the electricity consumption statistical data. Users in the same user cluster can perform user comparison 310 between similar users or comparison 311 between users and their own history to determine abnormal users. Furthermore, the load factor or power factor 312 of the common transformer 304 can be determined, and the threshold value comparison process 313 is executed based on the load factor or power factor 312 to determine abnormal users.
  • the embodiment of the present invention also proposes a device for identifying abnormal electricity usage.
  • Fig. 4 is a structural diagram of a device for identifying an abnormality in electricity usage according to an embodiment of the present invention.
  • the electrical abnormality identification device 400 includes:
  • the classification module 401 is configured to classify transformers.
  • the classification categories include public transformers or special transformers, where the public transformer corresponds to multiple first users, and the special transformer corresponds to one second user;
  • the first identification module 402 is used to identify the first abnormal user from the first user based on the smart meter reading of the public transformer and the smart meter reading of the first user; or, based on the smart meter reading of the first user, from the first user Identify the first abnormal user among users;
  • the second identification module 403 is configured to identify the second abnormal user from the second user based on the smart meter reading of the second user or the smart meter reading of the dedicated transformer.
  • the first identification module 402 is configured to determine the first electricity consumption statistical data in the first time interval based on the smart meter reading of the public transformer, and determine the first electricity consumption statistical data in the first time interval based on the smart meter reading of the first user.
  • the first identification module 402 is configured to determine the first electricity consumption statistics data in the first time interval based on the smart meter reading of the first user; cluster the first users based on the first electricity consumption statistics data To form user clusters, the first user who cannot be clustered into any user cluster is determined as the first abnormal user; the first user in each user cluster whose distance from the cluster center is greater than the first threshold is determined as the first user Abnormal user.
  • the first identification module 402 is configured to determine the first electricity consumption statistical data in the first time interval based on the smart meter reading of the public transformer, and determine the first electricity consumption statistics based on the smart meter reading of the first user.
  • the second electricity consumption statistics data in the time interval based on the correlation detection between the first electricity consumption statistics data and the second electricity consumption statistics data, identify the first abnormal user from the first users;
  • the first users are clustered to form user clusters, and the first users that cannot be clustered into any user cluster are determined as the first abnormal users; the first users in each user cluster whose distance from the cluster center is greater than the first threshold , Determined as the first abnormal user.
  • the second identification module 403 is configured to determine the third electricity consumption statistical data in the second time interval based on the smart meter reading of the second user, and gather the second user based on the third electricity consumption statistical data.
  • Class to form user clusters the second user that cannot be clustered into any user cluster is determined as the second abnormal user, and the second user in each user cluster whose distance from the cluster center is greater than the second threshold is determined as the second Abnormal user.
  • the second identification module 403 is used to determine the load factor of the special transformer based on the smart meter reading of the special transformer, and determine the second user corresponding to the special transformer whose load factor is greater than the predetermined first threshold of the load factor Determine as the second abnormal user, determine the second user corresponding to the dedicated transformer whose load factor is less than the predetermined second threshold of the load factor as the second abnormal user; or, determine the power of the dedicated transformer based on the smart meter reading of the dedicated transformer
  • the second user corresponding to the dedicated transformer whose power factor is greater than the predetermined first threshold value of the power factor is determined as the second abnormal user, and the dedicated transformer whose power factor is less than the predetermined second threshold value of the power factor is determined to be the second abnormal user.
  • the corresponding second user is determined to be the second abnormal user.
  • the second identification module 403 is also used to determine the load factor of the special transformer based on the smart meter reading of the special transformer, and determine the second load factor corresponding to the special transformer whose load factor is greater than the predetermined first threshold of the load factor.
  • the user is determined to be the second abnormal user, and the second user corresponding to the dedicated transformer whose load rate is less than the predetermined second threshold of the load rate is determined as the second abnormal user; the power factor of the dedicated transformer is determined based on the smart meter reading of the dedicated transformer , The second user corresponding to the dedicated transformer with a power factor greater than the predetermined first threshold of power factor is determined as the second abnormal user, and the second user corresponding to the dedicated transformer with a power factor less than the predetermined second threshold of power factor is determined as the second abnormal user.
  • the second user is determined to be the second abnormal user.
  • the embodiment of the present invention also proposes an abnormal power consumption identification device with a processor and a memory structure.
  • Fig. 5 is a structural diagram of a device for recognizing abnormal electricity usage with a processor-memory structure according to an embodiment of the present invention.
  • the device 500 for identifying abnormal electricity usage includes a processor 501 and a memory 502;
  • An application program that can be executed by the processor 501 is stored in the memory 502 to enable the processor 501 to execute the method for identifying an abnormality in power usage as described above.
  • the memory 502 can be specifically implemented as a variety of storage media such as electrically erasable programmable read-only memory (EEPROM), flash memory (Flash memory), and programmable program read-only memory (PROM).
  • the processor 501 may be implemented to include one or more central processing units or one or more field programmable gate arrays, where the field programmable gate array integrates one or more central processing unit cores.
  • the central processing unit or central processing unit core may be implemented as a CPU or MCU.
  • a hardware module may include specially designed permanent circuits or logic devices (such as dedicated processors, such as FPGAs or ASICs) to complete specific operations.
  • the hardware module may also include a programmable logic device or circuit (for example, including a general-purpose processor or other programmable processors) temporarily configured by software to perform specific operations.
  • a programmable logic device or circuit for example, including a general-purpose processor or other programmable processors temporarily configured by software to perform specific operations.
  • a mechanical method, or a dedicated permanent circuit, or a temporarily configured circuit (such as software configuration) to implement the hardware module it can be determined according to cost and time considerations.
  • the present invention also provides a machine-readable storage medium that stores instructions for making a machine execute the method described herein.
  • a system or device equipped with a storage medium may be provided, and the software program code for realizing the function of any one of the above embodiments is stored on the storage medium, and the computer (or CPU or MPU of the system or device) ) Read and execute the program code stored in the storage medium.
  • an operating system operating on the computer can also be used to complete part or all of the actual operations through instructions based on the program code.
  • Implementations of storage media used to provide program codes include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), Magnetic tape, non-volatile memory card and ROM.
  • the program code can be downloaded from a server computer or cloud via a communication network.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Remote Monitoring And Control Of Power-Distribution Networks (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种用电异常识别方法、装置和计算机可读存储介质。方法包括：将变压器进行分类，所述分类的类别包括公共变压器或专用变压器，其中所述公共变压器对应于多个第一用户，所述专用变压器对应于一个第二用户（101）；基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户；或，基于第一用户的智能表读数，从所述第一用户中识别出第一异常用户（102）；基于第二用户的智能表读数或专用变压器的智能表读数，从所述第二用户中识别出第二异常用户（103）。对于具有多个第一用户的公共变压器和仅具有一个第二用户的专用变压器，分别采用各自的个性化异常识别策略，从而提高识别准确率且提高识别速度。

Description

用电异常识别方法、装置和计算机可读存储介质 技术领域
本发明涉及电力技术领域，特别是涉及一种用电异常识别方法、装置和计算机可读存储介质。
背景技术
随着社会经济的发展和用电量的增大，用电异常问题(比如，窃电)变得越来越突出，不但困扰供电企业的发展，也严重影响了国家的经济建设和社会的稳定。
智能电表是智能电网的智能终端。除了具备传统电能表基本用电量的计量功能以外，为了适应智能电网和新能源的使用，智能电表还具有双向多种费率计量功能、用户端控制功能、多种数据传输模式的双向数据通信功能等智能化的功能。智能电表可以预防一些低级别的窃电行为。目前已经出现了针对智能电表的窃电行为。
在现有技术中，在窃电行为的识别过程中，并不对变压器进行区分，而是针对所有变压器的各个用户采用相同的异常识别策略。然而，变压器所服务的用户数量和行业分类可能不同，这种无区别的窃电识别策略导致识别准确率不高且识别速度较慢。
发明内容
本发明实施方式提出一种用电异常识别方法、装置和计算机可读存储介质。
本发明实施方式的技术方案如下：
一种用电异常识别方法，包括：
将变压器进行分类，所述分类的类别包括公共变压器或专用变压器，其中所述公共变压器对应于多个第一用户，所述专用变压器对应于一个第二用户；
基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户；或，基于第一用户的智能表读数，从所述第一用户中识别出第一异常用户；
基于第二用户的智能表读数或专用变压器的智能表读数，从所述第二用户中识别出第二异常用户。
可见，在本发明实施方式中，对于具有多个第一用户的公共变压器和仅具有一个第二用户的专用变压器，分别采用各自的个性化异常识别策略，从而提高识别准确率且提高识别速度。
在一个实施方式中，所述基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户包括：
基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据；
基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户。
因此，对于公共变压器，考虑到其用户数较多，直接基于公共变压器的用电统计数据与第一用户的用电统计数据的相关性检测，可以快速确定出第一异常用户。
在一个实施方式中，所述基于第一用户的智能表读数，从所述第一用户中识别出第一异常用户包括：
基于所述第一用户的智能表读数确定在第一时间区间内的第一用电统计数据；
基于所述第一用电统计数据对所述第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为所述第一异常用户；
将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为所述第一异常用户。
可见，对于公共变压器，采用聚类算法确定出第一异常用户，以保证识别的准确率。
在一个实施方式中，所述基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户包括：
基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据，基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户；
基于所述第一用电统计数据对所述第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为所述第一异常用户，将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为所述第一异常用户。
因此，对于公共变压器，综合相关性检测和聚类算法这两个维度确定出第一异常用户，避免遗漏异常用户。
在一个实施方式中，所述基于第二用户的智能表读数从所述第二用户中识别出第二异常用户包括：
基于所述第二用户的智能表读数确定所述第二用户的用电模式；
基于所述第二用户的智能表读数确定在第二时间区间内的第三用电统计数据；
基于所述第三用电统计数据对第二用户进行聚类以形成用户簇，将不能聚类到任意用户簇中的第二用户确定为所述第二异常用户；
将每个用户簇中与簇中心距离大于第二门限值的第二用户确定为所述第二异常用户。
因此，对于专用变压器，采用聚类算法确定出第二异常用户，以保证识别的准确率。
在一个实施方式中，所述基于专用变压器的智能表读数，从所述第二用户中识别出第二异常用户包括：
基于专用变压器的智能表读数确定所述专用变压器的负载率，将所述负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户；或
基于专用变压器的智能表读数确定所述专用变压器的功率因数，将所述功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户。
可见，对于专用变压器，考虑到用户数较少，直接基于功率因数或负载率的门限值判定，可以快速确定第二异常用户。
在一个实施方式中，该方法还包括：
基于专用变压器的智能表读数确定所述专用变压器的负载率，将所述负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户；
基于专用变压器的智能表读数确定所述专用变压器的功率因数，将所述功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户。
因此，对于专用变压器，综合功率因数、负载率和聚类算法这三个维度确定出第二异常用户，避免遗漏异常用户。
一种用电异常识别装置，包括：
分类模块，用于将变压器进行分类，所述分类的类别包括公共变压器或专用变压器，其中所述公共变压器对应于多个第一用户，所述专用变压器对应于一个第二用户；
第一识别模块，用于基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户；或，基于第一用户的智能表读数，从所述第一用户中识别出第一异常用户；
第二识别模块，用于基于第二用户的智能表读数或专用变压器的智能表读数，从所述第二用户中识别出第二异常用户。
可见，在本发明实施方式中，对于具有多个第一用户的公共变压器和仅具有一个第二用户的专用变压器，分别采用各自的个性化异常识别策略，从而提高识别准确率且提高识别速度。
在一个实施方式中，所述第一识别模块，用于基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据；基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户。
可见，对于公共变压器，考虑到其用户数较多，基于公共变压器的用电模式与第一用户的用电模式的相关性检测，从而可以快速确定出第一异常用户。
在一个实施方式中，所述第一识别模块，用于基于所述第一用户的智能表读数确定在第一时间区间内的第一用电统计数据；基于所述第一用电统计数据对所述第一用户聚类以形成用户簇，将不能聚类到任意 用户簇中的第一用户确定为所述第一异常用户；将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为所述第一异常用户。
因此，对于公共变压器，采用聚类算法确定出第一异常用户，以保证识别的准确率。
在一个实施方式中，所述第一识别模块，用于基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据，基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户；基于所述第一用电统计数据对所述第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为所述第一异常用户；将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为所述第一异常用户。
因此，对于公共变压器，综合相关性检测和聚类算法这两个维度确定出第一异常用户，避免遗漏异常用户。
在一个实施方式中，所述第二识别模块，用于基于所述第二用户的智能表读数确定在第二时间区间内的第三用电统计数据，基于所述第三用电统计数据对第二用户进行聚类以形成用户簇，将不能聚类到任意用户簇中的第二用户确定为所述第二异常用户，将每个用户簇中与簇中心距离大于第二门限值的第二用户确定为所述第二异常用户。
因此，对于专用变压器，采用聚类算法确定出第二异常用户，以保证识别的准确。
在一个实施方式中，所述第二识别模块，用于基于专用变压器的智能表读数确定所述专用变压器的负载率，将所述负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户；或，基于专用变压器的智能表读数确定所述专用变压器的功率因数，将所述功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户。
可见，对于专用变压器，考虑到用户数较少，基于功率因数或负载率的门限值判定，可以快速确定第二异常用户。
在一个实施方式中，所述第二识别模块，还用于基于专用变压器的智能表读数确定所述专用变压器的负载率，将所述负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户；基于专用变压器的智能表读数确定所述专用变压器的功率因数，将所述功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户。
因此，对于专用变压器，综合功率因数、负载率和聚类算法这三个维度确定出第二异常用户，避免遗 漏异常用户。
一种用电异常识别装置，包括处理器和存储器；
所述存储器中存储有可被所述处理器执行的应用程序，用于使得所述处理器执行如上任一项所述的用电异常识别方法。
可见，本发明实施方式还实现了具有处理器-存储器结构的用电异常识别装置。
一种计算机可读存储介质，其中存储有计算机可读指令，该计算机可读指令用于执行如上任一项所述的用电异常识别方法。
可见，本发明实施方式还实现了包含用于执行用电异常识别方法的计算机可读指令的计算机可读存储介质。
附图说明
图1为本发明实施例的用电异常识别方法的流程图。
图2为本发明实施例的配电网的示范性拓扑示意图。
图3为根据本发明用电异常识别处理过程的示范性处理示意图。
图4为本发明实施例的用电异常识别装置的结构图。
图5为本发明实施例处理器-存储器结构的、用电异常识别装置的结构图。
其中，附图标记如下：
标号 含义
101～103 步骤
200 配电线缆
201 公共变压器
202 第一专用变压器
203 第二专用变压器
204 公共变压器的总表
205 第一集团用户
206 第二集团用户
207 第一家庭用户
208 第二家庭用户
209 便利店用户
210 第一集团用户的智能电表
211 第二集团用户的智能电表
212 第一家庭用户的智能电表
213 第二家庭用户的智能电表
214 便利店用户的智能电表
215 第三专用变压器
216 第三集团用户
217 第三集团用户的智能电表
301 配电网中的变压器
302 分类操作
303 公共变压器
304 专用变压器
305 公共变压器的总表读数
306 公共变压器用户的智能电表读数
307 相关性检测处理
308 聚类处理
309 聚类处理
310 同类之间的用户比较
311 用户与自身历史比较
312 负载率或功率因数
313 门限值比较处理
314 汇总处理
400 用电异常识别装置
401 分类模块
402 第一识别模块
403 第二识别模块
500 用电异常识别装置
501 处理器
502 存储器
具体实施方式
为了使本发明的技术方案及优点更加清楚明白，以下结合附图及实施方式，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施方式仅仅用以阐述性说明本发明，并不用于限定本发明的保护范围。
为了描述上的简洁和直观，下文通过描述若干代表性的实施方式来对本发明的方案进行阐述。实施方 式中大量的细节仅用于帮助理解本发明的方案。但是很明显，本发明的技术方案实现时可以不局限于这些细节。为了避免不必要地模糊了本发明的方案，一些实施方式没有进行细致地描述，而是仅给出了框架。下文中，“包括”是指“包括但不限于”，“根据……”是指“至少根据……，但不限于仅根据……”。由于汉语的语言习惯，下文中没有特别指出一个成分的数量时，意味着该成分可以是一个也可以是多个，或可理解为至少一个。
考虑到现有技术中对所有变压器的各个用户统一采用相同的异常识别策略的缺点，本发明实施方式将变压器分类为公共变压器或专用变压器，并针对公共变压器或专用变压器分别采用各自的异常识别策略，从而提高识别准确率。
图1为本发明实施例的用电异常识别方法的流程图。
如图1所示，该方法包括：
步骤101：将变压器进行分类，分类的类别包括公共变压器或专用变压器，其中公共变压器对应于多个第一用户，专用变压器对应于一个第二用户。
在这里，变压器优选实施为配电网中的配电变压器。比如，可以实施为6千伏(kV)～20kV的中压配电变压器或220伏特(V)的低压配电变压器。
基于变压器所服务的用户的数量，将变压器分类为公共变压器或专用变压器。其中，公共变压器用于为多个用户提高电力，每个从公共变压器获取电力的用户即为一个第一用户。专用变压器用于为一个用户提供电力，该用户即为第二用户。通常而言，第二用户为用电量庞大的集团用户，比如商场、大型工厂或集团公司；第一用户为用电量较少的个体用户，比如居民家庭、便利店、个体商贩，等等。
基于配电网络的具体拓扑结构，分类出的公共变压器的数目可以为一或多个。同样，专用变压器的数目也可以为一或多个。当专用变压器的数目为多个时，第二用户的数目相应为多个。
步骤102：基于公共变压器的智能表读数和第一用户的智能表读数，从第一用户中识别出第一异常用户；或，基于第一用户的智能表读数，从第一用户中识别出第一异常用户。
在这里，考虑到公共变压器的用户数为多个，基于适用于公共变压器的异常识别策略，从第一用户中识别出第一异常用户，以加快识别速度。
在一个实施方式中，基于公共变压器的智能表读数和第一用户的智能表读数，从第一用户中识别出第一异常用户包括：基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于第一用户的智能表读数确定在第一时间区间内的第二用电统计数据；基于第一用电统计数据与第二用电统计数据的相关性检测，从第一用户中识别出第一异常用户。
优选地，第一时间区间可以为天、周、月、季度或自然年等时间。第一用电统计数据为公共变压器在第一时间区间的用电统计数据，反映了公共变压器在第一时间区间内的用电模式；第二用电统计数据为第一用户在第一时间区间的用电统计数据，反映了第一用户在第一时间区间内的用电模式。第一用电统计数 据和第二用电统计数据可以具体实施为：一天的电量曲线；一周的电量曲线；一个自然年的天电量平均值曲线，等等。
通过对第一用电统计数据与第二用电统计数据的相关性检测，可以确定出第一异常用户。比如：将全部第一用户在一周内的电量曲线叠加为综合电量曲线；将公共变压器在该周内的电量曲线与综合电量曲线进行相关性比较(比如，采用自相关法或修正协方差法比较形状或幅度差异)，以确定出非技术性的电损失曲线(举例，线路损失曲线)。然后，将非技术性的电损失曲线与各个第一用户的电量曲线分别进行相关性比较(比如，采用自相关法或修正协方差法比较功率谱差异)，以确定出第一异常用户。
可见，对于公共变压器，考虑到其用户数较多，直接基于公共变压器的用电统计数据与第一用户的用电统计数据的相关性检测，快速确定出第一异常用户。
在一个实施方式中，基于第一用户的智能表读数，从第一用户中识别出第一异常用户包括：基于第一用户的智能表读数确定在第一时间区间内的第一用电统计数据；基于在第一时间区间内的第一用电统计数据对第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为第一异常用户；将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为第一异常用户。
其中，基于第一用电统计数据对第一用户聚类以形成用户簇的过程具体包括：将第一用户的第一用电统计数据作为聚类对象，假定存在M个第一用户，则存在M个相应的第一用电统计数据，相应的，聚类对象也为M个；随机选取K个(K小于M)聚类对象作为初始的聚类中心，然后计算每个聚类对象与各个聚类中心之间的距离，把每个聚类对象分配给距离它最近的聚类中心，聚类中心以及分配给聚类中心的聚类对象就代表一个用户簇。一旦全部聚类对象都被分配，根据用户簇中现有的聚类对象重新计算每个聚类的聚类中心。这个过程将不断重复，直到满足某个终止条件。终止条件可以是以下任何一个：(1)没有(或最小数目)聚类对象被重新分配给不同的用户簇；(2)没有(或最小数目)聚类中心再发生变化；(3)误差平方和局部最小。优选地，可以采用k均值聚类算法(k-means clustering algorithm)、K-MEDOIDS算法或CLARANS算法执行上述聚类过程。
可见，对于公共变压器，还可以采用聚类算法确定出第一异常用户，以保证识别的准确率。
在一个实施方式中，基于公共变压器的智能表读数和第一用户的智能表读数，从第一用户中识别出第一异常用户包括：基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于第一用户的智能表读数确定在第一时间区间内的第二用电统计数据，基于第一用电统计数据与第二用电统计数据的相关性检测，从第一用户中识别出第一异常用户；基于第一用电统计数据对第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为第一异常用户，将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为第一异常用户。
可见，对于公共变压器，还可以综合相关性检测和聚类算法这两个维度确定出第一异常用户，避免遗漏异常用户。
步骤103：基于第二用户的智能表读数或专用变压器的智能表读数，从第二用户中识别出第二异常用户。
在这里，考虑到每个专用变压器的用户数为1个，因此基于适用于专用变压器的异常识别策略，从第二用户中识别出第二异常用户。
在一个实施方式中，基于第二用户的智能表读数从第二用户中识别出第二异常用户包括：基于第二用户的智能表读数确定在第二时间区间内的第三用电统计数据；基于第三用电统计数据对第二用户进行聚类以形成用户簇，将不能聚类到任意用户簇中的第二用户确定为第二异常用户；将每个用户簇中与簇中心距离大于第二门限值的第二用户确定为第二异常用户。
优选地，第二时间区间可以为天、周、月、季度或自然年等时间。第三用电统计数据为第二用户在第二时间区间的用电统计数据，反映了第二用户在第二时间区间内的用电模式。第三用电统计数据可以具体实施为：一天的电量曲线；一周的电量曲线；一个自然年的天电量平均值曲线，等等。
其中，基于第三用电统计数据对第二用户聚类以形成用户簇的过程具体包括：将第二用户的第三用电统计数据作为一个聚类对象，假定存在J个第二用户，则存在相应的J个第三用电统计数据，相应的，聚类对象也为J个；随机选取T个(T小于J)聚类对象作为初始的聚类中心，然后计算每个聚类对象与各个聚类中心之间的距离，把每个聚类对象分配给距离它最近的聚类中心，聚类中心以及分配给聚类中心的聚类对象就代表一个用户簇。一旦全部聚类对象都被分配，根据聚类中现有的聚类对象重新计算每个用户簇的聚类中心。这个过程将不断重复，直到满足某个终止条件。终止条件可以是以下任何一个：(1)没有(或最小数目)聚类对象被重新分配给不同的用户簇；(2)没有(或最小数目)聚类中心再发生变化；(3)误差平方和局部最小。优选地，可以采用k均值聚类算法(k-means clustering algorithm)、K-MEDOIDS算法或CLARANS算法执行上述聚类过程。
因此，对于专用变压器，可以直接采用聚类算法确定出第二异常用户，以保证识别的准确。
在一个实施方式中，基于专用变压器的智能表读数，从第二用户中识别出第二异常用户包括：基于专用变压器的智能表读数确定专用变压器的负载率，将负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为第二异常用户，将负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为第二异常用户；或，基于专用变压器的智能表读数确定专用变压器的功率因数，将功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为第二异常用户，将功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为第二异常用户。
因此，对于专用变压器，考虑到用户数较少，可以基于功率因数或负载率的门限值判定，快速确定第二异常用户。
在一个实施方式中，基于专用变压器的智能表读数，从第二用户中识别出第二异常用户包括：基于第二用户的智能表读数确定在第二时间区间内的第三用电统计数据；基于第三用电统计数据对第二用户进行 聚类以形成用户簇，将不能聚类到任意用户簇中的第二用户确定为第二异常用户；将每个用户簇中与簇中心距离大于第二门限值的第二用户确定为第二异常用户；基于专用变压器的智能表读数基于第二用户的智能表读数确定所述第二用户的用电模式，基于第二用户的用电模式对第二用户进行聚类以形成用户簇，将不能聚类到任意用户簇中的第二用户确定为第二异常用户，将每个用户簇中用电模式异常的第二用户确定为第二异常用户；基于专用变压器的智能表读数确定专用变压器的负载率，将负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为第二异常用户，将负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为第二异常用户；确定专用变压器的功率因数，将功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为第二异常用户，将功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为第二异常用户。
因此，对于专用变压器，还可以综合功率因数、负载率和聚类算法这三个维度确定出第二异常用户，避免遗漏异常用户。
执行完步骤103之后，可以将步骤102确定出的第一异常用户和步骤103确定出第二异常用户进行汇总，最终通过显示界面等方式将全部异常用户名单呈现给检查人员。
图2为本发明实施例的配电网的示范性拓扑示意图。
由图2可见，配电线缆200连接有公共变压器201、第一专用变压器202、第二专用变压器203和第三专用变压器215。配电线缆200为公共变压器201、第一专用变压器202、第二专用变压器203和第三专用变压器215提供电力。
公共变压器201服务三个用户，分别为第一家庭用户207、第二家庭用户208和便利店用户209。公共变压器201连接有总表204。第一家庭用户207具有自身的智能电表212，第二家庭用户208具有自身的智能电表213，便利店用户209具有自身的智能电表214。智能电表212的读数，用于指示第一家庭用户207的用电量；智能电表213的读数，用于指示第二家庭用户208的用电量；智能电表214的读数，用于指示便利店用户209的用电量。智能电表214的读数，用于指示公共变压器201的用电量。
专用变压器202服务一个用户，为第一集团用户205。第一集团用户205具有自身的智能电表210。智能电表210的读数，用于指示第一集团用户205的用电量。
专用变压器203服务一个用户，为第二集团用户206。第二集团用户206具有自身的智能电表211。智能电表211的读数，用于指示第二集团用户206的用电量。
专用变压器215服务一个用户，为第三集团用户216。第三集团用户216具有自身的智能电表217。智能电表217的读数，用于指示第三集团用户216的用电量。
适用于公共变压器201的异常识别策略包括下列策略中的至少一个：
策略1：首先，基于公共变压器的总表204的读数确定公共变压器201的统计数据(比如，一天的电量曲线)。然后，基于第一家庭用户207的智能电表212的读数确定第一家庭用户207的统计数据(同样， 一天的电量曲线)，基于第二家庭用户208的智能电表213的读数确定第二家庭用户208的统计数据(同样，一天的电量曲线)，基于便利店用户209的智能电表214的读数确定便利店用户209的统计数据(同样，一天的电量曲线)。接着，将第一家庭用户207的统计数据、第二家庭用户208的统计数据和便利店用户209的统计数据叠加为综合电量统计数据；将公共变压器201的统计数据与综合电量统计数据进行相关性比较(比如，比较形状或幅度差异)，以确定非技术性的电损失曲线；再分别将第一家庭用户207的统计数据、第二家庭用户208和便利店用户209的统计数据与非技术性的电损失曲线进行相关性比较(比如，比较功率谱差异)，以识别出异常用户。
策略2：基于第一家庭用户207的智能电表212的读数确定第一家庭用户207的统计数据(比如，一个自然年的天电量平均值曲线)，基于第二家庭用户208的智能电表213的读数确定第二家庭用户208的统计数据(同样，比如，一个自然年的天电量平均值曲线)，基于便利店用户209的智能电表214的读数确定便利店用户209的统计数据(同样，比如，一个自然年的天电量平均值曲线)。然后，针对第一家庭用户207的统计数据、第二家庭用户208的统计数据和便利店用户209的统计数据进行聚类以形成用户簇，将不能聚类到任意用户簇中的用户确定为异常用户。举例，聚类结果显示第一家庭用户207和第二家庭用户208可以聚到相同的用户簇(即家庭用户簇)中，而便利店用户209不能聚到该用户簇中，则可以将便利店用户209确定为异常用户。另外，还将每个用户簇中与簇中心距离大于预定门限值的用户确定为异常用户。比如，在第一家庭用户207和第二家庭用户208所在的用户簇中，确定出用电模式异常的异常用户。
策略3：同时执行上述策略1和策略2。具体包括：基于公共变压器的总表204的读数确定公共变压器201的统计数据(比如，一天的电量曲线)。然后，基于第一家庭用户207的智能电表212的读数确定第一家庭用户207的统计数据(同样，一天的电量曲线)，基于第二家庭用户208的智能电表213的读数确定第二家庭用户208的统计数据(同样，一天的电量曲线)，基于便利店用户209的智能电表214的读数确定便利店用户209的统计数据(同样，一天的电量曲线)。接着，将第一家庭用户207的统计数据、第二家庭用户208的统计数据和便利店用户209的统计数据叠加为综合电量统计数据；将公共变压器201的统计数据与综合电量统计数据进行相关性比较(比如，比较形状或幅度差异)，以确定非技术性的电损失曲线；再分别将第一家庭用户207的统计数据、第二家庭用户208和便利店用户209的统计数据的统计数据与非技术性的电损失曲线进行相关性比较(比如，比较功率谱差异)，以识别出异常用户。而且，还基于第一家庭用户207的智能电表212的读数确定第一家庭用户207的统计数据(比如，一个自然年的天电量平均值曲线)，基于第二家庭用户208的智能电表213的读数确定第二家庭用户208的统计数据(同样，比如，一个自然年的天电量平均值曲线)，基于便利店用户209的智能电表214的读数确定便利店用户209的统计数据(同样，比如，一个自然年的天电量平均值曲线)。然后，针对第一家庭用户207的统计数据、第二家庭用户208的统计数据和便利店用户209的统计数据进行聚类以形成用户簇，将不能聚类到任意用户簇中的用户确定为异常用户。举例，聚类结果显示第一家庭用户207和第二家庭用户208可以 聚到相同的用户簇(即家庭用户簇)中，而便利店用户209不能聚到该用户簇中，则可以将便利店用户209确定为异常用户。另外，还将每个用户簇中与簇中心距离大于预定门限值的用户确定为异常用户。比如，在第一家庭用户207和第二家庭用户208所在的用户簇中，确定出用电模式异常的异常用户。
适用于专用变压器202、专用变压器203和专用变压215的异常识别策略包括下列中的至少一个：
策略1：首先，基于第一集团用户205的智能电表210的读数确定第一集团用户205的统计数据(比如，一天的电量曲线)，基于第二集团用户206的智能电表211的读数确定第二集团用户206的统计数据(同样，一天的电量曲线)，基于第三集团用户216的智能电表217的读数确定第三集团用户216的统计数据(同样，一天的电量曲线)。然后，基于统计数据对第一集团用户205、第二集团用户206和第三集团用户216进行聚类以形成用户簇，将不能聚类到任意用户簇中的第二用户确定为异常用户。比如，假定第一集团用户205和第二集团用户206可以聚类到对应于大型工厂的用户簇，而第三集团用户216不能聚类到该对应于大型工厂的用户簇中，可以将第三集团用户216确定为异常用户。而且，将用户簇中与簇中心距离大于预定门限值的用户确定为异常用户。比如，在第一集团用户205和第二集团用户206所在的用户簇中，确定出用电模式异常的异常用户。
策略2：基于第一集团用户205的智能电表210的读数确定第一专用变压器202的负载率，基于第二集团用户206的智能电表211的读数确定第二专用变压器203的负载率，基于第三集团用户217的智能电表216的读数确定第三专用变压器215的负载率。分别将第一专用变压器202的负载率、第二专用变压器203的负载率和第三专用变压器215的负载率与预定的负载率门限值进行比较，以确定出异常用户。比如，当发现第一专用变压器202的负载率大于负载率的上限门限值时，将第一集团用户205确定为异常用户；当发现第一专用变压器202的负载率小于负载率的下限门限值时，将第一集团用户205确定为异常用户。
策略3：基于第一集团用户205的智能电表210的读数确定第一专用变压器202的功率因数，基于第二集团用户206的智能电表211的读数确定第二专用变压器203的功率因数，基于第三集团用户217的智能电表216的读数确定第三专用变压器215的功率因数。分别将第一专用变压器202的功率因数、第二专用变压器203的功率因数和第三专用变压器215的功率因数与预定的功率因数门限值进行比较，以确定出异常用户。比如，当发现第二专用变压器203的功率因数大于功率因数上限门限值时，将第二集团用户206确定为异常用户；当发现第二专用变压器203的功率因数小于功率因数下限门限值时，将第二集团用户206确定为异常用户。
策略4：同时执行上述策略1、策略2和策略3。具体包括：基于第一集团用户205的智能电表210的读数确定第一集团用户205的统计数据(比如，一天的电量曲线)，基于第二集团用户206的智能电表211的读数确定第二集团用户206的统计数据(同样，一天的电量曲线)，基于第三集团用户216的智能电表217的读数确定第三集团用户216的统计数据(同样，一天的电量曲线)。然后，基于统计数据对第一集团用户205、第二集团用户206和第三集团用户216进行聚类以形成用户簇，将不能聚类到任意用户簇中的 第二用户确定为异常用户。比如，假定第一集团用户205和第二集团用户206可以聚类到对应于大型工厂的用户簇，而第三集团用户216不能聚类到该对应于大型工厂的用户簇中，可以将第三集团用户216确定为异常用户。而且，将用户簇中与簇中心距离大于预定门限值的用户确定为异常用户。比如，在第一集团用户205和第二集团用户206所在的用户簇中，确定出用电模式异常的异常用户。
而且，基于第一集团用户205的智能电表210的读数确定第一专用变压器202的负载率，基于第二集团用户206的智能电表211的读数确定第二专用变压器203的负载率，基于第三集团用户217的智能电表216的读数确定第三专用变压器215的负载率。分别将第一专用变压器202的负载率、第二专用变压器203的负载率和第三专用变压器215的负载率与预定的负载率门限值进行比较，将大于负载率上限门限值或小于负载率下限门限值的专用变压器所对应的用户确定为异常用户。
另外，基于第一集团用户205的智能电表210的读数确定第一专用变压器202的功率因数，基于第二集团用户206的智能电表211的读数确定第二专用变压器203的功率因数，基于第三集团用户217的智能电表216的读数确定第三专用变压器215的功率因数。分别将第一专用变压器202的功率因数、第二专用变压器203的功率因数和第三专用变压器215的功率因数与预定的功率因数门限值进行比较，将大于功率因数上限门限值或小于功率因数下限门限值的专用变压器所对应的用户确定为异常用户。
基于上述描述，图3为根据本发明用电异常识别处理过程的示范性处理示意图。
如图3所示，针对配电网中的变压器301，执行分类操作302，从而将变压器301分类为公共变压器303或专用变压器304。
对于公共变压器303，分别采集公共变压器的总表读数305和公共变压器用户的智能电表读数306。针对公共变压器的总表读数305所体现的用电模式和公共变压器用户的智能电表读数306所体现的用电模式执行相关性检测处理307，以确定出异常用户。而且，针对公共变压器用户的智能电表读数所体现的用户模式执行聚类处理308，以确定出异常用户。
对于公共变压器304，基于用户的智能表读数确定用户的用电统计数据，基于用电统计数据用户执行聚类处理309。在相同用户簇内的用户，可以执行同类之间的用户比较310或用户与自身历史的比较311，以确定出异常用户。而且，可以确定出公共变压器304的负载率或功率因数312，基于负载率或功率因数312执行门限值比较处理313以确定出异常用户。
在汇总处理314中，将公共变压器303和公共变压器304的全部异常用户进行汇总，最终通过显示界面等方式呈现给检查人员。
本发明实施方式还提出了一种用电异常识别装置。
图4为本发明实施例的用电异常识别装置的结构图。
如图4所示，电异常识别装置400包括：
分类模块401，用于将变压器进行分类，分类的类别包括公共变压器或专用变压器，其中公共变压器 对应于多个第一用户，所述专用变压器对应于一个第二用户；
第一识别模块402，用于基于公共变压器的智能表读数和第一用户的智能表读数，从第一用户中识别出第一异常用户；或，基于第一用户的智能表读数，从第一用户中识别出第一异常用户；
第二识别模块403，用于基于第二用户的智能表读数或专用变压器的智能表读数，从第二用户中识别出第二异常用户。
在一个实施方式中，第一识别模块402，用于基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据；基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户。
在一个实施方式中，第一识别模块402，用于基于第一用户的智能表读数确定在第一时间区间内的第一用电统计数据；基于第一用电统计数据对第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为第一异常用户；将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为第一异常用户。
在一个实施方式中，第一识别模块402，用于基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在第一时间区间内的第二用电统计数据，基于第一用电统计数据与第二用电统计数据的相关性检测，从第一用户中识别出第一异常用户；基于第一用电统计数据对第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为第一异常用户；将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为第一异常用户。
在一个实施方式中，第二识别模块403，用于基于第二用户的智能表读数确定在第二时间区间内的第三用电统计数据，基于第三用电统计数据对第二用户进行聚类以形成用户簇，将不能聚类到任意用户簇中的第二用户确定为第二异常用户，将每个用户簇中与簇中心距离大于第二门限值的第二用户确定为第二异常用户。
在一个实施方式中，第二识别模块403，用于基于专用变压器的智能表读数确定专用变压器的负载率，将负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为第二异常用户，将负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为第二异常用户；或，基于专用变压器的智能表读数确定专用变压器的功率因数，将功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为第二异常用户。
在一个实施方式中，第二识别模块403，还用于基于专用变压器的智能表读数确定专用变压器的负载率，将负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为第二异常用户，将负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为第二异常用户；基于专用变压器的 智能表读数确定专用变压器的功率因数，将功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为第二异常用户，将功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为第二异常用户。
本发明实施方式还提出了一种具有处理器和存储器结构的用电异常识别装置。
图5为本发明实施例处理器-存储器结构的、用电异常识别装置的结构图。
如图5所示，用电异常识别装置500包括处理器501和存储器502；
存储器502中存储有可被处理器501执行的应用程序，用于使得处理器501执行如上任一项所述的用电异常识别方法。
其中，存储器502具体可以实施为电可擦可编程只读存储器(EEPROM)、快闪存储器(Flash memory)、可编程程序只读存储器(PROM)等多种存储介质。处理器501可以实施为包括一或多个中央处理器或一或多个现场可编程门阵列，其中现场可编程门阵列集成一或多个中央处理器核。具体地，中央处理器或中央处理器核可以实施为CPU或MCU。
需要说明的是，上述各流程和各结构图中不是所有的步骤和模块都是必须的，可以根据实际的需要忽略某些步骤或模块。各步骤的执行顺序不是固定的，可以根据需要进行调整。各模块的划分仅仅是为了便于描述采用的功能上的划分，实际实现时，一个模块可以分由多个模块实现，多个模块的功能也可以由同一个模块实现，这些模块可以位于同一个设备中，也可以位于不同的设备中。
各实施方式中的硬件模块可以以机械方式或电子方式实现。例如，一个硬件模块可以包括专门设计的永久性电路或逻辑器件(如专用处理器，如FPGA或ASIC)用于完成特定的操作。硬件模块也可以包括由软件临时配置的可编程逻辑器件或电路(如包括通用处理器或其它可编程处理器)用于执行特定操作。至于具体采用机械方式，或是采用专用的永久性电路，或是采用临时配置的电路(如由软件进行配置)来实现硬件模块，可以根据成本和时间上的考虑来决定。
本发明还提供了一种机器可读的存储介质，存储用于使一机器执行如本文所述方法的指令。具体地，可以提供配有存储介质的系统或者装置，在该存储介质上存储着实现上述实施例中任一实施方式的功能的软件程序代码，且使该系统或者装置的计算机(或CPU或MPU)读出并执行存储在存储介质中的程序代码。此外，还可以通过基于程序代码的指令使计算机上操作的操作系统等来完成部分或者全部的实际操作。还可以将从存储介质读出的程序代码写到插入计算机内的扩展板中所设置的存储器中或者写到与计算机相连接的扩展单元中设置的存储器中，随后基于程序代码的指令使安装在扩展板或者扩展单元上的CPU等来执行部分和全部实际操作，从而实现上述实施方式中任一实施方式的功能。
用于提供程序代码的存储介质实施方式包括软盘、硬盘、磁光盘、光盘(如CD-ROM、CD-R、CD-RW、DVD-ROM、DVD-RAM、DVD-RW、DVD+RW)、磁带、非易失性存储卡和ROM。可选择地，可以由通信网络从服务器计算机或云上下载程序代码。
上文通过附图和优选实施例对本发明进行了详细展示和说明，然而本发明不限于这些已揭示的实施例，基与上述多个实施例本领域技术人员可以知晓，可以组合上述不同实施例中的代码审核手段得到本发明更多的实施例，这些实施例也在本发明的保护范围之内。

Claims (16)

  1. 一种用电异常识别方法，其特征在于，包括：
    将变压器进行分类，所述分类的类别包括公共变压器或专用变压器，其中所述公共变压器对应于多个第一用户，所述专用变压器对应于一个第二用户(101)；
    基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户；或，基于第一用户的智能表读数，从所述第一用户中识别出第一异常用户(102)；
    基于第二用户的智能表读数或专用变压器的智能表读数，从所述第二用户中识别出第二异常用户(103)。
  2. 根据权利要求1所述的用电异常识别方法，其特征在于，所述基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户(102)包括：
    基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据；
    基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户。
  3. 根据权利要求1所述的用电异常识别方法，其特征在于，所述基于第一用户的智能表读数，从所述第一用户中识别出第一异常用户(102)包括：
    基于所述第一用户的智能表读数确定在第一时间区间内的第一用电统计数据；
    基于所述第一用电统计数据对所述第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为所述第一异常用户；
    将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为所述第一异常用户。
  4. 根据权利要求1所述的用电异常识别方法，其特征在于，所述基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户(102)包括：
    基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据，基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户；
    基于所述第一用电统计数据对所述第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为所述第一异常用户，将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为所述第一异常用户。
  5. 根据权利要求1所述的用电异常识别方法，其特征在于，所述基于第二用户的智能表读数从所述第二用户中识别出第二异常用户(103)包括：
    基于所述第二用户的智能表读数确定在第二时间区间内的第三用电统计数据；
    基于所述第三用电统计数据对第二用户进行聚类以形成用户簇，将不能聚类到任意用户簇中的第二用 户确定为所述第二异常用户；
    将每个用户簇中与簇中心距离大于第二门限值的第二用户确定为所述第二异常用户。
  6. 根据权利要求1所述的用电异常识别方法，其特征在于，所述基于专用变压器的智能表读数，从所述第二用户中识别出第二异常用户(103)包括：
    基于专用变压器的智能表读数确定所述专用变压器的负载率，将所述负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户；或
    基于专用变压器的智能表读数确定所述专用变压器的功率因数，将所述功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户。
  7. 根据权利要求5所述的用电异常识别方法，其特征在于，该方法还包括：
    基于专用变压器的智能表读数确定所述专用变压器的负载率，将所述负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户；
    基于专用变压器的智能表读数确定所述专用变压器的功率因数，将所述功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户。
  8. 一种用电异常识别装置(400)，其特征在于，包括：
    分类模块(401)，用于将变压器进行分类，所述分类的类别包括公共变压器或专用变压器，其中所述公共变压器对应于多个第一用户，所述专用变压器对应于一个第二用户；
    第一识别模块(402)，用于基于公共变压器的智能表读数和第一用户的智能表读数，从所述第一用户中识别出第一异常用户；或，基于第一用户的智能表读数，从所述第一用户中识别出第一异常用户；
    第二识别模块(403)，用于基于第二用户的智能表读数或专用变压器的智能表读数，从所述第二用户中识别出第二异常用户。
  9. 根据权利要求8所述的用电异常识别装置(400)，其特征在于，
    所述第一识别模块(402)，用于基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据；基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户。
  10. 根据权利要求8所述的用电异常识别装置(400)，其特征在于，
    所述第一识别模块(402)，用于基于所述第一用户的智能表读数确定在第一时间区间内的第一用电统计数据；基于所述第一用电统计数据对所述第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第 一用户确定为所述第一异常用户；将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为所述第一异常用户。
  11. 根据权利要求8所述的用电异常识别装置(400)，其特征在于，
    所述第一识别模块(402)，用于基于公共变压器的智能表读数确定在第一时间区间内的第一用电统计数据，基于所述第一用户的智能表读数确定在所述第一时间区间内的第二用电统计数据，基于所述第一用电统计数据与所述第二用电统计数据的相关性检测，从所述第一用户中识别出所述第一异常用户；基于所述第一用电统计数据对所述第一用户聚类以形成用户簇，将不能聚类到任意用户簇中的第一用户确定为所述第一异常用户；将每个用户簇中与簇中心距离大于第一门限值的第一用户，确定为所述第一异常用户。
  12. 根据权利要求8所述的用电异常识别装置(400)，其特征在于，
    所述第二识别模块(403)，用于基于所述第二用户的智能表读数确定在第二时间区间内的第三用电统计数据，基于所述第三用电统计数据对第二用户进行聚类以形成用户簇，将不能聚类到任意用户簇中的第二用户确定为所述第二异常用户，将每个用户簇中与簇中心距离大于第二门限值的第二用户确定为所述第二异常用户。
  13. 根据权利要求8所述的用电异常识别装置(400)，其特征在于，
    所述第二识别模块(403)，用于基于专用变压器的智能表读数确定所述专用变压器的负载率，将所述负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户；或，基于专用变压器的智能表读数确定所述专用变压器的功率因数，将所述功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户。
  14. 根据权利要求12所述的用电异常识别装置(400)，其特征在于，
    所述第二识别模块(403)，还用于基于专用变压器的智能表读数确定所述专用变压器的负载率，将所述负载率大于预定的负载率第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述负载率小于预定的负载率第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户；基于专用变压器的智能表读数确定所述专用变压器的功率因数，将所述功率因数大于预定的功率因数第一门限值的专用变压器所对应的第二用户确定为所述第二异常用户，将所述功率因数小于预定的功率因数第二门限值的专用变压器所对应的第二用户确定为所述第二异常用户。
  15. 一种用电异常识别装置(500)，其特征在于，包括处理器(501)和存储器(502)；
    所述存储器(502)中存储有可被所述处理器(501)执行的应用程序，用于使得所述处理器(501)执行如权利要求1至7中任一项所述的用电异常识别方法。
  16. 一种计算机可读存储介质，其特征在于，其中存储有计算机可读指令，该计算机可读指令用于执 行如权利要求1至7中任一项所述的用电异常识别方法。
PCT/CN2019/092387 2019-06-21 2019-06-21 用电异常识别方法、装置和计算机可读存储介质 Ceased WO2020252785A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
CN201980096669.6A CN113853528A (zh) 2019-06-21 2019-06-21 用电异常识别方法、装置和计算机可读存储介质
PCT/CN2019/092387 WO2020252785A1 (zh) 2019-06-21 2019-06-21 用电异常识别方法、装置和计算机可读存储介质
EP19933291.7A EP3968040A4 (en) 2019-06-21 2019-06-21 METHOD AND DEVICE FOR DETECTING ABNORMAL ELECTRICITY USE AND COMPUTER READABLE STORAGE MEDIA
US17/618,933 US20220260619A1 (en) 2019-06-21 2019-06-21 Abnormal electricity use recognition method and device, and computer-readable medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/092387 WO2020252785A1 (zh) 2019-06-21 2019-06-21 用电异常识别方法、装置和计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2020252785A1 true WO2020252785A1 (zh) 2020-12-24

Family

ID=74040002

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/092387 Ceased WO2020252785A1 (zh) 2019-06-21 2019-06-21 用电异常识别方法、装置和计算机可读存储介质

Country Status (4)

Country Link
US (1) US20220260619A1 (zh)
EP (1) EP3968040A4 (zh)
CN (1) CN113853528A (zh)
WO (1) WO2020252785A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112865078A (zh) * 2021-01-14 2021-05-28 南京软核科技有限公司 一种中压用户接入点与中压专变用户对应关系识别方法
CN113009221A (zh) * 2021-02-25 2021-06-22 深圳市国电科技通信有限公司 用于检测窃电的方法、智能断路器和电表
CN113792264A (zh) * 2021-09-10 2021-12-14 广东电网有限责任公司 一种用电量异常用户识别方法、装置、设备和介质
CN114118302A (zh) * 2022-01-25 2022-03-01 全球能源互联网研究院有限公司 一种电能表串表识别方法、装置及电子设备
CN114721326A (zh) * 2022-05-31 2022-07-08 国网浙江省电力有限公司 基于深度学习算法的营销稽查信息处理方法及装置
CN115293481A (zh) * 2022-05-30 2022-11-04 国网江苏省电力有限公司无锡供电分公司 一种基于集成学习lscp算法的台区户变关系纠查方法及系统
CN115508636A (zh) * 2022-08-24 2022-12-23 广西电网有限责任公司钦州供电局 一种应用于变压器变压的有载调压分析方法
CN116303480B (zh) * 2023-05-22 2023-08-04 新乡市万新电气有限公司 基于云计算的电能表误差校验方法
CN116910596A (zh) * 2023-07-26 2023-10-20 江苏方天电力技术有限公司 一种基于改进dbscan聚类的用户窃电分析方法、设备及存储介质
CN119066580A (zh) * 2024-08-22 2024-12-03 国网湖北省电力有限公司信息通信公司 用电异常用户数据存储方法、装置、电子设备和可读介质

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114487566A (zh) * 2021-12-31 2022-05-13 杭州拓深科技有限公司 基于电流指纹技术的学校违规电器自识别方法及计算设备
CN115469590B (zh) * 2022-10-31 2023-03-24 深圳金正方科技股份有限公司 智能电表接口的低功耗控制方法、装置、设备及存储介质
CN116340765B (zh) * 2023-02-16 2024-02-09 成都昶鑫电子科技有限公司 一种窃电用户预测方法、装置、存储介质及电子设备
CN116613894B (zh) * 2023-07-20 2023-10-10 国网汇通金财（北京）信息科技有限公司 一种用电数据的处理方法及装置
CN116660621B (zh) * 2023-07-27 2023-09-26 江西琰圭技术服务有限公司 一种本地采样分析的防窃电智能管理系统
CN117093883B (zh) * 2023-09-15 2026-04-17 成都秦川物联网科技股份有限公司 智慧燃气异常数据分析方法、物联网系统和装置及介质
CN118244191B (zh) * 2024-05-28 2024-08-02 国网甘肃省电力公司白银供电公司 一种不间断供电的电力采集计量系统
CN118364313B (zh) * 2024-06-18 2024-09-20 广州南洋电缆集团有限公司 一种电力系统的实时信息标定方法
CN118884069A (zh) * 2024-07-02 2024-11-01 广东电网有限责任公司 一种停电异常识别方法及装置
CN120258651B (zh) * 2025-06-09 2025-08-08 施维智能计量系统服务(长沙)有限公司 一种基于ltu和多维数据分析异常用电的方法、介质及终端

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101206795A (zh) * 2007-08-01 2008-06-25 许之端 支持配网台区测控网络的集控集抄装置
CN201936510U (zh) * 2010-11-03 2011-08-17 河北省电力公司职业技术培训中心 一种双向控制的反窃电仿真培训装置
CN204882735U (zh) * 2015-06-17 2015-12-16 姜进 一种适用于检测配变进出线负荷的比对系统
US20160320431A1 (en) * 2015-05-03 2016-11-03 Itron, Inc. Detection of Electrical Theft from a Transformer Secondary
CN107340492A (zh) * 2017-05-31 2017-11-10 国网上海市电力公司 基于大数据挖掘和场景预判的电能计量装置故障分析方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9418045B2 (en) * 2011-12-12 2016-08-16 Mbh Consulting Ltd. Systems, apparatus and methods for quantifying and identifying diversion of electrical energy
JP2014233154A (ja) * 2013-05-29 2014-12-11 株式会社東芝 配電系統監視システム及び配電系統監視装置
MX357831B (es) * 2013-06-13 2018-07-26 Astrolink Int Llc Perdidas no tecnicas en una rejilla electrica publica.
US10379146B2 (en) * 2015-09-23 2019-08-13 International Business Machines Corporation Detecting non-technical losses in electrical networks based on multi-layered statistical techniques from smart meter data
CN105932773B (zh) * 2016-05-06 2019-03-19 深圳市同昌汇能科技发展有限公司 一种智能收费防窃电系统
CN106203832B (zh) * 2016-07-12 2020-03-27 亿米特(上海)数据科技有限公司 智能防窃电分析系统及分析方法
CN107132410B (zh) * 2017-03-13 2018-09-25 广东电网有限责任公司信息中心 10kV线路线损率异常原因检测方法和系统
CN108256559B (zh) * 2017-12-27 2021-05-14 国网河南省电力公司电力科学研究院 一种基于局部离群点因子的低压窃电用户定位方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101206795A (zh) * 2007-08-01 2008-06-25 许之端 支持配网台区测控网络的集控集抄装置
CN201936510U (zh) * 2010-11-03 2011-08-17 河北省电力公司职业技术培训中心 一种双向控制的反窃电仿真培训装置
US20160320431A1 (en) * 2015-05-03 2016-11-03 Itron, Inc. Detection of Electrical Theft from a Transformer Secondary
CN204882735U (zh) * 2015-06-17 2015-12-16 姜进 一种适用于检测配变进出线负荷的比对系统
CN107340492A (zh) * 2017-05-31 2017-11-10 国网上海市电力公司 基于大数据挖掘和场景预判的电能计量装置故障分析方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3968040A4 *

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112865078A (zh) * 2021-01-14 2021-05-28 南京软核科技有限公司 一种中压用户接入点与中压专变用户对应关系识别方法
CN113009221A (zh) * 2021-02-25 2021-06-22 深圳市国电科技通信有限公司 用于检测窃电的方法、智能断路器和电表
CN113792264A (zh) * 2021-09-10 2021-12-14 广东电网有限责任公司 一种用电量异常用户识别方法、装置、设备和介质
CN114118302A (zh) * 2022-01-25 2022-03-01 全球能源互联网研究院有限公司 一种电能表串表识别方法、装置及电子设备
CN115293481A (zh) * 2022-05-30 2022-11-04 国网江苏省电力有限公司无锡供电分公司 一种基于集成学习lscp算法的台区户变关系纠查方法及系统
CN114721326A (zh) * 2022-05-31 2022-07-08 国网浙江省电力有限公司 基于深度学习算法的营销稽查信息处理方法及装置
CN115508636A (zh) * 2022-08-24 2022-12-23 广西电网有限责任公司钦州供电局 一种应用于变压器变压的有载调压分析方法
CN116303480B (zh) * 2023-05-22 2023-08-04 新乡市万新电气有限公司 基于云计算的电能表误差校验方法
CN116910596A (zh) * 2023-07-26 2023-10-20 江苏方天电力技术有限公司 一种基于改进dbscan聚类的用户窃电分析方法、设备及存储介质
CN119066580A (zh) * 2024-08-22 2024-12-03 国网湖北省电力有限公司信息通信公司 用电异常用户数据存储方法、装置、电子设备和可读介质

Also Published As

Publication number Publication date
US20220260619A1 (en) 2022-08-18
CN113853528A (zh) 2021-12-28
EP3968040A1 (en) 2022-03-16
EP3968040A4 (en) 2022-12-21

Similar Documents

Publication Publication Date Title
WO2020252785A1 (zh) 用电异常识别方法、装置和计算机可读存储介质
CN110223196B (zh) 基于典型行业特征库和反窃电样本库的反窃电分析方法
US11667210B2 (en) Detection of electric vehicle charging at premise
US10664931B2 (en) Data analyzing system and method
CN110806518A (zh) 一种台区线损异动分析模块及其操作方法
CN110687361A (zh) 一种用于识别低压配网台区相序的方法及系统
CN107330540B (zh) 一种考虑电压质量的配电网台区缺供电量预测方法
US11573587B2 (en) Apparatus and method for non-invasively analyzing behaviors of multiple power devices in circuit and monitoring power consumed by individual devices
CN114154999B (zh) 一种反窃电方法、装置、终端及存储介质
CN110231503B (zh) 高损台区基于格兰杰因果检验的窃电用户识别定位方法
CN111709668A (zh) 基于数据挖掘技术的电网设备参数风险识别方法及装置
CN113872204A (zh) 一种基于电网图拓扑计算的用电负荷性质确定方法及装置
CN115409381A (zh) 线损原因的确定方法、装置、电子设备及存储介质
CN112132631A (zh) 一种基于电力客户画像的标签用电管理方法
CN112152213B (zh) 基于线损的拓扑识别率判断方法
US10557720B2 (en) Unauthorized electrical grid connection detection and characterization system and method
CN115207909B (zh) 一种台区拓扑识别方法、装置、设备及存储介质
CN113902485B (zh) 一种专变电力用户行业识别方法、装置及设备
CN115422788A (zh) 一种配电网线损分析管理方法、装置、存储介质及系统
CN112633692B (zh) 查窃电阈值的获取方法和装置、查窃电装置和方法
CN114254965A (zh) 乡村振兴电力监测方法、装置、电子设备及存储介质
CN113850296A (zh) 一种基于fcm算法的终端窃电检测方法、系统、设备及可读存储介质
CN117220417B (zh) 消费端电负荷动态监测方法及系统
CN106651630A (zh) 关键用电行业识别方法和系统
CN110175199A (zh) 基于k均值聚类算法的能源企业关键用户识别分析方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19933291

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2019933291

Country of ref document: EP

Effective date: 20211207

NENP Non-entry into the national phase

Ref country code: DE

WWW Wipo information: withdrawn in national office

Ref document number: 2019933291

Country of ref document: EP