Low-voltage area household relationship identification method and device

Abstract

This application relates to a method and apparatus for identifying the relationship between a user's electricity meter and a transformer substation in a low-voltage distribution area. The method includes: resampling a first electrical data sequence of the user's electricity meter within a target time period and a second electrical data sequence of different candidate distribution areas within the same target time period to obtain a first target sequence and a second target sequence; constructing a first anchor event sequence for the user's electricity meter and a second anchor event sequence for the candidate distribution areas; determining an event matching score between the user's electricity meter and the candidate distribution areas at different candidate time offsets based on the first anchor event sequence and the second anchor event sequence of the candidate distribution areas; determining the optimal time offset for each candidate distribution area based on the event matching score; ensuring that the time difference between any two adjacent candidate time offsets is the same; and determining the target distribution area to which the user's electricity meter belongs from among the different candidate distribution areas based on the optimal time offsets of different candidate distribution areas. This method can improve the accuracy of identifying the relationship between the user's electricity meter and the transformer substation.

Description

Technical Field

[0001] This application relates to the field of power distribution network topology identification technology, and in particular to a method and device for identifying the relationship between low-voltage transformer substations and households. Background Technology

[0002] With the development of low-voltage distribution network technology and the large-scale construction of smart meters and advanced metering systems, smart meters on the user side periodically upload data such as voltage, current, active power, electrical energy, and events such as power outages, power restorations, undervoltage, and overvoltage to the distribution area integration terminal every 15, 30, or 60 minutes. The distribution area integration terminal performs line loss calculation, fault location, power outage range assessment, distribution transformer load analysis, anti-electricity theft analysis, and integrated operation and distribution data management based on the data sent by the smart meters.

[0003] However, in real engineering environments, the relationship between smart meters and transformer substations on the user side is determined through manual on-site verification, resulting in poor accuracy and real-time performance. Summary of the Invention

[0004] Therefore, it is necessary to provide a method and device for identifying the relationship between a household and a transformer in a low-voltage distribution area, which can improve the accuracy of household-transformer relationship identification, in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a method for identifying the relationship between households and transformers in a low-voltage distribution area, the method comprising:

[0006] Based on the target sampling period, the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period are resampled to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence.

[0007] Based on the first target sequence, a first anchor event sequence for the user's electricity meter is constructed, and based on each second target sequence, a second anchor event sequence for the corresponding candidate transformer area is constructed.

[0008] For each candidate transformer substation, based on the first anchor point event sequence and the second anchor point event sequence of the candidate substation, determine the event matching score between the user's meter and the candidate substation at different candidate time offsets; and,

[0009] Based on the matching scores of each event, the optimal time offset of the candidate station area is determined; wherein, the time difference between any two adjacent candidate time offsets is the same;

[0010] Based on the optimal time offset of different candidate distribution areas, the target distribution area to which the user's electricity meter belongs is determined from the different candidate distribution areas.

[0011] In one possible implementation, the event matching score between the user's electricity meter and the candidate distribution area under different candidate time offsets is determined based on the first anchor event sequence and the second anchor event sequence of the candidate distribution area. This includes: for each candidate time offset, using each moment in the second anchor event sequence of the candidate distribution area as a reference moment, and aligning the first anchor event sequence and the second anchor events of the candidate distribution area based on the candidate time offset; and determining each valid moment in the second anchor event sequence of the candidate distribution area based on the alignment result; and determining the event matching score between the user's electricity meter and the candidate distribution area under the candidate time offset based on each set of reference event data in the first anchor event sequence and the second anchor event sequence of the candidate distribution area. Each set of reference event data includes event data corresponding to a valid moment in the second anchor event sequence of the candidate distribution area and event data corresponding to a target moment in the first anchor event sequence. The target moment corresponding to each valid moment is the sum or difference between the valid moment and the candidate time offset.

[0012] In one possible implementation, the optimal time offset of the candidate station area is determined based on the event matching scores, including any of the following: selecting the maximum value among the event matching scores and determining the neighborhood time range to which the candidate time offset corresponding to the maximum value belongs; for each candidate time offset within the neighborhood time range, determining the comprehensive matching score corresponding to the candidate time offset based on the event matching score corresponding to the candidate time offset; and determining the optimal time offset of the candidate station area from each candidate time offset within the neighborhood time range based on each comprehensive matching score.

[0013] In one possible implementation, a comprehensive matching score is determined based on the event matching score corresponding to the candidate time offset, including: determining the offset rationality score based on the ratio of the candidate time offset to a preset duration; and determining the power-related score based on the first target sequence and the second target sequence corresponding to the candidate transformer area; weighting the event matching score, offset rationality score, and power-related score corresponding to the candidate time offset to obtain the comprehensive matching score; wherein the target sequence includes at least one of a voltage sequence and a power sequence; the target sequence includes a voltage sequence, and the power-related score includes a differential voltage-related score, and / or, the target sequence includes a power sequence, and the power-related score includes at least one of a power change-related score and a power balance improvement score.

[0014] In one possible implementation, the target transformer area to which the user's meter belongs is determined from the different candidate transformer areas based on the optimal time offset of the different candidate transformer areas. This includes: selecting the minimum value among the optimal time offsets of the different candidate transformer areas; determining the candidate transformer area corresponding to the minimum value as the alternative transformer area to which the user's meter belongs, and determining the target confidence level of the alternative transformer area; and determining the alternative transformer area as the target transformer area to which the user's meter belongs if the target confidence level is greater than a preset confidence threshold.

[0015] In one possible implementation, determining the target confidence level of the candidate transformer substation includes: determining at least one confidence level among a plurality of candidate confidence levels, and determining the target confidence level of the candidate transformer substation based on the weighted processing result of the determined confidence levels; wherein the plurality of candidate confidence levels includes at least two of the following: scoring interval confidence level, event matching confidence level, time offset stability confidence level, power correlation confidence level, and time reliability confidence level of the first electrical data sequence.

[0016] In one possible implementation, determining at least one confidence level among multiple candidate confidence levels includes at least one of the following: determining the peak significance of a candidate transformer substation based on the ratio of the comprehensive matching score corresponding to the candidate substation to the sum of the comprehensive matching scores corresponding to the optimal time offset of each candidate substation, and determining the time offset stability confidence level of the candidate substation based on the peak significance; determining the scoring interval confidence level based on the difference between the reference matching score and the comprehensive matching score corresponding to the candidate substation; wherein the reference matching score is only less than the comprehensive matching score corresponding to the candidate substation among the comprehensive matching scores corresponding to the optimal time offset of each candidate substation; determining the event matching score corresponding to the candidate substation as the event matching confidence level of the candidate substation; determining the power-related score as the power-related confidence level of the candidate substation; determining at least one time reliability assessment parameter for the first electrical data sequence, and determining the time reliability confidence level of the first electrical data sequence based on the weighted processing result of the determined time reliability assessment parameters; wherein the at least one time reliability assessment parameter includes at least one of sampling time stability score, data missing rate score, anchor event availability score, and clock state score.

[0017] In one possible implementation, determining at least one time reliability assessment parameter for the first electrical data sequence includes at least one of the following: determining a sampling time stability score for the first electrical data sequence based on each two adjacent sampling times in the first electrical data sequence and the theoretical sampling period of the first electrical data sequence; determining a missing rate score for the first electrical data sequence based on the ratio of the number of missing sampling points in the first target sequence to the total number of theoretical sampling points in the first target sequence; determining an anchor event availability score for the first electrical data sequence based on the ratio of the number of each anchor event in the first anchor event sequence to the expected number of events; and obtaining the operating status of the clock in the user's electricity meter and determining a clock status score for the first electrical data sequence based on the operating status.

[0018] Secondly, this application also provides a low-voltage transformer substation relationship identification device, the device comprising:

[0019] The resampling module is used to resample the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period according to the target sampling period, so as to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence.

[0020] The construction module is used to construct a first anchor point event sequence for the user's electricity meter based on a first target sequence, and to construct a second anchor point event sequence for the corresponding candidate transformer area based on each second target sequence.

[0021] The determination module is used to determine, for each candidate transformer substation, an event matching score between the user's meter and the candidate transformer substation at different candidate time offsets based on the first anchor point event sequence and the second anchor point event sequence of the candidate transformer substation; and to determine the optimal time offset of the candidate transformer substation based on each event matching score; wherein the time difference between any two adjacent candidate time offsets is the same.

[0022] The determination module is used to determine the target transformer area to which the user's meter belongs from different candidate transformer areas based on the optimal time offset of different candidate transformer areas.

[0023] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods provided in the first aspect above.

[0024] Fourthly, this application also provides a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods provided in the first aspect above.

[0025] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the methods provided in the first aspect above.

[0026] The aforementioned low-voltage transformer substation relationship identification method and apparatus, after acquiring the first electrical data sequence of the user's electricity meter within a target time period and the second electrical data sequences of multiple candidate transformer substations within the target time period, resamples to obtain the first target sequence and the second target sequence. From the first target sequence, a first anchor event sequence corresponding to the user's electricity meter event is extracted, and from the second target sequence, a second anchor event sequence corresponding to the candidate transformer substation event is extracted. Thus, the anchor events occurring at the user's electricity meter and the candidate transformer substations can be determined using the first and second anchor event sequences. Then, for each candidate transformer substation, an event matching score is calculated based on the anchor events under different candidate time offsets, thereby determining the optimal time offset for each candidate substation. By comparing the optimal time offsets, the target transformer substation with the smallest offset can be identified. Thus, by calculating the event matching score of the anchor events under different candidate time offsets, the transformer substation relationship between the user's electricity meter and the candidate transformer substations can be determined from the dimensions of time offset and anchor events, thereby improving the real-time performance and accuracy of transformer substation relationship identification. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a flowchart illustrating a method for identifying the relationship between a household and a transformer in a low-voltage distribution area, as shown in one embodiment.

[0029] Figure 2 for Figure 1 A flowchart illustrating the specific implementation method of S105 in the illustrated embodiment;

[0030] Figure 3 This is a flowchart illustrating a time-based confidence calculation method in one embodiment;

[0031] Figure 4 This is a flowchart illustrating the clock state score calculation method in another embodiment;

[0032] Figure 5 This is an exemplary schematic diagram of a low-voltage transformer substation relationship identification method in one embodiment;

[0033] Figure 6This is a structural block diagram of a low-voltage transformer substation relationship identification device in one embodiment;

[0034] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0036] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments or any combination of multiple embodiments.

[0037] In one exemplary embodiment, such as Figure 1 As shown, a method for identifying the relationship between households and transformers in a low-voltage distribution area is provided. This method is applied to electronic devices. In this embodiment, the electronic device can be a distribution area fusion terminal, a distribution transformer intelligent terminal, a concentrator, an edge computing gateway, a power consumption information collection master station, a distribution and operation integrated data platform, a distribution network operation monitoring platform, a cloud big data platform, or a provincial or municipal-level metering automation system. The method includes steps S101-S105:

[0038] S101. Based on the target sampling period, resample the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence.

[0039] In this context, a low-voltage distribution area refers to the power supply area covered by a single distribution transformer and its power supply lines within a low-voltage power distribution system. The target sampling period is preset based on experience. The household-transformer relationship refers to the attribution and correspondence between a user's electricity meter and the distribution area.

[0040] Optionally, the first target sequence and the second target sequence are differential sequences, which can reflect synchronization disturbances and event changes on the user side and the transformer area side. For example, for the voltage sequence on the user side, a differential voltage sequence can be constructed: For the power sequence on the user side, construct a differential power sequence: For the voltage sequence on the transformer substation side, a differential voltage sequence is constructed: For the power sequence on the transformer substation side, a differential power sequence is constructed: .

[0041] Furthermore, electronic devices can employ equal-interval interpolation resampling, supplementing data points through linear interpolation, spline interpolation, or other methods to obtain the first target sequence and the second target sequence. Optionally, a voltage sequence can be constructed from the collected electrical data. Calculate the mean and standard deviation, in In the case of determining For outliers, interpolation is used to correct them. For power data, threshold comparison or quantile methods are used to identify outliers, and interpolation is then used to correct them.

[0042] In one example, the target sampling period can be expressed as: ,in, For different sampling times. The time interval of the target sampling period can be 15 minutes.

[0043] Candidate transformer areas can be specifically represented as: , For the m-th candidate station area, the candidate station area can be determined by at least one of the following methods:

[0044] Method 1: The original household registration information, specifying the relevant substation area;

[0045] Method 2: Several network zones near the user's geographical location;

[0046] Method 3: Units within the coverage area of ​​the same concentrator or adjacent concentrators;

[0047] Method 4: Related substations within the same administrative district, building, meter box, or branch area;

[0048] Method 5: Adjacent transformer substations that have historically experienced substation switching or cross-power supply;

[0049] Method 6: Collection of suspected transformer areas designated by maintenance personnel.

[0050] Method 7: Collect all data from all stations.

[0051] S102. Based on the first target sequence, construct the first anchor point event sequence of the user's electricity meter, and based on each second target sequence, construct the second anchor point event sequence of the corresponding candidate transformer area.

[0052] Among them, the anchor event sequence refers to the event sequence in the target sequence that has significant changes in electrical characteristics.

[0053] Specifically, electronic devices can extract anchor events based on electrical data mutation thresholds, and mark voltage and power fluctuations as anchor events when they exceed preset thresholds.

[0054] In one example, the first anchor event includes power outage events, power restoration events, voltage surge events, and power step events. The second anchor event includes transformer area power outage events, transformer area power restoration events, transformer area low-voltage side voltage surge events, transformer area total power step events, three-phase voltage synchronous change events, voltage disturbance events caused by reactive power compensation switching, and transformer uniformity surge events.

[0055] For the first anchor point event, the electronic device detects that the user voltage is less than a preset voltage threshold, i.e. And the event lasts longer than the preset duration. Then it is determined that a power outage event occurred on the user side at time t. The value can be 20V, 30V, or set according to the nominal voltage ratio.

[0056] The electronic device satisfies the condition that the user-side voltage has recovered from below the power outage threshold to the normal voltage state. ,and This indicates that a power restoration event occurs at time t. Wherein, It can be set to 180V, 190V, or according to the nominal voltage ratio.

[0057] Differential voltage sequence for user-side voltage ,exist In the case of a voltage surge event, a threshold value is used to determine the occurrence of the voltage surge event. A fixed threshold can be used, or it can be set adaptively: , 2 to 4 are acceptable.

[0058] Power increment sequence in user-side electrical data ,exist In the case of a power step event, it is determined that a power step event has occurred. .

[0059] S103. For each candidate transformer substation, determine the event matching score between the user's meter and the candidate transformer substation under different candidate time offsets based on the first anchor point event sequence and the second anchor point event sequence of the candidate transformer substation.

[0060] The event matching score quantifies the degree of matching between the first anchor event sequence and the second anchor event sequence at a specified candidate time offset. The candidate time offset is determined based on the maximum allowed time offset and the search step size. Given the maximum allowed time offset, different numbers of search steps are superimposed on it to obtain different candidate time offsets.

[0061] In one example, the maximum allowed time offset is The search step size is The set of candidate time offsets is: If the sampling period is 15 minutes, then , , .

[0062] S104. Based on the matching scores of each event, determine the optimal time offset for the candidate station area.

[0063] In this case, the time difference between any two adjacent candidate time offsets is the same.

[0064] Specifically, electronic devices can select the candidate time offset with the highest event matching score as the optimal time offset.

[0065] S105. Based on the optimal time offset of different candidate distribution areas, determine the target distribution area to which the user's meter belongs from the different candidate distribution areas.

[0066] Among them, the electronic equipment selects the candidate station area corresponding to the optimal time offset with the smallest value as the target station area.

[0067] Using the method provided in this application, after acquiring the first electrical data sequence of the user's electricity meter within a target time period and the second electrical data sequences of multiple candidate transformer substations within the target time period, a first target sequence and a second target sequence are obtained through resampling. A first anchor event sequence corresponding to the user's electricity meter event is extracted from the first target sequence, and a second anchor event sequence corresponding to the candidate transformer substation event is extracted from the second target sequence. Thus, the anchor events occurring at the user's electricity meter and the candidate transformer substations can be determined using the first and second anchor event sequences. Then, for each candidate transformer substation, an event matching score under different candidate time offsets is calculated using the anchor events, thereby determining the optimal time offset corresponding to each candidate substation. By comparing the optimal time offsets, the target transformer substation with the smallest offset can be determined. Thus, by calculating the event matching score of anchor events under different candidate time offsets, the relationship between the user's electricity meter and the candidate transformer substations can be determined from the dimensions of time offset and anchor events, thereby improving the real-time performance and accuracy of the relationship.

[0068] For S103 above, and for each candidate transformer substation, based on the first anchor point event sequence and the second anchor point event sequence of the candidate substation, the event matching score between the user's meter and the candidate substation under different candidate time offsets is determined. Specifically, this can be implemented as follows:

[0069] Step 1: For each candidate time offset, take each moment in the second anchor point event sequence of the candidate station area as the reference time, and perform time alignment between the first anchor point event sequence and the second anchor point event of the candidate station area based on the candidate time offset.

[0070] Among them, electronic devices can use candidate time offsets to positively adjust the timing of the first anchor point event sequence, and superimpose different candidate time offsets on the basis of the reference time to perform timing alignment.

[0071] Step 2: Based on the alignment results, determine the valid moments in the second anchor point event sequence of the candidate station area.

[0072] Step 3: Based on the reference event data in the first anchor point event sequence and the second anchor point event sequence of the candidate transformer area, determine the event matching score between the user meter and the candidate transformer area under the candidate time offset.

[0073] Each set of reference event data includes event data corresponding to a valid time in the second anchor point event sequence of the candidate station area and event data corresponding to the target time in the first anchor point event sequence; the target time corresponding to each valid time is the sum or difference between the valid time and the candidate time offset.

[0074] A valid moment refers to the moment when the event data corresponding to the second anchor point event sequence in the candidate station area is valid, and the sum of the event data and the candidate time offset is also valid for the event data corresponding to the first anchor point event sequence.

[0075] Specifically, the event matching score is calculated using the following formula: ;in, This indicates the time offset between user meter i and candidate transformer area j. The event matching score is below. To prevent small positive numbers from being divided by zero.

[0076] It should be noted that the embodiments of this application do not limit the calculation method of event matching score. In actual implementation, the calculation method of event matching score includes, but is not limited to: Pearson correlation coefficient, Spearman correlation coefficient, cosine similarity, Euclidean distance, Manhattan distance, mutual information, maximum cross-correlation, DTW distance, frequency domain amplitude spectrum similarity, wavelet energy similarity, edit distance, and Jaccard similarity.

[0077] Using the method provided in this application, after generating the first anchor point event sequence of the user's electricity meter and the second anchor point event sequence of the candidate transformer area, for each candidate time offset, the two sets of events are time-aligned based on the transformer area event time and the candidate time offset. Valid matching times are filtered, and event data is extracted to form a reference event group. An event matching score under the corresponding offset is calculated based on the reference event data of each group. By unifying the time series benchmark and accurately matching bidirectional events corresponding to the same physical disturbance, the problem of event timestamp misalignment is avoided, improving the accuracy of event matching and accurately quantifying the synchronization degree between the user and the transformer area electrical events under different time shifts. The event matching score is calculated based on the aligned event data, fully exploring the synchronization correlation of anchor point events, reducing the probability of misjudgment of the user transformer due to time series deviations, and improving the accuracy of time offset estimation.

[0078] Regarding S104 above, based on the matching scores of each event, the optimal time offset for the candidate station area is determined, specifically including:

[0079] The maximum value among the event matching scores is selected, and the neighborhood time range to which the candidate time offset corresponding to the maximum value belongs is determined. For each candidate time offset within the neighborhood time range, a comprehensive matching score is determined based on the event matching score corresponding to the candidate time offset. Based on each comprehensive matching score, the optimal time offset for the candidate station area is determined from all candidate time offsets within the neighborhood time range.

[0080] The electronic device selects a fixed number of sampling periods as the neighborhood time range, or dynamically determines the neighborhood time range based on the difference between different event matching scores. When the event matching score fluctuates significantly, the neighborhood time range is expanded. The comprehensive matching score is used to determine the correlation between user meter data and transformer area convergence terminal data from multiple dimensions. These dimensions include anchor events, voltage, and power. The electronic device calculates the event matching score, voltage matching score, and power matching score separately, and then performs a weighted sum of these scores to obtain the comprehensive matching score.

[0081] Specifically, the candidate time offset with the highest comprehensive matching score is taken as the optimal time offset.

[0082] In one example, for each user and candidate stations ,exist Internal calculation of comprehensive matching score under different candidate time offsets Then, the optimal time offset is determined from the overall matching score: .

[0083] The method provided in this application first filters the neighborhood time range corresponding to the highest event matching score from all candidate time offsets. Then, it calculates the comprehensive matching score by integrating multiple dimensions within the local neighborhood, and finally determines the optimal time offset. In this way, the optimization interval is narrowed down by coarsely filtering with the event matching score, reducing the amount of calculation of all global multi-indicators. Then, the comprehensive evaluation is refined within the neighborhood time range, so that the finally selected optimal time offset closely matches the actual time sequence deviation of on-site acquisition, improving the stability and reliability of self-calibration.

[0084] Furthermore, candidate time offsets can be decomposed into system-level offsets, concentrator-level offsets, and meter-level offsets: .in, Main station or system-level time offset; For users Time offset of the concentrator or data batch to which it belongs; This is due to the time offset of the user's electricity meter itself; Indicates user Belongs to the same concentrator. user set Introduce intragroup consistency constraints:

[0085]

[0086] in, For concentrator The common time offset.

[0087] Based on the event matching score corresponding to the candidate time offset, the comprehensive matching score corresponding to the candidate time offset is determined, which can be specifically implemented as follows:

[0088] Step A: Determine the offset rationality score corresponding to the candidate time offset based on the ratio of the candidate time offset to the preset duration; and determine the power-related score corresponding to the candidate time offset based on the first target sequence and the second target sequence corresponding to the candidate transformer area.

[0089] The higher the offset reasonableness score, the larger the time offset. Power-related scores include scores for preset dimensions. These preset dimensions include voltage, current, and power. Users can pre-set preset dimensions according to actual business needs, enabling electronic devices to calculate power-related scores for these preset dimensions.

[0090] In one example, the preset dimensions include differential voltage, power change, and power balance improvement. The method for calculating the differential voltage-related score is as follows: .in, It can be the Pearson correlation coefficient, Spearman correlation coefficient, or other correlation indicators. For the user-side smart meter at time t and according to the candidate time offset The voltage difference after the offset.

[0091] The calculation method for the power change-related score is as follows:

[0092]

[0093] in, For the user-side smart meter at time t and according to the candidate time offset The power difference after the offset.

[0094] The set of users whose ownership has been confirmed or is of high confidence in the candidate transformer area is: Add users to be identified The power balance errors before and after are as follows:

[0095]

[0096]

[0097] The power balance improvement score is:

[0098]

[0099] Among them, When it is positive, it indicates that a user has been added. The power balance error of the candidate distribution area decreased, indicating that the user belonged to the candidate distribution area. The likelihood of this increases.

[0100] Step B: Weight the event matching score, offset rationality score, and power-related score corresponding to the candidate time offset to obtain the comprehensive matching score corresponding to the candidate time offset.

[0101] The target sequence includes at least one of a voltage sequence and a power sequence; the target sequence includes a voltage sequence, and the power-related score includes a differential voltage-related score; and / or, the target sequence includes a power sequence, and the power-related score includes at least one of a power change-related score and a power balance improvement score.

[0102] In one example, the overall matching score is calculated using the following formula:

[0103]

[0104] in, Users can adaptively adjust weights based on different data availability scenarios. For example, in situations with frequent power outages and restorations, weights can be increased. When the voltage data quality is high When the power data quality is high, improve and In the case of missing power data, then let And renormalize; reduce the frequency of anchor events in the case of sparse anchor events. .

[0105] The method provided in this application's embodiments uses a weighted average of three categories of indicators: comprehensive event matching score, offset rationality score, and power-related score. The power-related score covers multi-dimensional features including differential voltage correlation, power change correlation, and power balance improvement. An offset rationality score is introduced to penalize excessively large or unreasonable time offsets, ensuring the calculation results closely reflect actual business scenarios and preventing the algorithm from selecting excessively large time shifts that are detached from engineering reality in pursuit of high event matching accuracy. Differential voltage and power increment sequences are used to replace the original electrical data to amplify the synchronization characteristics of common disturbances in the distribution area; simultaneously, a power balance improvement index is introduced to calculate the comprehensive matching score from multiple dimensions of data, improving the accuracy of the calculation results.

[0106] In some embodiments of this application, regarding the above-mentioned S105, determining the target transformer area to which the user's meter belongs from different candidate transformer areas based on the optimal time offset of different candidate transformer areas can be specifically implemented as S1051-S1503, such as... Figure 2 As shown:

[0107] S1051. Select the minimum value among the optimal time offsets of different candidate station areas.

[0108] Among them, for users The set of candidate transformer areas: Each candidate station area corresponds to an optimal score: The user's final transformer ownership was determined as follows: The user's time offset relative to their assigned station area is: Simultaneously output the candidate station sorting: This is used for subsequent verification and result interpretation.

[0109] S1052. The candidate transformer area corresponding to the minimum value is determined as the alternative transformer area to which the user's meter belongs, and the target confidence level of the alternative transformer area is determined.

[0110] The electronic equipment calculates the event matching confidence, offset reasonableness confidence, and power-related confidence for each candidate time offset. Then, it performs a weighted sum of the event matching confidence, offset reasonableness confidence, and power-related confidence to obtain the target confidence of the candidate transformer area.

[0111] The above-mentioned S1052, determining the candidate transformer area corresponding to the minimum value as the alternative transformer area to which the user's meter belongs, and determining the target confidence level of the alternative transformer area, can be implemented as follows:

[0112] Determine at least one confidence level from multiple candidate confidence levels, and determine the target confidence level of the candidate station area based on the weighted processing result of the determined confidence levels.

[0113] Among them, multiple candidate confidence levels include at least two of the following: scoring interval confidence level, event matching confidence level, time offset stability confidence level, power correlation confidence level, and time reliability confidence level of the first electrical data sequence.

[0114] Specifically, the event matching confidence score is defined as: The confidence level for the reasonableness of the offset is defined as follows: ,in, Power-related confidence levels include voltage-related confidence levels and power balance confidence levels. Voltage-related confidence levels are defined as follows: Power balance confidence level is defined as follows: .

[0115] The target confidence level can be defined as: .

[0116] in, The rating interval is specifically defined as follows: The larger the rating interval, the more clearly the user belongs to a specific TV station area. If a piece of data is unavailable, its corresponding weight is set to zero, and the remaining weights are renormalized.

[0117] Therefore, the transformer substation with the smallest time offset among the optimal time offsets of each candidate substation is first selected as a candidate substation, and then the final substation determination is completed by combining the target confidence threshold. In this way, the candidate substation with the smallest time offset selected by Xi'an as a candidate substation conforms to the on-site data collection time sequence pattern. The addition of a confidence threshold determination logic, which confirms the target substation only when the target confidence level and the preset confidence threshold are met, improves the reliability of the output household transformer relationship.

[0118] S1053. If the target confidence level is greater than the preset confidence level threshold, the candidate transformer area is determined as the target transformer area to which the user's electricity meter belongs.

[0119] The preset reliability threshold is set in advance based on experience.

[0120] In one example, a preset confidence threshold of 0.8 is used. Target areas are identified when the target confidence level is greater than or equal to 0.8. Target areas are also identified when the target confidence level is less than 0.8 but greater than or equal to 0.6, and these areas are periodically sampled. When the target confidence level is less than 0.6, a checklist is output and manually verified. When the target confidence level is less than the minimum preset confidence threshold or the time offset stability confidence level is less than 0.5, it indicates a time anomaly in the meter, concentrator, or data acquisition system, and the time status of the meter, concentrator, and data acquisition system is checked. If the number of valid events is less than the event quantity threshold or the data missing rate exceeds the missing rate threshold, no judgment result is output, the data is re-collected, and recalculated according to the method provided in this application's embodiments.

[0121] It should be noted that the above methods for determining household change relationships are only examples. In actual implementation, methods for determining household change relationships include, but are not limited to: the comprehensive score maximum method, logistic regression, support vector machine, random forest, gradient boosting tree, XGBoost, LightGBM, neural network, graph neural network, semi-supervised label propagation, Bayesian classification model, and Gaussian mixture model.

[0122] The method provided in this application integrates at least two types of confidence scores—scoring interval, event matching confidence, time offset stability confidence, power-related confidence, and time reliability confidence—to calculate the target confidence score. This multi-indicator weighting quantifies the reliability of the identification results from multiple dimensions, comprehensively reflecting various risk factors in the entire identification process. It accurately distinguishes between high-confidence and low-confidence identification samples, providing a clear grading basis for subsequent manual review and data re-collection, and reducing the workload of on-site verification by maintenance personnel.

[0123] In some embodiments of this application, determining at least one confidence level among multiple candidate confidence levels includes at least one of the following methods, such as Figure 3 As shown, the method includes:

[0124] S301. Determine the peak significance of the candidate station area based on the ratio of the comprehensive matching score corresponding to the candidate station area to the sum of the comprehensive matching scores corresponding to the optimal time offset of each candidate station area, and determine the time offset stability confidence level of the candidate station area based on the peak significance.

[0125] Specifically, the peak significance is calculated using the following formula. Among them, in If the value is greater than or equal to a preset peak threshold, the optimal time offset is determined to be statistically significant. If the data offset is less than the preset peak threshold, it is determined that the optimal time offset does not have data significance, and the optimal time offset is added to the review list or data re-collection list.

[0126] S302. Determine the confidence level of the scoring interval based on the difference between the reference matching score and the comprehensive matching score corresponding to the candidate station area.

[0127] Among them, the reference matching score is the only one less than the comprehensive matching score corresponding to the candidate station area in the comprehensive matching score corresponding to the optimal time offset of each candidate station area.

[0128] Specifically, the calculation method for the confidence level of the rating interval is described in the relevant embodiments above, and will not be repeated here.

[0129] S303. The event matching score corresponding to the candidate station area is determined as the event matching confidence level of the candidate station area.

[0130] S304. The power-related score is determined as the power-related confidence level of the candidate transformer area.

[0131] S305. Determine at least one time reliability assessment parameter for the first electrical data sequence, and determine the time reliability confidence level of the first electrical data sequence based on the weighted processing result of each determined time reliability assessment parameter.

[0132] Among them, at least one time reliability assessment parameter includes at least one of the following: sampling time stability score, data missing rate score, anchor event availability score, and clock status score.

[0133] Using the method provided in this application, five confidence sub-indicators are independently calculated through peak significance, front-end and back-end area score difference, event matching score, power-related score, and time-series multi-parameter weighting. The uniqueness of the optimal time shift is quantified based on peak significance to determine the stability and reliability of the time offset estimation; the score difference between front-end and back-end candidate transformer areas intuitively reflects the degree of distinction between the target transformer area and other transformer areas; and the event matching results are directly reused as sub-confidence indicators, making the calculation logic simple and efficient. This incorporates the quality of the original meter data into the confidence evaluation system, accurately locating identification risks caused by data loss, sampling disorder, and clock failure, facilitating maintenance personnel to quickly pinpoint the root cause of low reliability in the identification results.

[0134] In some embodiments of this application, for the determination of at least one time reliability assessment parameter of the first electrical data sequence, such as Figure 4 As shown, it includes:

[0135] S401. Based on each two adjacent sampling times in the first electrical data sequence and the theoretical sampling period of the first electrical data sequence, determine the sampling time stability score of the first electrical data sequence.

[0136] The sampling time stability score is determined by calculating the time difference between two adjacent sampling times and then using the time difference and the theoretical sampling period.

[0137] Specifically, the calculation method for the sampling time stability score is as follows: .in, The smaller the time difference, the closer the sampling time stability score is to 1.

[0138] S402. Determine the missing rate score of the first electrical data sequence based on the ratio of the number of missing sampling points in the first target sequence to the total number of theoretical sampling points in the first target sequence.

[0139] The data missing rate for user i's smart meter is defined as follows: ,in, This represents the number of missing sampling points. This represents the total number of theoretical sampling points. Correspondingly, the missing rate score is defined as: The lower the missing rate, the higher the missing rate score.

[0140] S403. Determine the anchor event availability score of the first electrical data sequence based on the ratio of the number of anchor events in the first anchor event sequence to the expected number of events.

[0141] Specifically, the anchor event availability score is defined as follows: ,in, The number of anchor events, The expected number of events.

[0142] S404. Obtain the operating status of the clock in the user's electricity meter, and determine the clock status score of the first electrical data sequence based on the operating status.

[0143] After calculating the clock state score, anchor event availability score, missing rate score, and sampling time stability score, these scores are weighted and summed to obtain the time reliability assessment parameters. Specifically, the time reliability assessment parameters can be expressed as: .in, In the absence of a clock state, Set to zero.

[0144] Using the method provided in this application, four time-series reliability assessment parameters are independently calculated based on sampling interval fluctuation, data missing rate, number of effective anchor events, and meter clock status. Among them, the sampling time stability score quantifies the degree of sampling interval disorder; the missing rate score quantifies data completeness; the anchor event availability score judges whether the basic data for self-calibration is sufficient, and identifies users who do not have enough event anchors and cannot accurately calibrate the time in advance; the clock status score directly captures the hardware clock failure of the user's meter. Thus, the time-series reliability is improved, the time-series quality of the original meter data is fully quantified, and the accuracy of the calculation results is improved.

[0145] The following combination Figure 5 This application introduces a method for identifying the relationship between households and transformers in a low-voltage distribution area, as provided in its embodiments. Figure 5 As shown, the method includes:

[0146] S501. Obtain time-series data and event data of user smart meters and candidate distribution area master meters.

[0147] Among them, the timing data and event data are the electrical data and anchor point event data in the above embodiments.

[0148] Specifically, time-series data includes one or more of the following:

[0149] Table 1 Time Series Data

[0150] User voltage sequence Time series data 1 minute, 5 minutes, 15 minutes, 30 minutes or 60 minutes Required Used for voltage event extraction, voltage correlation, and self-calibration. User power sequence Time series data Same as above recommend Used for power step, power matching, and power balancing. User current sequence Time series data Same as above Optional Used for auxiliary judgment of load changes User electrical energy sequence Time series data Same as above Optional Used for auxiliary judgment of load trends User event logs Event Data Event level recommend Events including power outages, power restorations, undervoltage, and overvoltage. Transformer area total voltage meter Time series data Same as above Required Reference voltage curve for candidate transformer substations Total power of transformer area Time series data Same as above recommend Used for power matching and power balance verification in transformer substations. Taiwan District Incident Record Event Data Event level recommend Including power outages and restorations, voltage surges, and power surges in transformer substations. Concentrator Number Static archives —— Optional Used for hierarchical time offset modeling Original household registration change records Static archives —— Optional Used for candidate transformer area screening and conflict determination Geographic data Static archives —— Optional Used for candidate station area constraints

[0151] S502, Perform missing value processing, outlier processing, sampling period unification, and differential sequence construction on the data.

[0152] Here, "unified sampling period" refers to the process of resampling electrical data. "Differential sequence construction" refers to the process of constructing differential voltage sequences; the specific method is described in the relevant embodiments above.

[0153] S503, Calculate the timestamp reliability of each user's data.

[0154] The timestamp reliability is the clock status score in the above embodiments, and the specific calculation method is described in the relevant description in the above embodiments.

[0155] S504. Extract natural electrical event anchor points from user-side and transformer area-side data.

[0156] Specifically, the electronic equipment extracts power outage events, power restoration events, voltage surge events, and functional step events from electrical data on the user side and electrical data on the distribution area side, respectively. For specific methods, please refer to the relevant descriptions in the above embodiments.

[0157] S505. Construct user event sequences and station event sequences.

[0158] Specifically, the first anchor event sequence is constructed for user i: ; for candidate areas Construct the second anchor event sequence: .

[0159] in, For the power outage incident, This refers to the power restoration incident; This is a voltage surge event; This is a power step event; This refers to the event weight.

[0160] In one example, event weights can be set according to Table 2:

[0161] Table 2 Event Weights

[0162] Power outage 0.90 to 1.00 Resumption of power incident 0.90 to 1.00 Voltage mutation event 0.60 to 0.90 Total power step event in the distribution area 0.50 to 0.80 User power step event 0.30 to 0.70

[0163] S506. Perform a candidate time offset search for each user and each candidate station area.

[0164] The candidate time offset search is the process of constructing the candidate time offset set in the above embodiments. The specific method for constructing the candidate time offset set is described in the relevant descriptions in the above embodiments, and will not be repeated here.

[0165] It should be noted that the embodiments of this application do not impose specific limitations on the candidate time offset search method. In actual implementation, the candidate time offset search method includes, but is not limited to, exhaustive search, coarse-to-fine two-stage search, maximum cross-correlation search, dynamic time warping (DTW), Bayesian optimization, particle swarm optimization, genetic algorithm, gradient search, sliding window matching, hidden Markov model, and neural network attention alignment mechanism.

[0166] S507. Calculate the event matching, voltage correlation, power correlation, and power balance improvement scores under different candidate time offsets.

[0167] S508. Determine the optimal time offset for the user relative to the candidate station area.

[0168] S509. Determine the household change relationship based on the comprehensive score under the optimal time offset.

[0169] The comprehensive score is the comprehensive matching score in the above embodiment, and the candidate transformer area to which the user's meter belongs is determined based on the comprehensive matching score.

[0170] S510 outputs user-transformation relationship, time offset, identification confidence level, and low-confidence verification list.

[0171] Specifically, for each candidate station area, the identification confidence level of the candidate station area is calculated. The identification confidence level is the target confidence level in the above embodiment. The target station area is determined based on the target confidence level.

[0172] In one example, the output is shown in Table 3:

[0173] Table 3 Output Results

[0174] Household change relationship mapping table Dictionary / Table User 001 → Transformer T1 User's actual power supply transformer Optimal Time Offset numerical values User 001 → +15min Time offset of the user table relative to the station area table Overall confidence level numerical values User 001 → 0.91 Reliability of household transformer identification Time credibility numerical values User 001 → 0.82 Reliability of user time data Event matching degree numerical values User 001 → 0.88 Event anchor matching degree Voltage-related rating numerical values User 001 → 0.84 Voltage correlation after time shift correction Candidate station ranking List T1: 0.91, T2: 0.63 Interpretation of candidate results Low-confidence user list List User 023, User 047 Users who need to review List of users with time-related anomalies List User 051, User 089 There may be clock or acquisition anomalies. Concentrator Time Offset Mapping table Concentrator C1 → +12min Reference for Time Management of Data Acquisition Link

[0175] In one example, the user-transformer relationship identification method provided in this application is illustrated using a power outage and restoration event as an example. Three adjacent transformer substations exist within a certain power supply area: The user to be identified is .

[0176] Electronic devices acquire users The system retrieves the voltage, power, and event records for the past 7 days, and obtains the voltage, power, and event records for the master tables of the three candidate transformer areas.

[0177] At 02:15 on a certain day, in the Taiwan area A brief power outage occurred; power was restored at 02:42. (User) The event log shows that the power outage occurred at 02:30 and the power was restored at 02:57.

[0178] Construct a set of candidate time offsets: The event matching score is calculated for each candidate time offset by iterating through the candidate time offset set. At that time, the power outage and restoration incidents for users and the transformer substation The power outage and restoration events were aligned, and the differential voltage correlation score was the highest. Final determination of the electronic equipment: Output The overall confidence level is 0.93. Marked as high confidence.

[0179] In another example, the user-transformer relationship identification method provided in this application is illustrated using a voltage surge event as an example. A low-voltage distribution area experiences a step increase in the low-voltage side voltage at 10:00 AM on a certain day. The user to be identified... The voltage curve shows a similar voltage step at 10:30.

[0180] Electronic devices extract voltage surge events and search and calculate them within a candidate time offset set. When At that time, the user voltage surge event aligned with the transformer area voltage surge event, and the differential voltage correlation coefficient increased from 0.42 to 0.87 after time synchronization. Therefore, the electronic equipment output... The candidate time offset is 30 minutes, and the overall confidence level is 0.86.

[0181] In another example, taking a low-confidence identification scenario where multiple candidate transformer area scores are close as an example, the user change relationship identification method provided in this application embodiment is introduced. In this case, a certain user... For the two candidate transformer areas and The optimal comprehensive scores are as follows: ; The difference between the two is only: The difference is below the preset threshold of 0.10, and the number of valid event anchor points for the user is small. Therefore, the electronic device marks this user as a low-confidence user. Output results include: Preferred station area: Candidate station areas: Confidence level: low confidence; Recommended measures: extend the data observation window, check the data collection timestamp, manually verify the archives or connect the data on-site.

[0182] The method provided in this application embodiment enables automatic identification and updating of household-transformer relationships without the need for additional hardware equipment, relying on existing smart meters and transformer terminal data collection. Furthermore, it completes abnormal and missing data and handles outliers to ensure data accuracy. The time alignment based on user data timestamp reliability calculation and target sampling frequency further improves data accuracy. The accuracy of time series offset solution is improved through multi-dimensional fusion scoring, thereby improving the accuracy of identification results and ultimately achieving high-precision identification of household-transformer relationships in low-voltage transformer areas.

[0183] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0184] Based on the same inventive concept, this application also provides a low-voltage transformer substation relationship identification device for implementing the low-voltage transformer substation relationship identification method described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the low-voltage transformer substation relationship identification device provided below can be found in the limitations of the low-voltage transformer substation relationship identification method described above, and will not be repeated here.

[0185] In one exemplary embodiment, such as Figure 6 As shown, a low-voltage transformer substation relationship identification device is provided, wherein:

[0186] The resampling module 601 is used to resample the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period according to the target sampling period, so as to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence.

[0187] The construction module 602 is used to construct a first anchor event sequence of the user's electricity meter according to the first target sequence, and to construct a second anchor event sequence of the corresponding candidate transformer area according to each second target sequence.

[0188] The determining module 603 is configured to, for each candidate transformer substation, determine an event matching score between the user's meter and the candidate substation at different candidate time offsets, based on the first anchor event sequence and the second anchor event sequence of the candidate substation; and,

[0189] Based on the matching scores of each event, the optimal time offset of the candidate station area is determined; wherein, the time difference between any two adjacent candidate time offsets is the same;

[0190] The determining module 603 is used to determine the target transformer area to which the user's electricity meter belongs from different candidate transformer areas based on the optimal time offset of different candidate transformer areas.

[0191] In one possible implementation, module 603 is specifically used for:

[0192] For each candidate time offset, using each moment in the second anchor point event sequence of the candidate station area as a reference moment, and based on the candidate time offset, the first anchor point event sequence and the second anchor point events of the candidate station area are time-aligned; and,

[0193] Based on the alignment results, determine each valid moment in the second anchor point event sequence of the candidate station area;

[0194] Based on each set of reference event data in the first anchor event sequence and the second anchor event sequence of the candidate transformer area, the event matching score between the user meter and the candidate transformer area under the candidate time offset is determined;

[0195] Each set of reference event data includes event data corresponding to a valid time in the second anchor point event sequence of the candidate station area and event data corresponding to a target time in the first anchor point event sequence; the target time corresponding to each valid time is the sum or difference between the valid time and the candidate time offset.

[0196] In one possible implementation, module 603 is specifically used for:

[0197] Select the maximum value among the matching scores of each event, and determine the neighborhood time range to which the candidate time offset corresponding to the maximum value belongs;

[0198] For each candidate time offset within the neighborhood time range, a comprehensive matching score is determined based on the event matching score corresponding to the candidate time offset.

[0199] Based on the comprehensive matching scores, the optimal time offset of the candidate station area is determined from the candidate time offsets within the neighborhood time range.

[0200] In one possible implementation, module 603 is specifically used for:

[0201] Based on the ratio of the candidate time offset to a preset duration, a reasonableness score for the offset corresponding to the candidate time offset is determined; and,

[0202] Based on the first target sequence and the second target sequence corresponding to the candidate transformer area, the power-related score corresponding to the candidate time offset is determined;

[0203] The event matching score, offset rationality score, and power-related score corresponding to the candidate time offset are weighted to obtain the comprehensive matching score corresponding to the candidate time offset;

[0204] The target sequence includes at least one of a voltage sequence and a power sequence; the target sequence includes the voltage sequence, the power-related score includes a differential voltage-related score, and / or the target sequence includes the power sequence, the power-related score includes at least one of a power change-related score and a power balance improvement score.

[0205] In one possible implementation, module 603 is specifically used for:

[0206] Select the minimum value among the optimal time offsets of different candidate station areas;

[0207] The candidate transformer area corresponding to the minimum value is determined as the alternative transformer area to which the user's electricity meter belongs, and the target confidence level of the alternative transformer area is determined.

[0208] If the target confidence level is greater than a preset confidence threshold, the candidate transformer area is determined as the target transformer area to which the user's electricity meter belongs.

[0209] In one possible implementation, module 603 is specifically used for:

[0210] Determine at least one confidence level from a plurality of candidate confidence levels, and determine the target confidence level of the candidate station area based on the weighted processing result of the determined confidence levels;

[0211] The plurality of candidate confidence levels include at least two of the following: scoring interval confidence level, event matching confidence level, time offset stability confidence level, power correlation confidence level, and time reliability confidence level of the first electrical data sequence.

[0212] In one possible implementation, module 603 is specifically used for:

[0213] The peak significance of the candidate station is determined by the ratio of the comprehensive matching score corresponding to the candidate station area to the sum of the comprehensive matching scores corresponding to the optimal time offset of each candidate station area, and the time offset stability confidence of the candidate station area is determined based on the peak significance.

[0214] The confidence level of the scoring interval is determined based on the difference between the reference matching score and the comprehensive matching score corresponding to the candidate station area; wherein, the reference matching score is only less than the comprehensive matching score corresponding to the candidate station area in the comprehensive matching score corresponding to the optimal time offset of each candidate station area.

[0215] The event matching score corresponding to the candidate station area is determined as the event matching confidence level of the candidate station area;

[0216] The power-related score is determined as the power-related confidence level of the candidate transformer area;

[0217] Determine at least one time reliability assessment parameter for the first electrical data sequence, and determine the time reliability confidence level of the first electrical data sequence based on the weighted processing result of each determined time reliability assessment parameter; wherein the at least one time reliability assessment parameter includes at least one of sampling time stability score, data missing rate score, anchor event availability score, and clock state score.

[0218] In one possible implementation, module 603 is specifically used for:

[0219] Based on each two adjacent sampling times in the first electrical data sequence and the theoretical sampling period of the first electrical data sequence, the sampling time stability score of the first electrical data sequence is determined.

[0220] The missing rate score of the first electrical data sequence is determined based on the ratio of the number of missing sampling points in the first target sequence to the total number of theoretical sampling points in the first target sequence.

[0221] The anchor event availability score of the first electrical data sequence is determined based on the ratio of the number of anchor events in the first anchor event sequence to the expected number of events.

[0222] The operating status of the clock in the user's electricity meter is obtained, and based on the operating status, the clock status score of the first electrical data sequence is determined.

[0223] Each module in the aforementioned low-voltage transformer substation relationship identification device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0224] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for identifying the relationship between low-voltage transformer substations and households.

[0225] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0226] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0227] Based on the target sampling period, the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period are resampled to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence.

[0228] Based on the first target sequence, a first anchor event sequence for the user's electricity meter is constructed, and based on each second target sequence, a second anchor event sequence for the corresponding candidate transformer area is constructed.

[0229] For each candidate transformer substation, based on the first anchor event sequence and the second anchor event sequence of the candidate substation, an event matching score is determined between the user's meter and the candidate substation at different candidate time offsets; and,

[0230] Based on the matching scores of each event, the optimal time offset of the candidate station area is determined; wherein, the time difference between any two adjacent candidate time offsets is the same;

[0231] Based on the optimal time offset of different candidate distribution areas, the target distribution area to which the user's electricity meter belongs is determined from the different candidate distribution areas.

[0232] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0233] For each candidate time offset, using each moment in the second anchor point event sequence of the candidate station area as a reference moment, and based on the candidate time offset, the first anchor point event sequence and the second anchor point events of the candidate station area are time-aligned; and,

[0234] Based on the alignment results, determine each valid moment in the second anchor point event sequence of the candidate station area;

[0235] Based on each set of reference event data in the first anchor event sequence and the second anchor event sequence of the candidate transformer area, the event matching score between the user meter and the candidate transformer area under the candidate time offset is determined;

[0236] Each set of reference event data includes event data corresponding to a valid time in the second anchor point event sequence of the candidate station area and event data corresponding to a target time in the first anchor point event sequence; the target time corresponding to each valid time is the sum or difference between the valid time and the candidate time offset.

[0237] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0238] Select the maximum value among the matching scores of each event, and determine the neighborhood time range to which the candidate time offset corresponding to the maximum value belongs;

[0239] For each candidate time offset within the neighborhood time range, a comprehensive matching score is determined based on the event matching score corresponding to the candidate time offset.

[0240] Based on the comprehensive matching scores, the optimal time offset of the candidate station area is determined from the candidate time offsets within the neighborhood time range.

[0241] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0242] Based on the ratio of the candidate time offset to a preset duration, a reasonableness score for the offset corresponding to the candidate time offset is determined; and,

[0243] Based on the first target sequence and the second target sequence corresponding to the candidate transformer area, the power-related score corresponding to the candidate time offset is determined;

[0244] The event matching score, offset rationality score, and power-related score corresponding to the candidate time offset are weighted to obtain the comprehensive matching score corresponding to the candidate time offset;

[0245] The target sequence includes at least one of a voltage sequence and a power sequence; the target sequence includes the voltage sequence, the power-related score includes a differential voltage-related score, and / or the target sequence includes the power sequence, the power-related score includes at least one of a power change-related score and a power balance improvement score.

[0246] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0247] Select the minimum value among the optimal time offsets of different candidate station areas;

[0248] The candidate transformer area corresponding to the minimum value is determined as the alternative transformer area to which the user's electricity meter belongs, and the target confidence level of the alternative transformer area is determined.

[0249] If the target confidence level is greater than a preset confidence threshold, the candidate transformer area is determined as the target transformer area to which the user's electricity meter belongs.

[0250] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0251] Determine at least one confidence level from a plurality of candidate confidence levels, and determine the target confidence level of the candidate station area based on the weighted processing result of the determined confidence levels;

[0252] The plurality of candidate confidence levels include at least two of the following: scoring interval confidence level, event matching confidence level, time offset stability confidence level, power correlation confidence level, and time reliability confidence level of the first electrical data sequence.

[0253] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0254] The peak significance of the candidate station is determined by the ratio of the comprehensive matching score corresponding to the candidate station area to the sum of the comprehensive matching scores corresponding to the optimal time offset of each candidate station area, and the time offset stability confidence of the candidate station area is determined based on the peak significance.

[0255] The confidence level of the scoring interval is determined based on the difference between the reference matching score and the comprehensive matching score corresponding to the candidate station area; wherein, the reference matching score is only less than the comprehensive matching score corresponding to the candidate station area in the comprehensive matching score corresponding to the optimal time offset of each candidate station area.

[0256] The event matching score corresponding to the candidate station area is determined as the event matching confidence level of the candidate station area;

[0257] The power-related score is determined as the power-related confidence level of the candidate transformer area;

[0258] Determine at least one time reliability assessment parameter for the first electrical data sequence, and determine the time reliability confidence level of the first electrical data sequence based on the weighted processing result of each determined time reliability assessment parameter; wherein the at least one time reliability assessment parameter includes at least one of sampling time stability score, data missing rate score, anchor event availability score, and clock state score.

[0259] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0260] Based on each two adjacent sampling times in the first electrical data sequence and the theoretical sampling period of the first electrical data sequence, the sampling time stability score of the first electrical data sequence is determined.

[0261] The missing rate score of the first electrical data sequence is determined based on the ratio of the number of missing sampling points in the first target sequence to the total number of theoretical sampling points in the first target sequence.

[0262] The anchor event availability score of the first electrical data sequence is determined based on the ratio of the number of anchor events in the first anchor event sequence to the expected number of events.

[0263] The operating status of the clock in the user's electricity meter is obtained, and based on the operating status, the clock status score of the first electrical data sequence is determined.

[0264] In one embodiment, a readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0265] Based on the target sampling period, the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period are resampled to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence.

[0266] Based on the first target sequence, a first anchor event sequence for the user's electricity meter is constructed, and based on each second target sequence, a second anchor event sequence for the corresponding candidate transformer area is constructed.

[0267] For each candidate transformer substation, based on the first anchor event sequence and the second anchor event sequence of the candidate substation, an event matching score is determined between the user's meter and the candidate substation at different candidate time offsets; and,

[0268] Based on the matching scores of each event, the optimal time offset of the candidate station area is determined; wherein, the time difference between any two adjacent candidate time offsets is the same;

[0269] Based on the optimal time offset of different candidate distribution areas, the target distribution area to which the user's electricity meter belongs is determined from the different candidate distribution areas.

[0270] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0271] For each candidate time offset, using each moment in the second anchor point event sequence of the candidate station area as a reference moment, and based on the candidate time offset, the first anchor point event sequence and the second anchor point events of the candidate station area are time-aligned; and,

[0272] Based on the alignment results, determine each valid moment in the second anchor point event sequence of the candidate station area;

[0273] Based on each set of reference event data in the first anchor event sequence and the second anchor event sequence of the candidate transformer area, the event matching score between the user meter and the candidate transformer area under the candidate time offset is determined;

[0274] Each set of reference event data includes event data corresponding to a valid time in the second anchor point event sequence of the candidate station area and event data corresponding to a target time in the first anchor point event sequence; the target time corresponding to each valid time is the sum or difference between the valid time and the candidate time offset.

[0275] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0276] Select the maximum value among the matching scores of each event, and determine the neighborhood time range to which the candidate time offset corresponding to the maximum value belongs;

[0277] For each candidate time offset within the neighborhood time range, a comprehensive matching score is determined based on the event matching score corresponding to the candidate time offset.

[0278] Based on the comprehensive matching scores, the optimal time offset of the candidate station area is determined from the candidate time offsets within the neighborhood time range.

[0279] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0280] Based on the ratio of the candidate time offset to a preset duration, a reasonableness score for the offset corresponding to the candidate time offset is determined; and,

[0281] Based on the first target sequence and the second target sequence corresponding to the candidate transformer area, the power-related score corresponding to the candidate time offset is determined;

[0282] The event matching score, offset rationality score, and power-related score corresponding to the candidate time offset are weighted to obtain the comprehensive matching score corresponding to the candidate time offset;

[0283] The target sequence includes at least one of a voltage sequence and a power sequence; the target sequence includes the voltage sequence, the power-related score includes a differential voltage-related score, and / or the target sequence includes the power sequence, the power-related score includes at least one of a power change-related score and a power balance improvement score.

[0284] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0285] Select the minimum value among the optimal time offsets of different candidate station areas;

[0286] The candidate transformer area corresponding to the minimum value is determined as the alternative transformer area to which the user's electricity meter belongs, and the target confidence level of the alternative transformer area is determined.

[0287] If the target confidence level is greater than a preset confidence threshold, the candidate transformer area is determined as the target transformer area to which the user's electricity meter belongs.

[0288] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0289] Determine at least one confidence level from a plurality of candidate confidence levels, and determine the target confidence level of the candidate station area based on the weighted processing result of the determined confidence levels;

[0290] The plurality of candidate confidence levels include at least two of the following: scoring interval confidence level, event matching confidence level, time offset stability confidence level, power correlation confidence level, and time reliability confidence level of the first electrical data sequence.

[0291] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0292] The peak significance of the candidate station is determined by the ratio of the comprehensive matching score corresponding to the candidate station area to the sum of the comprehensive matching scores corresponding to the optimal time offset of each candidate station area, and the time offset stability confidence of the candidate station area is determined based on the peak significance.

[0293] The confidence level of the scoring interval is determined based on the difference between the reference matching score and the comprehensive matching score corresponding to the candidate station area; wherein, the reference matching score is only less than the comprehensive matching score corresponding to the candidate station area in the comprehensive matching score corresponding to the optimal time offset of each candidate station area.

[0294] The event matching score corresponding to the candidate station area is determined as the event matching confidence level of the candidate station area;

[0295] The power-related score is determined as the power-related confidence level of the candidate transformer area;

[0296] Determine at least one time reliability assessment parameter for the first electrical data sequence, and determine the time reliability confidence level of the first electrical data sequence based on the weighted processing result of each determined time reliability assessment parameter; wherein the at least one time reliability assessment parameter includes at least one of sampling time stability score, data missing rate score, anchor event availability score, and clock state score.

[0297] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0298] Based on each two adjacent sampling times in the first electrical data sequence and the theoretical sampling period of the first electrical data sequence, the sampling time stability score of the first electrical data sequence is determined.

[0299] The missing rate score of the first electrical data sequence is determined based on the ratio of the number of missing sampling points in the first target sequence to the total number of theoretical sampling points in the first target sequence.

[0300] The anchor event availability score of the first electrical data sequence is determined based on the ratio of the number of anchor events in the first anchor event sequence to the expected number of events.

[0301] The operating status of the clock in the user's electricity meter is obtained, and based on the operating status, the clock status score of the first electrical data sequence is determined.

[0302] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0303] Based on the target sampling period, the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period are resampled to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence.

[0304] Based on the first target sequence, a first anchor event sequence for the user's electricity meter is constructed, and based on each second target sequence, a second anchor event sequence for the corresponding candidate transformer area is constructed.

[0305] For each candidate transformer substation, based on the first anchor event sequence and the second anchor event sequence of the candidate substation, an event matching score is determined between the user's meter and the candidate substation at different candidate time offsets; and,

[0306] Based on the matching scores of each event, the optimal time offset of the candidate station area is determined; wherein, the time difference between any two adjacent candidate time offsets is the same;

[0307] Based on the optimal time offset of different candidate distribution areas, the target distribution area to which the user's electricity meter belongs is determined from the different candidate distribution areas.

[0308] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0309] For each candidate time offset, using each moment in the second anchor point event sequence of the candidate station area as a reference moment, and based on the candidate time offset, the first anchor point event sequence and the second anchor point events of the candidate station area are time-aligned; and,

[0310] Based on the alignment results, determine each valid moment in the second anchor point event sequence of the candidate station area;

[0311] Based on each set of reference event data in the first anchor event sequence and the second anchor event sequence of the candidate transformer area, the event matching score between the user meter and the candidate transformer area under the candidate time offset is determined;

[0312] Each set of reference event data includes event data corresponding to a valid time in the second anchor point event sequence of the candidate station area and event data corresponding to a target time in the first anchor point event sequence; the target time corresponding to each valid time is the sum or difference between the valid time and the candidate time offset.

[0313] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0314] Select the maximum value among the matching scores of each event, and determine the neighborhood time range to which the candidate time offset corresponding to the maximum value belongs;

[0315] For each candidate time offset within the neighborhood time range, a comprehensive matching score is determined based on the event matching score corresponding to the candidate time offset.

[0316] Based on the comprehensive matching scores, the optimal time offset of the candidate station area is determined from the candidate time offsets within the neighborhood time range.

[0317] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0318] Based on the ratio of the candidate time offset to a preset duration, a reasonableness score for the offset corresponding to the candidate time offset is determined; and,

[0319] Based on the first target sequence and the second target sequence corresponding to the candidate transformer area, the power-related score corresponding to the candidate time offset is determined;

[0320] The event matching score, offset rationality score, and power-related score corresponding to the candidate time offset are weighted to obtain the comprehensive matching score corresponding to the candidate time offset;

[0321] The target sequence includes at least one of a voltage sequence and a power sequence; the target sequence includes the voltage sequence, the power-related score includes a differential voltage-related score, and / or the target sequence includes the power sequence, the power-related score includes at least one of a power change-related score and a power balance improvement score.

[0322] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0323] Select the minimum value among the optimal time offsets of different candidate station areas;

[0324] The candidate transformer area corresponding to the minimum value is determined as the alternative transformer area to which the user's electricity meter belongs, and the target confidence level of the alternative transformer area is determined.

[0325] If the target confidence level is greater than a preset confidence threshold, the candidate transformer area is determined as the target transformer area to which the user's electricity meter belongs.

[0326] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0327] Determine at least one confidence level from a plurality of candidate confidence levels, and determine the target confidence level of the candidate station area based on the weighted processing result of the determined confidence levels;

[0328] The plurality of candidate confidence levels include at least two of the following: scoring interval confidence level, event matching confidence level, time offset stability confidence level, power correlation confidence level, and time reliability confidence level of the first electrical data sequence.

[0329] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0330] The peak significance of the candidate station is determined by the ratio of the comprehensive matching score corresponding to the candidate station area to the sum of the comprehensive matching scores corresponding to the optimal time offset of each candidate station area, and the time offset stability confidence of the candidate station area is determined based on the peak significance.

[0331] The confidence level of the scoring interval is determined based on the difference between the reference matching score and the comprehensive matching score corresponding to the candidate station area; wherein, the reference matching score is only less than the comprehensive matching score corresponding to the candidate station area in the comprehensive matching score corresponding to the optimal time offset of each candidate station area.

[0332] The event matching score corresponding to the candidate station area is determined as the event matching confidence level of the candidate station area;

[0333] The power-related score is determined as the power-related confidence level of the candidate transformer area;

[0334] Determine at least one time reliability assessment parameter for the first electrical data sequence, and determine the time reliability confidence level of the first electrical data sequence based on the weighted processing result of each determined time reliability assessment parameter; wherein the at least one time reliability assessment parameter includes at least one of sampling time stability score, data missing rate score, anchor event availability score, and clock state score.

[0335] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0336] Based on each two adjacent sampling times in the first electrical data sequence and the theoretical sampling period of the first electrical data sequence, the sampling time stability score of the first electrical data sequence is determined.

[0337] The missing rate score of the first electrical data sequence is determined based on the ratio of the number of missing sampling points in the first target sequence to the total number of theoretical sampling points in the first target sequence.

[0338] The anchor event availability score of the first electrical data sequence is determined based on the ratio of the number of anchor events in the first anchor event sequence to the expected number of events.

[0339] The operating status of the clock in the user's electricity meter is obtained, and based on the operating status, the clock status score of the first electrical data sequence is determined.

[0340] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0341] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0342] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for identifying the relationship between households and transformers in a low-voltage distribution area, characterized in that, The method includes: Based on the target sampling period, the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period are resampled to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence. Based on the first target sequence, a first anchor event sequence for the user's electricity meter is constructed, and based on each second target sequence, a second anchor event sequence for the corresponding candidate transformer area is constructed. For each candidate transformer substation, based on the first anchor event sequence and the second anchor event sequence of the candidate substation, an event matching score is determined between the user's meter and the candidate substation at different candidate time offsets; and, Based on the matching scores of each event, the optimal time offset of the candidate station area is determined; wherein, the time difference between any two adjacent candidate time offsets is the same; Based on the optimal time offset of different candidate distribution areas, the target distribution area to which the user's electricity meter belongs is determined from the different candidate distribution areas.

2. The method according to claim 1, characterized in that, The step of determining the event matching score between the user's electricity meter and the candidate distribution area under different candidate time offsets based on the first anchor point event sequence and the second anchor point event sequence of the candidate distribution area includes: For each candidate time offset, using each moment in the second anchor point event sequence of the candidate station area as a reference moment, and based on the candidate time offset, the first anchor point event sequence and the second anchor point events of the candidate station area are time-aligned; and, Based on the alignment results, determine each valid moment in the second anchor point event sequence of the candidate station area; Based on each set of reference event data in the first anchor event sequence and the second anchor event sequence of the candidate transformer area, the event matching score between the user meter and the candidate transformer area under the candidate time offset is determined; Each set of reference event data includes event data corresponding to a valid time in the second anchor point event sequence of the candidate station area and event data corresponding to a target time in the first anchor point event sequence; the target time corresponding to each valid time is the sum or difference between the valid time and the candidate time offset.

3. The method according to claim 1, characterized in that, The determination of the optimal time offset for the candidate station area based on the matching scores of each event includes any one of the following: Select the maximum value among the matching scores of each event, and determine the neighborhood time range to which the candidate time offset corresponding to the maximum value belongs; For each candidate time offset within the neighborhood time range, a comprehensive matching score is determined based on the event matching score corresponding to the candidate time offset. Based on the comprehensive matching scores, the optimal time offset of the candidate station area is determined from the candidate time offsets within the neighborhood time range.

4. The method according to claim 3, characterized in that, The step of determining the comprehensive matching score corresponding to the candidate time offset based on the event matching score corresponding to the candidate time offset includes: Based on the ratio of the candidate time offset to a preset duration, a reasonableness score for the offset corresponding to the candidate time offset is determined; and, Based on the first target sequence and the second target sequence corresponding to the candidate transformer area, the power-related score corresponding to the candidate time offset is determined; The event matching score, offset rationality score, and power-related score corresponding to the candidate time offset are weighted to obtain the comprehensive matching score corresponding to the candidate time offset; The target sequence includes at least one of a voltage sequence and a power sequence; the target sequence includes the voltage sequence, the power-related score includes a differential voltage-related score, and / or the target sequence includes the power sequence, the power-related score includes at least one of a power change-related score and a power balance improvement score.

5. The method according to claim 4, characterized in that, The step of determining the target transformer station to which the user's electricity meter belongs from different candidate transformer stations based on the optimal time offset of different candidate transformer stations includes: Select the minimum value among the optimal time offsets of different candidate station areas; The candidate transformer area corresponding to the minimum value is determined as the alternative transformer area to which the user's electricity meter belongs, and the target confidence level of the alternative transformer area is determined. If the target confidence level is greater than a preset confidence threshold, the candidate transformer area is determined as the target transformer area to which the user's electricity meter belongs.

6. The method according to claim 5, characterized in that, Determining the target confidence level of the candidate transformer area includes: Determine at least one confidence level from a plurality of candidate confidence levels, and determine the target confidence level of the candidate station area based on the weighted processing result of the determined confidence levels; The plurality of candidate confidence levels include at least two of the following: scoring interval confidence level, event matching confidence level, time offset stability confidence level, power correlation confidence level, and time reliability confidence level of the first electrical data sequence.

7. The method according to claim 6, characterized in that, Determining at least one confidence level from a plurality of candidate confidence levels includes at least one of the following: The peak significance of the candidate station is determined by the ratio of the comprehensive matching score corresponding to the candidate station area to the sum of the comprehensive matching scores corresponding to the optimal time offset of each candidate station area, and the time offset stability confidence of the candidate station area is determined based on the peak significance. The confidence level of the scoring interval is determined based on the difference between the reference matching score and the comprehensive matching score corresponding to the candidate station area; wherein, the reference matching score is only less than the comprehensive matching score corresponding to the candidate station area in the comprehensive matching score corresponding to the optimal time offset of each candidate station area. The event matching score corresponding to the candidate station area is determined as the event matching confidence level of the candidate station area; The power-related score is determined as the power-related confidence level of the candidate transformer area; Determine at least one time reliability assessment parameter for the first electrical data sequence, and determine the time reliability confidence level of the first electrical data sequence based on the weighted processing result of each determined time reliability assessment parameter; wherein the at least one time reliability assessment parameter includes at least one of sampling time stability score, data missing rate score, anchor event availability score, and clock state score.

8. The method according to claim 7, characterized in that, Determining at least one time reliability assessment parameter of the first electrical data sequence includes at least one of the following: Based on each two adjacent sampling times in the first electrical data sequence and the theoretical sampling period of the first electrical data sequence, the sampling time stability score of the first electrical data sequence is determined. The missing rate score of the first electrical data sequence is determined based on the ratio of the number of missing sampling points in the first target sequence to the total number of theoretical sampling points in the first target sequence. The anchor event availability score of the first electrical data sequence is determined based on the ratio of the number of anchor events in the first anchor event sequence to the expected number of events. The operating status of the clock in the user's electricity meter is obtained, and based on the operating status, the clock status score of the first electrical data sequence is determined.

9. A low-voltage transformer substation relationship identification device, characterized in that, The device includes: The resampling module is used to resample the first electrical data sequence of the user's electricity meter in the target time period and the second electrical data sequence of different candidate transformer areas in the target time period according to the target sampling period, so as to obtain the first target sequence corresponding to the first electrical data sequence and the second target sequence corresponding to each second electrical data sequence. The construction module is used to construct a first anchor event sequence for the user's electricity meter based on the first target sequence, and to construct a second anchor event sequence for the corresponding candidate transformer area based on each second target sequence. The determination module is configured to, for each candidate transformer substation, determine an event matching score between the user's meter and the candidate substation at different candidate time offsets, based on the first anchor event sequence and the second anchor event sequence of the candidate substation; and, Based on the matching scores of each event, the optimal time offset of the candidate station area is determined; wherein, the time difference between any two adjacent candidate time offsets is the same; The determining module is used to determine the target transformer area to which the user's electricity meter belongs from different candidate transformer areas based on the optimal time offset of different candidate transformer areas.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images