Methods, devices, storage media and equipment for filling reliability data in low-voltage distribution networks

Abstract

This application provides a method, apparatus, storage medium, and device for filling in reliability data in low-voltage distribution networks. This application acquires operational data collected by a sensor system during a power outage event. First, redundant data is filtered out based on the similarity between similar data entries. The data is then dimensionality reduced. The filling errors of different filling algorithms are learned and trained using sample data. Next, an evaluation value for the error of each filling algorithm is calculated using a differentiated objective function corresponding to the target scenario. The filling algorithm with the best filling effect is selected to fill in the data to be processed once. Abnormal data is identified in the filled data, and the abnormal data is treated as data to be filled again to obtain the final filled data. This effectively improves the quality of reliability data filling used to evaluate the reliability indicators of low-voltage distribution networks.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, storage medium and device for filling reliability data in low-voltage distribution networks. Background Technology

[0002] With the continuous improvement of informatization, intelligent equipment in low-voltage distribution networks is gradually increasing, and post-reliability statistics and analysis are gradually transitioning from medium-voltage distribution networks to low-voltage distribution networks, placing higher demands on the refined management of reliability. Therefore, research on statistical and analytical methods for low-voltage distribution network reliability based on intelligent terminal data fusion is of great practical significance for improving power supply reliability in different types of areas.

[0003] However, due to factors such as short-term failures of sensor systems used for data monitoring and collection in smart grids, or external interference and transmission errors, the collected reliability data is not complete and reliable, resulting in data anomalies and missing data in the original reliability data. Data quality needs to be improved through big data cleaning and filling of distribution networks. Summary of the Invention

[0004] This application provides a method, apparatus, storage medium, and device for filling in reliability data in a low-voltage distribution network, which can improve the quality of data filling and thus improve the accuracy of subsequent reliability index parameters.

[0005] In a first aspect, this application provides a method for filling reliability data in a low-voltage distribution network, the method comprising:

[0006] Acquire operational data collected by the sensor system when a power outage occurs in a low-voltage distribution network;

[0007] Calculate the similarity between similar entries in the operational data;

[0008] For each pair of data whose similarity exceeds a preset similarity boundary value, redundancy removal is performed to obtain the data to be processed;

[0009] Extract sample data from the data to be processed;

[0010] Multiple imputation algorithms were used to impute the sample data, and the imputation error of each algorithm was calculated.

[0011] An error evaluation value is calculated for each filling algorithm based on a preset differentiated objective function corresponding to the target scene; the error evaluation value is used to reflect the filling effect of each filling algorithm in the target scene, and the magnitude of the error evaluation value is negatively correlated with the filling effect;

[0012] The filling algorithm with the lowest error evaluation value is determined as the target filling algorithm;

[0013] The data to be processed is filled in based on the target filling algorithm;

[0014] Anomaly detection algorithms based on time series are used to identify anomalous data in the imputed data to be processed.

[0015] The abnormal data in the data to be processed is treated as the data to be filled and the filling process is performed again to obtain the filled data.

[0016] In one embodiment, the method further includes:

[0017] The reliability index of the distribution network is calculated based on the supplementary data.

[0018] In one embodiment, the process of imputing the sample data using multiple imputation algorithms and calculating the imputation error for each algorithm includes:

[0019] The sample data were filled using the mean imputation method, median imputation method, K-nearest neighbor imputation method, weighted K-nearest neighbor imputation method, and a hybrid iterative imputation method combining missing forest and weighted K-nearest neighbor, respectively, and the imputation results for each imputation algorithm were obtained.

[0020] Calculate the filling error for the filling results of each filling algorithm.

[0021] In one embodiment, the calculation of the error evaluation value for each filling algorithm based on a preset differentiated objective function corresponding to the target scene includes:

[0022] Obtain the corresponding differentiated objective function based on the target scenario;

[0023] The error evaluation value is calculated based on the differentiated objective function:

[0024]

[0025] Where F is the error evaluation value; i is the i-th data in the sample data; m is the total number of sample data; a i f represents the preset evaluation coefficient corresponding to the target scenario; i This refers to the filling error in the filling result.

[0026] In one embodiment, the imputation error is at least one of the following: absolute error, root mean square error, absolute percentage error, absolute error standard deviation, root mean square error standard deviation, and absolute percentage error standard deviation, calculated based on the true data value and the imputed data value.

[0027] In one embodiment, calculating the similarity between similar entries in the runtime data includes:

[0028] The similarity between any two data points is calculated using the following formula:

[0029] F(a,b)=p(a|b)p(b|a)tf ab

[0030] Where 'a' is one of the two data points used to calculate similarity, and 'b' is the other data point used to calculate similarity; p(a|b) is the probability that 'a' and 'b' both occur; p(b|a) is the probability that 'b' and 'a' both occur; tf ab Let be the correlation coefficient between a and b, and

[0031]

[0032] Where, n c The total number of data entries in class c; tf a,c f is the total number of times 'a' appears in category c; b,c This represents the total number of times 'b' appears in category c.

[0033] In one embodiment, the time-series anomaly detection algorithm for identifying anomalous data in the imputed data to be processed includes:

[0034] Calculation of normal data range based on interquartile range detection method:

[0035] Data in the supplemented data that does not belong to the normal data range after being sorted in ascending order are marked as abnormal data.

[0036] Secondly, this application provides a low-voltage distribution network reliability data filling device, comprising:

[0037] The data acquisition module is used to acquire the operational data collected by the sensor system when a power outage occurs in the low-voltage distribution network;

[0038] The similarity calculation module is used to calculate the similarity between similar entries in the running data;

[0039] The data dimensionality reduction module is used to remove redundancy from two data sets whose similarity exceeds a preset similarity boundary value, thus obtaining the data to be processed.

[0040] The data extraction module is used to extract sample data from the data to be processed;

[0041] The first calculation module is used to fill in the sample data using various filling algorithms and to calculate the filling error of each algorithm.

[0042] The second calculation module is used to calculate an error evaluation value for the filling error of each filling algorithm based on a preset differentiated objective function corresponding to the target scene; the error evaluation value is used to reflect the filling effect of each filling algorithm in the target scene, and the magnitude of the error evaluation value is negatively correlated with the filling effect;

[0043] An algorithm determination module is used to determine the filling algorithm with the lowest error evaluation value as the target filling algorithm;

[0044] A primary filling module is used to perform data filling processing on the data to be processed based on the target filling algorithm;

[0045] The abnormal data identification module is used to identify abnormal data in the imputed data to be processed based on a time series anomaly detection algorithm.

[0046] The secondary filling module is used to re-fill the abnormal data in the data to be processed as the data to be filled to obtain the filled data.

[0047] Thirdly, this application provides a storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the low-voltage distribution network reliability data filling method as described in any of the preceding claims.

[0048] Fourthly, this application provides a computer device, including: one or more processors, and a memory;

[0049] The memory stores computer-readable instructions, and when the one or more processors execute the computer-readable instructions, they perform the steps of the low-voltage distribution network reliability data filling method as described above.

[0050] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0051] The low-voltage distribution network reliability data filling method, apparatus, storage medium, and equipment provided in this application acquire operational data collected by sensor systems during power outage events. First, redundant data is filtered out based on the similarity between similar data entries. The data is then dimensionality reduced. The filling errors of different filling algorithms are obtained through training using sample data. Next, an evaluation value for the error of each filling algorithm is calculated using a differentiated objective function corresponding to the target scenario. The filling algorithm with the best filling effect is selected to perform data filling on the data to be processed. Abnormal data is identified in the filled data, and the abnormal data is used as data to be filled again to obtain the final filled data. This effectively improves the quality of reliability data filling used to evaluate the reliability indicators of low-voltage distribution networks. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a flowchart illustrating a low-voltage distribution network reliability data filling method in one embodiment.

[0054] Figure 2 This is a structural block diagram of a low-voltage distribution network reliability data filling device in one embodiment;

[0055] Figure 3 This is a diagram of the internal structure of a computer device in one embodiment. Detailed Implementation

[0056] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0057] like Figure 1 As shown, this application provides a method for filling in reliability data in a low-voltage distribution network, the method including steps S101 to S110, wherein:

[0058] Step S101: Obtain the operational data collected by the sensor system when a power outage event occurs in the low-voltage distribution network.

[0059] The sensor system is used to monitor various operating parameters of the low-voltage distribution network. The network's operating status during power outages is a crucial parameter for evaluating the reliability of the low-voltage distribution network; therefore, the sensor system collects various operating data of the low-voltage distribution network during power outages.

[0060] Step S102: Calculate the similarity between similar entries in the running data.

[0061] During data cleaning, the data is categorized according to predefined rules, dividing it into several entries for easier subsequent processing. Redundancy is assessed only within the same category of entries; therefore, during data cleaning, data similarity is calculated for each category. The process iterates through the data in each category until every data point has been compared with the remaining data in the same category.

[0062] Step S103: For each pair of data whose similarity exceeds the preset similarity boundary value, perform redundancy removal processing to obtain the data to be processed.

[0063] Two data points are randomly selected from the same category to calculate the similarity. If the similarity exceeds the preset similarity threshold, i.e. the similarity is too high, the two data points need to be redundant, i.e., one of the data points is deleted.

[0064] Step S104: Extract sample data from the data to be processed.

[0065] Extracting sample data from the data to be processed for screening and imputation algorithms can better reflect the patterns in the data and improve the accuracy of the imputation predictions.

[0066] Step S105: Fill in the sample data using various filling algorithms and calculate the filling error for each algorithm.

[0067] Based on the characteristics of the data, select the most suitable imputation algorithms for the data to be processed, use each imputation algorithm to imput the sample data, and calculate the corresponding imputation error.

[0068] In one embodiment, the process of imputing the sample data using multiple imputation algorithms and calculating the imputation error for each algorithm includes:

[0069] The sample data were imputed using the mean imputation method, median imputation method, K-nearest neighbor imputation method, weighted K-nearest neighbor imputation method, and a hybrid iterative imputation method combining missing forest and weighted K-nearest neighbor. The imputation results for each imputation algorithm were obtained. The imputation error was calculated for the imputation results of each algorithm.

[0070] In one embodiment, the imputation error is at least one of the following: absolute error, root mean square error, absolute percentage error, absolute error standard deviation, root mean square error standard deviation, and absolute percentage error standard deviation, calculated based on the true data value and the imputed data value.

[0071] Step S106: Calculate the error evaluation value for each filling algorithm based on a preset differentiated objective function corresponding to the target scene.

[0072] Among them, the error evaluation value is used to reflect the filling effect of each filling algorithm in the target scene, and the magnitude of the error evaluation value is negatively correlated with the filling effect.

[0073] Step S107: The filling algorithm with the lowest error evaluation value is determined as the target filling algorithm.

[0074] Step S108: Perform data imputation processing on the data to be processed based on the target imputation algorithm.

[0075] Currently, common data imputation methods involve learning and training on sample data to predict and fill in missing data. However, different algorithms perform differently in different scenarios, and choosing a single evaluation metric can only assess the imputation effect under a single condition. In this embodiment, by learning and training multiple imputation algorithms and then using differentiated objective functions for different scenarios to evaluate the imputation effect, it is possible to select an algorithm that is more suitable for the target scenario, optimize the imputation effect, and improve the quality of data imputation.

[0076] Step S109: Based on the time series anomaly detection algorithm, identify abnormal data in the filled data to be processed.

[0077] In existing technologies, the threshold for filtering outlier data relies entirely on manual setting. While this is relatively simple in terms of program setup, it is prone to misjudgments and its excessive reliance on human intervention makes it difficult to effectively identify outlier data, thus rendering it unsuitable for cleaning multi-source big data in power distribution networks. This embodiment utilizes a time-series anomaly detection algorithm to filter outlier data. Based on the overall distribution of the collected data, it can select outliers, thereby improving the accuracy of anomaly identification.

[0078] Step S110: The abnormal data in the data to be processed is taken as the data to be filled and the filling process is performed again to obtain the filled data.

[0079] Abnormal data can affect subsequent reliability index evaluations. Directly deleting abnormal data will result in some data loss, negatively impacting the accuracy of reliability index evaluations. In this embodiment, abnormal data is treated as data to be filled in and subjected to secondary filling processing, which can improve data integrity, improve data filling quality, and thus improve the accuracy of subsequent reliability index evaluations.

[0080] In this embodiment, by acquiring the operational data collected by the sensor system when a power outage occurs, redundant data is first filtered out based on the similarity between similar data entries, and the data is dimensionality reduced. The filling errors of different filling algorithms are obtained by learning and training using sample data. Then, the error evaluation value of each filling atomization is calculated by using a differentiated objective function corresponding to the target scenario. The filling algorithm with the best filling effect is selected to fill the data to be processed once. Abnormal data is identified in the filled data to be processed. Abnormal data is used as data to be filled and then re-filled to obtain the final filled data, thereby effectively improving the quality of reliability data filling used to evaluate the reliability indicators of low-voltage distribution networks.

[0081] In one embodiment, the low-voltage distribution network reliability data filling method further includes:

[0082] The reliability index of the distribution network is calculated based on the supplementary data.

[0083] In one embodiment, the calculation of the error evaluation value for each filling algorithm based on a preset differentiated objective function corresponding to the target scene includes:

[0084] Obtain the corresponding differentiated objective function based on the target scenario;

[0085] Calculate the error evaluation value based on the differentiated objective function:

[0086]

[0087] Where F is the error evaluation value; i is the i-th data in the sample data; m is the total number of sample data; a i f represents the preset evaluation coefficient corresponding to the target scenario; i This refers to the filling error in the filling result.

[0088] In one embodiment, calculating the similarity between similar entries in the runtime data includes:

[0089] The similarity between any two data points is calculated using the following formula:

[0090] F(a,b)=p(a|b)p(b|a)tf ab

[0091] Where 'a' is one of the two data points used to calculate similarity, and 'b' is the other data point used to calculate similarity; p(a|b) is the probability that 'a' and 'b' both occur; p(b|a) is the probability that 'b' and 'a' both occur; tf ab Let be the correlation coefficient between a and b, and

[0092]

[0093] Where, n c The total number of data entries in class c; tf a,c f is the total number of times 'a' appears in category c; b,c This represents the total number of times 'b' appears in category c.

[0094] In this embodiment, feature identification is performed based on the direction of redundancy reduction based on correlation. The data to be processed is compared with standard data, which can reasonably distinguish the problem of redundancy that is difficult to reduce due to changes in units, formats, etc., which are not related to content changes. This allows for optimal redundancy removal and a better reduction in redundancy level.

[0095] In one embodiment, the time-series anomaly detection algorithm for identifying anomalous data in the imputed data to be processed includes:

[0096] Calculation of normal data range based on interquartile range detection method:

[0097] Data that does not fall within the normal data range after being sorted in ascending order in the filled-in data set is marked as abnormal data.

[0098] The normal data range is x∈(Q1-1.5IQR,Q3+1.5IQR), where (Q1-1.5IQR) is the minimum observed value, (Q3+1.5IQR) is the maximum observed value, Q1 is the data in the 25th percentile after ascending sort of the same category of data, Q3 is the data in the 75th percentile after ascending sort of the same category of data, and IQR is the difference between the minimum and maximum observed values.

[0099] In this embodiment, compared with the traditional abnormal data monitoring method that requires setting upper and lower limits for abnormal data, outlier data can be selected based on the overall data collected, thereby improving the accuracy of abnormal data identification.

[0100] 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 of other steps.

[0101] The following describes the low-voltage distribution network reliability data filling device provided in the embodiments of this application. The low-voltage distribution network reliability data filling device described below and the low-voltage distribution network reliability data filling method described above can be referred to in correspondence.

[0102] like Figure 2 As shown in the figure, this application embodiment provides a low-voltage distribution network reliability data filling device 200, including:

[0103] Data acquisition module 201 is used to acquire the operating data collected by the sensor system when a power outage event occurs in the low-voltage distribution network;

[0104] Similarity calculation module 202 is used to calculate the similarity between similar entries in the running data;

[0105] The data dimensionality reduction module 203 is used to perform redundancy removal on two data sets whose similarity exceeds a preset similarity boundary value to obtain the data to be processed;

[0106] Data extraction module 204 is used to extract sample data from the data to be processed;

[0107] The first calculation module 205 is used to fill in the sample data using various filling algorithms and to calculate the filling error of each algorithm.

[0108] The second calculation module 206 is used to calculate the error evaluation value of each filling algorithm based on a preset differentiated objective function corresponding to the target scene; the error evaluation value is used to reflect the filling effect of each filling algorithm in the target scene, and the magnitude of the error evaluation value is negatively correlated with the filling effect;

[0109] The algorithm determination module 207 is used to determine the filling algorithm with the lowest error evaluation value as the target filling algorithm;

[0110] The primary filling module 208 is used to perform data filling processing on the data to be processed based on the target filling algorithm;

[0111] The abnormal data identification module 209 is used to identify abnormal data in the filled data to be processed based on the time series anomaly detection algorithm;

[0112] The secondary filling module 210 is used to re-fill the abnormal data in the data to be processed as the data to be filled to obtain the filled data.

[0113] In one embodiment, the low-voltage distribution network reliability data filling device further includes:

[0114] The reliability index calculation module is used to calculate the reliability index of the distribution network based on the supplementary data.

[0115] In one embodiment, the first computing module is configured to perform the following steps:

[0116] The sample data were filled using the mean imputation method, median imputation method, K-nearest neighbor imputation method, weighted K-nearest neighbor imputation method, and a hybrid iterative imputation method combining missing forest and weighted K-nearest neighbor. The imputation results for each imputation algorithm were obtained.

[0117] Calculate the filling error for the filling results of each filling algorithm.

[0118] In one embodiment, the second computing module is configured to perform the following steps:

[0119] Obtain the corresponding differentiated objective function based on the target scenario;

[0120] Calculate the error evaluation value based on the differentiated objective function:

[0121]

[0122] Where F is the error evaluation value; i is the i-th data in the sample data; m is the total number of sample data; a i f represents the preset evaluation coefficient corresponding to the target scenario; i This refers to the filling error in the filling result.

[0123] In one embodiment, the similarity calculation module is configured to perform the following steps:

[0124] The similarity between any two data points is calculated using the following formula:

[0125] F(a,b)=p(a|b)p(b|a)tf ab

[0126] Where 'a' is one of the two data points used to calculate similarity, and 'b' is the other data point used to calculate similarity; p(a|b) is the probability that 'a' and 'b' both occur; p(b|a) is the probability that 'b' and 'a' both occur; tf ab Let be the correlation coefficient between a and b, and

[0127]

[0128] Where, n c The total number of data entries in class c; tf a,c f is the total number of times 'a' appears in category c; b,c This represents the total number of times 'b' appears in category c.

[0129] In one embodiment, the abnormal data identification module is configured to perform the following steps:

[0130] Calculation of normal data range based on interquartile range detection method:

[0131] Data that does not fall within the normal data range after being sorted in ascending order in the filled-in data set is marked as abnormal data.

[0132] The division of modules in the aforementioned low-voltage distribution network reliability data filling device is merely illustrative. In other embodiments, the low-voltage distribution network reliability data filling device can be divided into different modules as needed to complete all or part of its functions. Each module in the aforementioned low-voltage distribution network reliability data filling 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 in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0133] In one embodiment, this application also provides a storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the low-voltage distribution network reliability data filling method as described in any of the foregoing embodiments.

[0134] In one embodiment, this application also provides a computer device storing computer-readable instructions, wherein when the one or more processors execute the computer-readable instructions, they perform the steps of the low-voltage distribution network reliability data filling method as described in any of the above embodiments.

[0135] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 3As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for filling reliability data in a low-voltage power distribution network. The display screen can be an LCD screen or an e-ink display screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0136] Those skilled in the art will understand that Figure 3 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.

[0137] 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 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, etc., and are not limited to these.

[0138] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0139] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0140] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0141] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for filling reliability data in a low-voltage distribution network, characterized in that, The method includes: Acquire operational data collected by the sensor system when a power outage occurs in a low-voltage distribution network; Calculate the similarity between similar entries in the operational data; For each pair of data whose similarity exceeds a preset similarity boundary value, redundancy removal is performed to obtain the data to be processed; Extract sample data from the data to be processed; Multiple imputation algorithms were used to impute the sample data, and the imputation error of each algorithm was calculated. An error evaluation value is calculated for each filling algorithm based on a preset differentiated objective function corresponding to the target scene; the error evaluation value is used to reflect the filling effect of each filling algorithm in the target scene, and the magnitude of the error evaluation value is negatively correlated with the filling effect; The filling algorithm with the lowest error evaluation value is determined as the target filling algorithm; The data to be processed is filled in based on the target filling algorithm; Anomaly detection algorithms based on time series anomalies are used to identify anomalous data in the imputed data to be processed. The abnormal data in the data to be processed is treated as the data to be filled and the filling process is performed again to obtain the filled data; The process involves applying multiple imputation algorithms to the sample data and calculating the imputation error for each algorithm, including: The sample data were filled using the mean imputation method, median imputation method, K-nearest neighbor imputation method, weighted K-nearest neighbor imputation method, and a hybrid iterative imputation method combining missing forest and weighted K-nearest neighbor, respectively, and the imputation results for each imputation algorithm were obtained. Calculate the filling error for the filling results of each filling algorithm; The calculation of the filling error evaluation value for each filling algorithm based on a preset differentiated objective function corresponding to the target scene includes: Obtain the corresponding differentiated objective function based on the target scenario; The error evaluation value is calculated based on the differentiated objective function: Where F is the error evaluation value; For the first in the sample data One data point; The total number of sample data; These are the preset evaluation coefficients corresponding to the target scenario; The filling error in the filling result; The imputation error is at least one of the following: absolute error, root mean square error, absolute percentage error, absolute error standard deviation, root mean square error standard deviation, and absolute percentage error standard deviation, calculated based on the true data value and the imputed data value.

2. The low-voltage distribution network reliability data filling method according to claim 1, characterized in that, The method further includes: The reliability index of the distribution network is calculated based on the supplementary data.

3. The low-voltage distribution network reliability data filling method according to claim 1, characterized in that, The calculation of the similarity between similar entries in the running data includes: The similarity between any two data points is calculated using the following formula: Where 'a' is one of the two data points used to calculate similarity, and 'b' is the other of the two data points used to calculate similarity. Let be the probability that 'a' occurs at the same time as 'b'. Let a be the probability that b occurs at the same time as a. Let be the correlation coefficient between a and b, and in, The total number of data entries in category c; This represents the total number of times 'a' appears in category c. This represents the total number of times 'b' appears in category c.

4. The low-voltage distribution network reliability data filling method according to claim 1, characterized in that, The time-series anomaly detection algorithm is used to identify anomalous data in the imputed data to be processed, including: Calculation of normal data range based on interquartile range detection method: Data in the supplemented data that does not belong to the normal data range after being sorted in ascending order are marked as abnormal data.

5. A low-voltage distribution network reliability data filling device, characterized in that, include: The data acquisition module is used to acquire the operational data collected by the sensor system when a power outage occurs in the low-voltage distribution network; The similarity calculation module is used to calculate the similarity between similar entries in the running data; The data dimensionality reduction module is used to remove redundancy from two data sets whose similarity exceeds a preset similarity boundary value, thus obtaining the data to be processed. The data extraction module is used to extract sample data from the data to be processed; The first calculation module is used to fill in the sample data using various filling algorithms and to calculate the filling error of each algorithm. The second calculation module is used to calculate an error evaluation value for the filling error of each filling algorithm based on a preset differentiated objective function corresponding to the target scene; the error evaluation value is used to reflect the filling effect of each filling algorithm in the target scene, and the magnitude of the error evaluation value is negatively correlated with the filling effect; An algorithm determination module is used to determine the filling algorithm with the lowest error evaluation value as the target filling algorithm; A primary filling module is used to perform data filling processing on the data to be processed based on the target filling algorithm; The abnormal data identification module is used to identify abnormal data in the filled-in data based on the time series anomaly detection algorithm. The secondary filling module is used to re-fill the abnormal data in the data to be processed as the data to be filled to obtain the filled data. The first computing module is configured to perform the following steps: The sample data were filled using the mean imputation method, median imputation method, K-nearest neighbor imputation method, weighted K-nearest neighbor imputation method, and a hybrid iterative imputation method combining missing forest and weighted K-nearest neighbor. The imputation results for each imputation algorithm were obtained. Calculate the filling error for the filling results of each filling algorithm; The second computing module is configured to perform the following steps: Obtain the corresponding differentiated objective function based on the target scenario; Calculate the error evaluation value based on the differentiated objective function: Where F is the error evaluation value; For the first in the sample data One data point; The total number of sample data; These are the preset evaluation coefficients corresponding to the target scenario; The filling error in the filling result; The imputation error is at least one of the following: absolute error, root mean square error, absolute percentage error, absolute error standard deviation, root mean square error standard deviation, and absolute percentage error standard deviation, calculated based on the true data value and the imputed data value.

6. A storage medium, characterized in that: The storage medium stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the low-voltage distribution network reliability data filling method as described in any one of claims 1 to 4.

7. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, and when the one or more processors execute the computer-readable instructions, they perform the steps of the low-voltage distribution network reliability data filling method as described in any one of claims 1 to 4.

Images