A method, device, system and medium for verifying the relationship between household and transformer in low-voltage distribution network

Abstract

The present invention provides a method, device, system, and medium for verifying the relationship between users and transformers in a low-voltage distribution network. The method includes: extracting load events based on the load waveforms of users and transformers; extracting load features based on the extracted load events; characterizing the extracted load features and arranging them in a predetermined order to obtain a load event character string sequence for the user to be identified; calculating the longest common subsequence between the load event character string sequence of the user to be identified and the load event character string sequence of each transformer, calculating the time deviation entropy value corresponding to the user to be identified and each transformer based on the longest common subsequence, and determining the transformer to which the user to be identified belongs based on the time deviation entropy value; and verifying the transformer to which the determined user to be identified belongs with the transformer to which the user to be identified belongs recorded in the system archive. The present invention can solve the problems of the prior art of adding additional measurement equipment and requiring high data accuracy, thereby improving the accuracy of topology recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network, and in particular to a method, device, system and medium for verifying the relationship between household and transformer in a low-voltage distribution network. Background Art

[0002] Customer-to-substation relationships are the most crucial foundational archival information for low-voltage substations. They are widely used in load management, substation coordination, power flow calculations, state estimation, lean line loss management, and fault location. However, the chaotic maintenance of these records has long plagued power grid companies. With the advancement of lean substation management and the transformation to digital operations and maintenance, traditional manual maintenance methods such as "outage verification," "on-site observation," and "topology identification tools" are time-consuming and labor-intensive, and their update speed cannot meet the analytical requirements of applications. For example, daily line loss analysis and diagnosis, optimized allocation of available capacity, and substation source-load coordination require accurate, real-time customer-to-substation relationships. To address this engineering application challenge, automatic identification of customer-to-substation relationships has become a research hotspot. From a technical perspective, existing automatic identification technologies for customer-to-substation relationships fall into two main categories: signal matching, including communication signal matching and injected power signal matching; and data-driven approaches. These leverage massive amounts of measurement data collected by AMI, analyze circuit operating mechanisms, and utilize voltage correlation and energy conservation as key insights, incorporating machine learning methods for modeling.

[0003] Signal matching methods are primarily engineering methods, with limited academic research. While this type of method has high recognition efficiency, it requires equipment installation and high engineering construction costs. Regarding communication signal matching, the HPLC communication signal from the low-voltage substation carrier communication terminal is used to determine household changes and phases through methods such as communication signal tag matching, communication message characteristics, signal transmission time measurement, and power frequency signal zero-crossing detection. This type of method is only applicable to substations with full HPLC communication coverage and is susceptible to electromagnetic coupling and noise interference. Furthermore, it is not widely used in engineering projects because it affects the success rate of AMI system acquisition. Regarding power signal injection research, a signal generator generates a specific interharmonic power signal for topology matching. This method requires extensive engineering implementation, is costly, presents safety risks, and can fail in complex fishbone grid structures.

[0004] For the data-driven method, the AMI system has accumulated a large amount of user-side load measurement data, and applied machine learning, big data analysis and other methods to explore the non-metering value of existing metering data, which is of great value for engineering applications. The academic community has carried out extensive research. One type is based on the spatiotemporal distribution and correlation characteristics of node voltage, and uses correlation analysis such as Pearson correlation coefficient and grey correlation degree to judge the subordination of household transformers; in addition, household transformer relationship identification is essentially to solve the classification problem. Other studies introduce clustering algorithms into it. This type of method mainly conducts research from two aspects: effective voltage feature extraction method and clustering algorithm optimization. However, it has high requirements on voltage data quality. Strict data screening conditions will greatly reduce the computability. In addition, for scenarios such as small power substations, photovoltaic substations and adjacent substations under the same feeder, the algorithm is limited in applicability due to objective factors such as electrical distance and photovoltaic grid-connected voltage. Summary of the Invention

[0005] In view of this, the present invention provides a method, device, system and medium for verifying the relationship between household transformers in a low-voltage distribution network to solve the problems of adding additional measurement equipment and requiring high data accuracy in the prior art, thereby improving the accuracy of topology recognition technology.

[0006] A method for verifying the relationship between household transformers in a low-voltage distribution network comprises the following steps:

[0007] Extracting load events based on the load waveforms of users and transformers, wherein the load events include load event data in which the active power value exceeds a preset threshold;

[0008] Extracting load features according to the extracted load events, wherein the load features include load event occurrence time, load type, load active power value, and corresponding load reactive power value;

[0009] The extracted load features are digitized and arranged in a predetermined order to obtain a load event character string sequence of the user to be identified;

[0010] Calculate the longest common subsequence between the load event string sequence of the user to be identified and the load event string sequence of each transformer, calculate the time deviation entropy value corresponding to the user to be identified and each transformer based on the longest common subsequence, and determine the transformer to which the user to be identified belongs based on the time deviation entropy value;

[0011] The transformer to which the determined user to be identified belongs is compared with the transformer to which the user to be identified belongs recorded in the system archive.

[0012] Furthermore, the extracted load features are characterized and arranged in sequence according to a pre-set order to obtain a load event character string sequence of the user to be identified, including:

[0013] Divide the time of each day into fixed time intervals and name each interval with a number. Convert the load event occurrence time into the number TS of the interval where the load event occurs.

[0014] Identify the load type TYPE of the corresponding load event, convert the step type character into S, the impact type character into P, and other types that are not step type characters or impact type characters into E;

[0015] The load active power value is divided into intervals of 100 watts and is converted into characters and recorded as QS;

[0016] The reactive power value of the load is divided into intervals of 100Var and is processed into characters, which are recorded as Q.

[0017] The four characterized characteristic characters corresponding to a single load event are combined into a character string EVENT, EVENT = {TS, TYPE, PS, QS};

[0018] Arrange the character strings corresponding to each load event at the user transformer end to be identified in the order of the load event occurrence time to obtain the load event character string sequence of the user to be identified; if the load event occurrence time TS is the same, arrange them in descending order of priority of load type S, P, E, with S as the highest priority character; if the arrangement order still cannot be distinguished, the one with the smaller PS is arranged first; if the PS is the same, the one with the smaller QS is arranged first.

[0019] Furthermore, the step of calculating the longest common subsequence between the load event character string sequence of the user to be identified and the load event character string sequence of each transformer, calculating the time deviation entropy value corresponding to the user to be identified and each transformer based on the longest common subsequence, and determining the transformer to which the user to be identified belongs based on the time deviation entropy value includes:

[0020] The load event string sequence of the user to be identified is matched with the load event string sequence of each transformer by the longest common subsequence matching calculation. The longest common subsequence refers to a sequence W that is a subsequence of two or more known sequences and is the longest among all sequences that meet this condition. Then W is called the longest common subsequence of the known sequences.

[0021] Backtracking is performed based on the obtained longest common subsequence to extract the corresponding events from the load event sequence of the user to be identified and the load event sequence of each transformer. The user-transformer load events that constitute the elements in the subsequence are obtained, and several event pairs are obtained. Each event pair consists of an event in the user sequence and a corresponding event in the transformer sequence, forming a set of one-to-one corresponding event pairs. The strings formed by the two load events in the event pair are exactly the same.

[0022] For each event pair, the time deviation between the user event and the transformer event is calculated in seconds. The time deviation is the absolute difference between the occurrence times of the two events.

[0023] The time deviation values ​​of all event pairs are formed into a set, and the entropy of all time deviation values ​​is calculated;

[0024] The time deviation entropy values ​​corresponding to the user and each transformer are calculated, and the transformer corresponding to the minimum time difference entropy value is determined as the transformer belonging to the user.

[0025] Furthermore, the calculation formula of the entropy of the time deviation value is: Where n is the number of different values ​​of the time deviation, p i is the probability of each deviation value appearing in the set.

[0026] A low-voltage distribution network household transformer relationship verification device, comprising:

[0027] A load event extraction module is used to extract load events based on the load waveforms of users and transformers, wherein the load events include load event data in which the active power value exceeds a preset threshold;

[0028] A load event extraction module is used to extract load characteristics according to the extracted load events, wherein the load characteristics include load event occurrence time, load type, load active power value and corresponding load reactive power value;

[0029] A character processing module is used to characterize the extracted load features and arrange them in a predetermined order to obtain a load event character string sequence of the user to be identified;

[0030] A user attribution discrimination module is configured to calculate the longest common subsequence between the load event string sequence of the user to be identified and the load event string sequence of each transformer, calculate the time deviation entropy value corresponding to the user to be identified and each transformer based on the longest common subsequence, and determine the transformer to which the user to be identified belongs based on the time deviation entropy value;

[0031] The verification module is used to verify the transformer to which the determined user to be identified belongs with the transformer to which the user to be identified belongs recorded in the system archive.

[0032] Furthermore, the character processing module is specifically used to:

[0033] Divide the time of each day into fixed time intervals and name each interval with a number. Convert the load event occurrence time into the number TS of the interval where the load event occurs.

[0034] Identify the load type TYPE of the corresponding load event, convert the step type character into S, the impact type character into P, and other types that are not step type characters or impact type characters into E;

[0035] The load active power value is divided into intervals of 100 watts and is converted into characters and recorded as QS;

[0036] The reactive power value of the load is divided into intervals of 100Var and is processed into characters, which are recorded as Q.

[0037] The four characterized characteristic characters corresponding to a single load event are combined into a character string EVENT, EVENT = {TS, TYPE, PS, QS};

[0038] Arrange the character strings corresponding to each load event at the user transformer end to be identified in the order of the load event occurrence time to obtain the load event character string sequence of the user to be identified; if the load event occurrence time TS is the same, arrange them in descending order of priority of load type S, P, E, with S as the highest priority character; if the arrangement order still cannot be distinguished, the one with the smaller PS is arranged first; if the PS is the same, the one with the smaller QS is arranged first.

[0039] Furthermore, the user attribution determination module is specifically configured to:

[0040] The load event string sequence of the user to be identified is matched with the load event string sequence of each transformer by the longest common subsequence matching calculation. The longest common subsequence refers to a sequence W that is a subsequence of two or more known sequences and is the longest among all sequences that meet this condition. Then W is called the longest common subsequence of the known sequences.

[0041] Backtracking is performed based on the obtained longest common subsequence to extract the corresponding events from the load event sequence of the user to be identified and the load event sequence of each transformer. The user-transformer load events that constitute the elements in the subsequence are obtained, and several event pairs are obtained. Each event pair consists of an event in the user sequence and a corresponding event in the transformer sequence, forming a set of one-to-one corresponding event pairs. The strings formed by the two load events in the event pair are exactly the same.

[0042] For each event pair, the time deviation between the user event and the transformer event is calculated in seconds. The time deviation is the absolute difference between the occurrence times of the two events.

[0043] The time deviation values ​​of all event pairs are formed into a set, and the entropy of all time deviation values ​​is calculated;

[0044] The time deviation entropy values ​​corresponding to the user and each transformer are calculated, and the transformer corresponding to the minimum time difference entropy value is determined as the transformer belonging to the user.

[0045] Furthermore, the calculation formula of the entropy of the time deviation value is: Where n is the number of different values ​​of the time deviation, p i is the probability of each deviation value appearing in the set.

[0046] A low-voltage distribution network user-transformer relationship verification system, comprising: a computer-readable storage medium and a processor;

[0047] The computer-readable storage medium is used to store executable instructions;

[0048] The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the low-voltage distribution network user-transformer relationship verification method.

[0049] A non-transient computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for verifying the relationship between household transformers in a low-voltage distribution network.

[0050] The present invention has the following beneficial effects:

[0051] 1. Some existing low-voltage transformer identification methods require the addition of additional measuring equipment, resulting in high installation and operation and maintenance costs, and are prone to failure in complex scenarios, and there are certain operational safety risks. The present invention mainly utilizes the characteristics of low-voltage user electrical appliances when they are turned on. These characteristics belong to natural power consumption characteristics and do not require additional equipment to be installed. In addition, these characteristics have deterministic signal properties and high recognition accuracy, which effectively solves the accuracy bottleneck of existing topology recognition technology and has great engineering application value.

[0052] 2. Some existing methods for identifying low-voltage transformers in low-voltage areas use a data-driven approach, using AMI measurement data for identification. However, this approach places high demands on data accuracy, as noise in AMI data can significantly affect accuracy. This invention uses a character sequence matching approach for identification, characterizing numerical data based on their intervals. This fuzzifies numerical accuracy and reduces the need for data precision, enabling good recognition results even in areas with complex data. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 The present invention is a flowchart of a method for verifying the relationship between household transformers in a low-voltage distribution network. DETAILED DESCRIPTION

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0055] See also Figure 1 The first aspect of the present invention provides a method for verifying the relationship between household transformers in a low-voltage distribution network, comprising the following steps:

[0056] Step 1: Extract load events based on the load waveforms of the user and transformer. The load events include load event data where the active power value exceeds a preset threshold. The load event data includes: the time period of the event, the active power and reactive power of the event, and the type of event.

[0057] Specifically, all power waveform data recorded on the same day from user and transformer load waveforms is selected to detect and extract load events. Load events with active power values ​​greater than or equal to 800 watts and their corresponding occurrence times are recorded. The time format is year-month-day-hour-minute-second, using the 24-hour timekeeping method.

[0058] Step 2: Extract load features based on the load events extracted in step 1. The load features include load event occurrence time, load type, load active power value, and corresponding load reactive power value.

[0059] Based on the load event extracted in step 1, the corresponding load characteristics of the identified user-side terminal (referring to the user-side terminal and the transformer-side terminal) are extracted and recorded. In this embodiment of the present invention, the load event occurrence time, load type, load active power value, and corresponding load reactive power value are selected to construct the load characteristics.

[0060] Step 3: Tokenize the load features extracted in step 2 and arrange them in a pre-ordered sequence to obtain a load event string sequence of the user to be identified. The user to be identified refers to the set of users whose transformers need to be confirmed.

[0061] Step 3 specifically includes:

[0062] Step 3.1: The load event occurrence time is divided into intervals of 15 minutes and characterized. That is, a numerical range is divided into several intervals, and each value is expressed using the corresponding character of the interval.

[0063] The specific processing method is: if the load event occurs at T.start, then T = seconds(T.start). The seconds(*) function is used to convert the corresponding duration into a numerical value in seconds, such as seconds(1 minute - 25 seconds) = 85 seconds. The characterized T is recorded as TS, TS = floor(T / 900) + 1. The floor(*) function rounds the input value down to the maximum integer not greater than the input value, such as floor(3.2) = 3. Numerical expression is a form of character expression. For example, the character expression for 0:01:25 is "1", and 1:15:28 is "6". Each day is divided into 96 time periods, each 15 minutes long. The character expression represents the time period number of the time period. The purpose of this step is to convert the second-level time records into minute-level data, that is, minute-by-minute time intervals of each 15 minutes. In short, the number of the 15-minute period in which the time occurs is recorded. TS is the final character representation, which can be {1, 2, 3, ....95, 96}.

[0064] Step 3.2: Identify the load type TYPE of the corresponding load event, characterize the step type as S, the impulse type as P, and other types that do not belong to the above two types as E.

[0065] Step 3.3: The load active power value is converted into characters with 100 watts as a division interval. The active power value calculated from the recorded data is recorded as P, which is converted into characters as PS, where PS = floor(P / 100) + 1.

[0066] Step 3.4: The load reactive power value is digitized with 100Var as a division interval. The active power value calculated from the recorded data is recorded as Q, which is then digitized as QS, where QS = floor(Q / 100) + 1.

[0067] Step 3.5: Combine the four characterized characteristic characters corresponding to a single load event into a character string EVENT, EVENT = {TS, TYPE, PS, QS}.

[0068] Step 3.6: Arrange the character strings corresponding to each load event at the user transformer to be identified in the order of the load event occurrence time to obtain the load event character string sequence for the user to be identified; if the load event occurrence time TS is the same, arrange them in descending order of load type S, P, and E, with S being the highest priority character; if the arrangement order still cannot be distinguished, the one with the smaller PS is arranged first; if the PS is the same, the one with the smaller QS is arranged first.

[0069] Step 4: Calculate the longest common subsequence between the load event string sequence of the user to be identified and the load event string sequence of each transformer, calculate the time deviation entropy value corresponding to the user to be identified and each transformer based on the longest common subsequence, and determine the transformer to which the user to be identified belongs based on the time deviation entropy value.

[0070] Step 4 specifically includes:

[0071] Step 4.1: Define the longest common subsequence: For a sequence W, if it is a subsequence of two or more known sequences and is the longest among all sequences that meet this condition, then W is called the longest common subsequence of the known sequences.

[0072] Step 4.2: Perform a longest common subsequence matching (LCS) between the load event string sequence of the user to be identified and the load event string sequence of each transformer. Using a dynamic programming algorithm, find the longest common subsequence between the load event string sequence of the user to be identified and the load event string sequence of each transformer.

[0073] Step 4.3: Backtrack based on the longest common subsequence obtained in Step 4.2 to extract the corresponding events from the load event sequence of the user to be identified and the load event sequences of each transformer. These events constitute the user-transformer load events of the elements in the subsequence, resulting in several event pairs. Each event pair consists of an event from the user sequence and a corresponding event from the transformer sequence, forming a one-to-one correspondence. The strings formed by the two load events in the event pair are identical.

[0074] Step 4.4: For each event pair, calculate the time offset between the user event and the transformer event in seconds. The time offset is the absolute difference between the occurrence times of the two events, in seconds.

[0075] Step 4.5: The time deviation values ​​of all event pairs are combined into a set, and the entropy of all time deviation values ​​is calculated. The entropy calculation formula is: Where n is the number of different values ​​of the time deviation, p i is the probability of each deviation value appearing in the set.

[0076] Step 4.6: Calculate the time deviation entropy value corresponding to the user and each transformer, and determine the transformer corresponding to the minimum time difference entropy value as the transformer belonging to the user.

[0077] Step 5: Verify household change relationship

[0078] Step 5.1: Apply step 4 to the load event sequence of each user to be identified to obtain the user change identification results of all users.

[0079] Step 5.2: Compare the household transformer identification results of all users with the system files. If the transformer number of a user given by the identification result is inconsistent with the transformer number of the user recorded in the system file, the user is identified as an abnormal user.

[0080] This embodiment of the present invention selected five substations, randomly selecting 20 users from each substation. Based on the data content and format required by the present invention, load data from 100 users and five transformer terminals was sampled over a four-day period. For the daily data from a particular substation, two users were randomly selected from each of the remaining four substations, and the data from the same day was mixed into the user data for that substation. The verification results using this invention are shown in Table 1:

[0081] Table 1

[0082]

[0083] According to experimental data, this method can detect incorrect users every day, with an average detection success rate exceeding 85%. In addition, this method only has missed detections within 4 days, with no incorrectly detected users, indicating high confidence in the method.

[0084] Another aspect of the present invention provides a low-voltage distribution network household transformer relationship verification device, comprising:

[0085] A load event extraction module is used to extract load events based on the load waveforms of users and transformers, wherein the load events include load event data in which the active power value exceeds a preset threshold;

[0086] A load event extraction module is used to extract load characteristics according to the extracted load events, wherein the load characteristics include load event occurrence time, load type, load active power value and corresponding load reactive power value;

[0087] A character processing module is used to characterize the extracted load features and arrange them in a predetermined order to obtain a load event character string sequence of the user to be identified;

[0088] A user attribution discrimination module is configured to calculate the longest common subsequence between the load event string sequence of the user to be identified and the load event string sequence of each transformer, calculate the time deviation entropy value corresponding to the user to be identified and each transformer based on the longest common subsequence, and determine the transformer to which the user to be identified belongs based on the time deviation entropy value;

[0089] The verification module is used to verify the transformer to which the determined user to be identified belongs with the transformer to which the user to be identified belongs recorded in the system archive.

[0090] Another aspect of the present invention provides a low-voltage distribution network user-transformer relationship verification system, comprising: a computer-readable storage medium and a processor;

[0091] The computer-readable storage medium is used to store executable instructions;

[0092] The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the low-voltage distribution network user-transformer relationship verification method described in the first aspect.

[0093] Another aspect of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for verifying the relationship between household transformers in a low-voltage distribution network as described in the first aspect is implemented.

[0094] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0095] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0096] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for verifying the relationship between household transformers in a low-voltage distribution network, characterized in that: The steps include: Extracting load events based on the load waveforms of users and transformers, wherein the load events include load event data in which the active power value exceeds a preset threshold; Extracting load features according to the extracted load events, wherein the load features include load event occurrence time, load type, load active power value, and corresponding load reactive power value; The extracted load features are digitized and arranged in a predetermined order to obtain a load event character string sequence of the user to be identified; Calculate the longest common subsequence between the load event string sequence of the user to be identified and the load event string sequence of each transformer, calculate the time deviation entropy value corresponding to the user to be identified and each transformer based on the longest common subsequence, and determine the transformer to which the user to be identified belongs based on the time deviation entropy value; Verify the transformer to which the determined user to be identified belongs with the transformer to which the user to be identified belongs recorded in the system archive; The method of calculating the longest common subsequence of the load event character string sequence of the user to be identified and the load event character string sequence of each transformer, calculating the time deviation entropy value corresponding to the user to be identified and each transformer according to the longest common subsequence, and determining the transformer to which the user to be identified belongs according to the time deviation entropy value includes: The load event string sequence of the user to be identified is matched with the load event string sequence of each transformer by the longest common subsequence matching calculation. The longest common subsequence refers to a sequence W that is a subsequence of two or more known sequences and is the longest among all sequences that meet this condition. Then W is called the longest common subsequence of the known sequences. Backtracking is performed based on the obtained longest common subsequence to extract the corresponding events from the load event sequence of the user to be identified and the load event sequence of each transformer. The user-transformer load events that constitute the elements in the subsequence are obtained, and several event pairs are obtained. Each event pair consists of an event in the user sequence and a corresponding event in the transformer sequence, forming a set of one-to-one corresponding event pairs. The strings formed by the two load events in the event pair are exactly the same. For each event pair, the time deviation between the user event and the transformer event is calculated in seconds. The time deviation is the absolute difference between the occurrence times of the two events. The time deviation values ​​of all event pairs are formed into a set, and the entropy of all time deviation values ​​is calculated; The time deviation entropy values ​​corresponding to the user and each transformer are calculated, and the transformer corresponding to the minimum time difference entropy value is determined as the transformer belonging to the user.

2. The method for verifying the relationship between household transformers in a low-voltage distribution network according to claim 1, wherein: The extracted load features are characterized and arranged in sequence according to a pre-set order to obtain a load event character string sequence for the user to be identified, including: Divide the time of each day into fixed time intervals and name each interval with a number. Convert the load event occurrence time into the number TS of the interval where the load event occurs. Identify the load type TYPE of the corresponding load event, convert the step type character into S, the impact type character into P, and other types that are not step type characters or impact type characters into E; The load active power value is divided into intervals of 100 watts and is converted into characters and recorded as QS; The reactive power value of the load is divided into intervals of 100 watts and is converted into characters, which are recorded as Q. Combine the four characterized characteristic characters corresponding to a single load event into a character string EVENT, EVENT={TS, TYPE, PS, QS}; Arrange the character strings corresponding to each load event at the user transformer end to be identified in the order of the load event occurrence time to obtain the load event character string sequence of the user to be identified; if the load event occurrence time TS is the same, arrange them in descending order of priority of load type S, P, E, with S as the highest priority character; if the arrangement order still cannot be distinguished, the one with the smaller PS is arranged first; if the PS is the same, the one with the smaller QS is arranged first.

3. The method for verifying the relationship between household transformers in a low-voltage distribution network according to claim 1, wherein: The calculation formula for the entropy of the time deviation value is: ,in is the number of different values ​​of the time deviation, is the probability of each deviation value appearing in the set.

4. A low-voltage distribution network household transformer relationship verification device, characterized in that: include: A load event extraction module is used to extract load events based on the load waveforms of users and transformers, wherein the load events include load event data in which the active power value exceeds a preset threshold; A load event extraction module is used to extract load characteristics according to the extracted load events, wherein the load characteristics include load event occurrence time, load type, load active power value and corresponding load reactive power value; A character processing module is used to characterize the extracted load features and arrange them in a predetermined order to obtain a load event character string sequence of the user to be identified; A user attribution discrimination module is configured to calculate the longest common subsequence between the load event string sequence of the user to be identified and the load event string sequence of each transformer, calculate the time deviation entropy value corresponding to the user to be identified and each transformer based on the longest common subsequence, and determine the transformer to which the user to be identified belongs based on the time deviation entropy value; A verification module, configured to verify the transformer to which the determined user to be identified belongs with the transformer to which the user to be identified belongs recorded in the system archive; The user attribution determination module is specifically used to: The load event string sequence of the user to be identified is matched with the load event string sequence of each transformer by the longest common subsequence matching calculation. The longest common subsequence refers to a sequence W that is a subsequence of two or more known sequences and is the longest among all sequences that meet this condition. Then W is called the longest common subsequence of the known sequences. Backtracking is performed based on the obtained longest common subsequence to extract the corresponding events from the load event sequence of the user to be identified and the load event sequence of each transformer. The user-transformer load events that constitute the elements in the subsequence are obtained, and several event pairs are obtained. Each event pair consists of an event in the user sequence and a corresponding event in the transformer sequence, forming a set of one-to-one corresponding event pairs. The strings formed by the two load events in the event pair are exactly the same. For each event pair, the time deviation between the user event and the transformer event is calculated in seconds. The time deviation is the absolute difference between the occurrence times of the two events. The time deviation values ​​of all event pairs are formed into a set, and the entropy of all time deviation values ​​is calculated; The time deviation entropy values ​​corresponding to the user and each transformer are calculated, and the transformer corresponding to the minimum time difference entropy value is determined as the transformer belonging to the user.

5. The low-voltage distribution network household-transformer relationship verification device according to claim 4, characterized in that: The character processing module is specifically used to: Divide the time of each day into fixed time intervals and name each interval with a number. Convert the load event occurrence time into the number TS of the interval where the load event occurs. Identify the load type TYPE of the corresponding load event, convert the step type character into S, the impact type character into P, and other types that are not step type characters or impact type characters into E; The load active power value is divided into intervals of 100 watts and is converted into characters and recorded as QS; The reactive power value of the load is divided into intervals of 100 watts and is converted into characters, which are recorded as Q. Combine the four characterized characteristic characters corresponding to a single load event into a character string EVENT, EVENT={TS, TYPE, PS, QS}; Arrange the character strings corresponding to each load event at the user transformer end to be identified in the order of the load event occurrence time to obtain the load event character string sequence of the user to be identified; if the load event occurrence time TS is the same, arrange them in descending order of priority of load type S, P, E, with S as the highest priority character; if the arrangement order still cannot be distinguished, the one with the smaller PS is arranged first; if the PS is the same, the one with the smaller QS is arranged first.

6. The low-voltage distribution network household-transformer relationship verification device according to claim 4, characterized in that: The calculation formula for the entropy of the time deviation value is: ,in is the number of different values ​​of the time deviation, is the probability of each deviation value appearing in the set.

7. A low-voltage distribution network household transformer relationship verification system, comprising: Computer-readable storage medium and processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the low-voltage distribution network user-transformer relationship verification method according to any one of claims 1 to 3.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for verifying the relationship between users and transformers in a low-voltage distribution network according to any one of claims 1 to 3 is implemented.

Images