Charging pile fault data generation method and system

Abstract

The invention discloses a charging pile fault data generation method and system, and the method comprises the steps: collecting and recording the fault data of a charging pile, and carrying out the preprocessing; analyzing through data visualization to obtain distribution, characteristics and fault modes of the fault data in different dimensions; key features of the data are determined, a fault data probability model is established, and electric vehicle charging pile fault data are generated through a dynamic fault data synthesis algorithm; and carrying out visualization and quality evaluation. According to the method, real and effective charging pile fault data are generated, data enhancement is performed in combination with data visualization analysis and statistics, a large amount of high-quality fault data can be generated, the method can be used for training and optimizing a fault prediction model, prediction accuracy and reliability are improved, and the fault prediction efficiency is improved. Powerful data support can be provided for operation and maintenance of the charging pile, managers are helped to better understand fault characteristics and rules of the charging pile, targeted maintenance is formulated, and the safety and reliability of the charging system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle fault warning, and in particular to a charging pile fault data generation method and system. Background Art

[0002] With the rapid development of the electric vehicle industry, the number and scale of charging facilities, as key infrastructure supporting its normal operation, are showing a rapid growth trend. However, with the large-scale deployment and use of charging piles, their failure rate is becoming increasingly apparent, becoming an important factor affecting the normal charging of electric vehicles and traffic operation efficiency. Charging pile failures may be caused by environmental factors, such as extreme weather, long-term wear, etc., or by human factors such as operating errors or improper maintenance. If these failures cannot be identified and handled in a timely and effective manner, they may not only affect the overall operating efficiency of the charging system, but also pose a threat to the property and personal safety of users.

[0003] Traditional data generation methods are usually based on fixed patterns or rules. Although this method is simple and easy, it is often difficult to truly reflect the complexity and variability of charging piles in the actual operation process. In addition, due to the lack of sufficient real fault data support, traditional fault prediction methods often find it difficult to achieve ideal prediction results. In recent years, with the rapid development of computer technology, data visualization analysis and statistics have shown great potential in the field of fault prediction. By collecting and analyzing various types of fault data generated during the operation of charging piles, the inherent laws and characteristics of charging pile faults can be learned, thereby achieving effective prediction of charging pile faults. Therefore, how to generate real and effective charging pile fault data and improve the accuracy and robustness of charging pile fault prediction has become a problem that needs to be solved urgently. Based on this, this patent proposes an innovative charging pile fault data generation method, which aims to generate high-quality fault data by simulating the actual operating environment and fault mode of the charging pile, and provide strong support for subsequent fault prediction model training and optimization. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a charging pile fault data generation method to solve the problem of missing fault data of electric vehicle charging piles.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for generating charging pile fault data, comprising:

[0008] Collect and record the fault data of charging piles, and pre-process the collected data;

[0009] Analyze the preprocessed fault data through data visualization to obtain the distribution, characteristics and fault modes of the fault data in different dimensions;

[0010] Based on the results of the data visualization analysis, key features of the data are determined, a fault data probability model is established, and electric vehicle charging pile fault data is generated through a dynamic fault data synthesis algorithm;

[0011] The generated fault data is stored, visualized and quality evaluated.

[0012] As a preferred solution of the charging pile fault data generation method of the present invention, collecting fault data and preprocessing it includes:

[0013] Monitor the working status and environmental conditions of the charging pile in real time, convert the analog signal of the sensor into a digital signal, and record fault data;

[0014] The fault data includes the time, location, cause and manufacturer of the fault, and saves the relevant original data;

[0015] Preprocessing the collected data, wherein the preprocessing includes data cleaning and data aggregation;

[0016] The data cleaning includes removing duplicate, erroneous, incomplete and outlier data;

[0017] The data aggregation includes grouping the data according to site type and urban area, counting the number of failures in each time period of each grouped data, and counting the number of failures occurring in each 10-minute time period in a day.

[0018] As a preferred solution of the charging pile fault data generation method of the present invention, the data visualization analysis includes:

[0019] Through data visualization of charts, the distribution and characteristics of fault data are displayed to obtain analysis results;

[0020] Classify the fault data according to site type, and count the frequency ratio and cumulative time of different fault types;

[0021] Classify the fault data according to regions, and count the frequency ratios and cumulative time of different fault types;

[0022] The charging pile failure data is classified according to site classification and region classification, and the time distribution of charging pile failures is counted.

[0023] As a preferred solution of the charging pile fault data generation method of the present invention, the generating of electric vehicle charging pile fault data comprises:

[0024] Input parameters, including site type, region, and the number of charging pile fault data to be generated, to obtain the corresponding fault data set;

[0025] According to the charging pile fault data set, random sampling of data is performed to select fault data corresponding to the input area, and the corresponding fault causes are extracted;

[0026] Establish a fault data probability model for the fault data of site types and regions, and generate site type and regional directional fault data;

[0027] By establishing the probability model, a site type directional dynamic fault data and a regional directional fault data are generated respectively, multi-dimensional data fusion is performed, the start time and duration of the two data are selected, the arithmetic mean is calculated respectively, and the mean value is used as the value of the corresponding field of the newly generated fault data;

[0028] According to the total amount of charging pile fault data generated by the user, a loop iteration is performed until the user's needs are met. Each iteration repeats the steps of probability model construction and multi-dimensional data fusion to record the new fault data records that meet the requirements.

[0029] As a preferred solution of the charging pile fault data generation method of the present invention, establishing a fault data probability model includes:

[0030] According to the site type or region and the fault cause of the randomly selected data, set the query conditions for retrieval to obtain data that matches the specified site type or specified region and fault cause;

[0031] The amount of data obtained is compared with the data volume threshold. If the amount of data obtained is less than the data volume threshold, the search range is expanded until the amount of data is greater than the data volume threshold. If it is greater than the data volume threshold, a probability model is constructed to calculate the frequency of each fault cause. The calculated frequency value is used as the probability model, which is expressed as:

[0032] f i =n i / N

[0033] Among them, n i represents the frequency of occurrence of the i-th fault cause, N represents the frequency of all fault causes, and f i represents the frequency of occurrence of the i-th fault cause;

[0034] Randomly generate fault data through the probability module, and determine the fault cause of the next generated fault data according to the probability of each fault cause;

[0035] If new data is added to the database, the frequency is recalculated and the probability model is updated.

[0036] As a preferred solution of the charging pile fault data generation method of the present invention, the quality assessment includes:

[0037] Evaluate the quality of the generated data to determine whether there are missing values ​​in the data set, whether the range of values ​​is reasonable, and whether the records in the data set are consistent in format and units;

[0038] Compare the similarity between the generated data and the real data through statistical tests to confirm whether the distribution of the generated data is consistent with the distribution of the actual data set;

[0039] Compare the similarity between the variables in the generated data and the real data set to determine whether the difference is too large.

[0040] As a preferred solution of the charging pile fault data generation method of the present invention, the similarity calculation includes:

[0041] The similarity between data is evaluated by KL divergence, which is expressed as:

[0042]

[0043] Among them, p(x) represents the actual failure data of electric vehicle charging piles in a certain area, and q(x) represents the probability distribution of generated failure data.

[0044] In a second aspect, the present invention provides a charging pile fault data generation system, comprising:

[0045] The acquisition module is used to collect and record the fault data of the charging pile and pre-process the collected data;

[0046] The analysis module is used to analyze the pre-processed fault data through data visualization to obtain the distribution, characteristics and fault modes of the fault data in different dimensions;

[0047] A construction module is used to determine the key features of the data based on the results of the data visualization analysis, establish a fault data probability model, and generate electric vehicle charging pile fault data through a dynamic fault data synthesis algorithm;

[0048] The output module is used to store the generated fault data and perform visualization and quality assessment.

[0049] In a third aspect, the present invention provides an electronic device, comprising:

[0050] Memory and processor;

[0051] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the charging pile fault data generating method are implemented.

[0052] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the charging pile fault data generating method.

[0053] Compared with the prior art, the present invention has the following beneficial effects: the present invention can generate a large amount of high-quality fault data by generating real and effective charging pile fault data and combining data visualization analysis and statistics for data enhancement. These data can not only be used to train and optimize fault prediction models to improve prediction accuracy and reliability, but also provide strong data support for the operation and maintenance of charging piles. The method of the present invention can help charging pile managers better understand the fault characteristics and laws of charging piles, formulate targeted maintenance strategies, and improve the safety and reliability of the charging system. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0055] Figure 1 It is a schematic diagram of the overall process of a method for generating charging pile fault data according to an embodiment of the present invention;

[0056] Figure 2 A schematic diagram of the frequency ratios of different fault types of urban public charging piles according to a charging pile fault data generation method according to an embodiment of the present invention;

[0057] Figure 3 A schematic diagram of the accumulated duration of different fault types of a highway charging pile according to a method for generating charging pile fault data according to an embodiment of the present invention;

[0058] Figure 4 A schematic diagram of the frequency ratio of different fault types of charging piles in a certain district or county according to a method for generating charging pile fault data according to an embodiment of the present invention;

[0059] Figure 5A schematic diagram of the accumulated duration of different fault types of charging piles in a prefecture-level city according to a method for generating charging pile fault data according to an embodiment of the present invention;

[0060] Figure 6 A schematic diagram of a distribution ratio of fault time of a city public charging pile according to a method for generating fault data of a charging pile according to an embodiment of the present invention;

[0061] Figure 7 A schematic diagram of a distribution ratio of charging pile failure time in a prefecture-level city according to a charging pile failure data generation method according to an embodiment of the present invention;

[0062] Figure 8 A schematic diagram of a flow chart of generating single fault data of a charging pile fault data generating method according to an embodiment of the present invention;

[0063] Fig. 9 A schematic diagram of the frequency of different fault types of charging pile fault data in a certain district or county (generated data compared with real data) according to a charging pile fault data generation method according to an embodiment of the present invention;

[0064] Fig.10 A schematic diagram of a distribution ratio diagram of the start time of charging pile fault data in a certain district or county (generated data compared with real data) according to a charging pile fault data generation method according to an embodiment of the present invention;

[0065] Fig.11 This is a schematic diagram of the existence time distribution ratio of charging pile fault data in a certain district or county (comparison between generated and real data) according to the charging pile fault data generation method described in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0067] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0068] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0069] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0070] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0071] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0072] Example 1

[0073] Reference Figure 1-Figure 8 , is an embodiment of the present invention, and provides a charging pile fault data generation method, comprising:

[0074] S101, collecting and recording fault data of the charging pile, and pre-processing the collected data.

[0075] Preferably, the working status and environmental conditions of the charging pile are monitored in real time, the analog signal of the sensor is converted into a digital signal, and the fault data is recorded;

[0076] Fault data includes the time, location, cause and manufacturer of the fault, and saves relevant original data;

[0077] Preprocess the collected data, including data cleaning and data aggregation;

[0078] Data cleaning includes removing duplicate, erroneous, incomplete and outlier data to ensure data quality;

[0079] Data aggregation includes grouping data according to site type and urban area, counting the number of failures in each time period of each grouped data, and counting the number of failures that occur in each 10-minute time period of a day.

[0080] Specifically, necessary sensors and data acquisition equipment are installed on the actual operating electric vehicle charging piles to collect the fault data of the charging piles in real time, including temperature sensors, current sensors, voltage sensors, humidity sensors, etc., and data acquisition cards or modules are installed to convert the analog signals of the sensors into digital signals and transmit them to the data server. A fault recording system is set up in the data server to record the time, location, cause of the fault (such as over-temperature fault of the charging cabinet, fault of the charging module, etc.), manufacturer and other information of the fault, and save the relevant original data.

[0081] S102, analyzing the pre-processed fault data through data visualization to obtain the distribution, characteristics and fault modes of the fault data in different dimensions.

[0082] Preferably, the distribution and characteristics of the fault data are displayed through data visualization of charts to obtain analysis results;

[0083] Classify fault data according to site type, and count the frequency ratio and cumulative time of different fault types;

[0084] Classify the fault data by region, and count the frequency percentage and cumulative time of different fault types;

[0085] The charging pile failure data is classified according to site classification and region classification, and the time distribution of charging pile failures is counted.

[0086] Specifically, the fault data is classified by site type, and the frequency ratio and cumulative time of different fault types are counted. In this embodiment, Figure 2 The figure shows the frequency distribution of different fault categories of urban public charging piles. Figure 3 The figure shows the cumulative time distribution of different fault categories of highway charging piles, where the horizontal axis is the category of fault cause, and the vertical axis is the fault frequency ratio and the cumulative time of fault.

[0087] The fault data is classified by region, and the frequency ratio and cumulative time of different fault types are counted. In this embodiment, Figure 4 The figure shows the frequency distribution of different fault types of charging piles in a certain district or county. Figure 5The figure shows the cumulative time distribution of different fault categories of charging piles in a prefecture-level city, where the horizontal axis is the category of fault cause, and the vertical axis is the fault frequency ratio and the cumulative time of faults.

[0088] For charging pile failure data, the time distribution of charging pile failures is counted according to site classification and regional classification. In this embodiment, Figure 6 The figure shows the distribution ratio of the failure time of public charging piles in the city. Figure 7 The figure shows the distribution ratio of charging pile failure time in a prefecture-level city, where the horizontal axis is each time point of the day (unit: h), and the vertical axis is the ratio of the number of charging pile failures at a certain time point to the total number of failures.

[0089] From the data visualization analysis, it can be seen that: from the perspective of site type, emergency stop button action failure and emergency stop button pressed failure frequently occur in public charging piles and internal charging piles of units, and abnormal electricity meter failures frequently occur in highway, bus and logistics site types; from the perspective of urban districts and counties, emergency stop button action failure, emergency stop button pressed failure and BMS communication abnormality and other types of failures frequently occur in various urban districts and counties; from the perspective of the distribution of charging pile failure time, most failures occur between 18:00 and 23:00 in the evening, and random failures at 0:00 in the morning decrease sharply. In general, emergency stop button action failure, BMS communication abnormality, charging module failure, charging interface electronic lock failure, emergency stop button pressed failure and other failure categories appear frequently and show a long-tail distribution, such as Figure 2-5 shown.

[0090] S103, based on the results of data visualization analysis, determine the key features of the data, establish a fault data probability model, and generate electric vehicle charging pile fault data through a dynamic fault data synthesis algorithm.

[0091] It should be noted that S101 collects real fault data to ensure that the existing real fault data contains sufficient information about site type, urban area, fault cause, and fault duration for subsequent data generation.

[0092] Preferably, establishing a fault data probability model includes:

[0093] Step 1: Input parameters, including site type, region, and the number of charging pile fault data to be generated, and obtain the corresponding fault data set;

[0094] Specifically, site type: such as public charging piles, rental charging piles, internal charging piles, high-speed charging piles, bus charging piles, other charging piles and logistics charging piles, etc.; Region: specific region name, such as: Wuhan City, Jingzhou City, Badong County, Jiangling County, etc.; The number of charging pile fault data to be generated: a positive integer, set according to actual needs.

[0095] Step 2: Based on the charging pile fault data set, perform random sampling of data, select the fault data corresponding to the input area, and extract the corresponding fault causes;

[0096] Specifically, in the existing electric vehicle charging pile fault data set, we perform a random sampling process to select fault data corresponding to the input area. Exemplarily, random sampling can be simple random sampling, stratified random sampling, systematic sampling, clustered random sampling, weighted random sampling, etc.; among them, simple random sampling uses a random number generator to assign a random number to each data point, and then selects several data points with the lowest random number; stratified random sampling, stratifies the data according to factors such as fault type, time, and location, and then randomly extracts a fixed number or proportion of samples from each layer; systematic sampling, determines the sampling interval (such as the total number divided by the number of samples), and extracts a sample every fixed number from the random starting point; clustered random sampling, divides the charging piles into groups according to geographical location or other attributes, randomly selects some groups, and analyzes all the charging pile fault data in these groups; weighted random sampling, assigns a weight to each fault data according to the fault frequency or importance, and uses weighted probability for selection when sampling.

[0097] In this embodiment, preferably, data is extracted through stratified random sampling, all fault records belonging to the input area are screened out from the data set, the data is stratified according to factors such as fault type, time, and location, and then a fixed number or proportion of samples are randomly selected from each layer to ensure that these fault data can fully reflect the charging pile fault conditions in the input area, and then 10 electric vehicle charging pile fault data are randomly selected to extract their corresponding fault causes.

[0098] It should be noted that this method ensures that the generated data are geographically consistent with the input regions, thereby improving the accuracy and reliability of the data.

[0099] Step 3: Establish a fault data probability model for the fault data of site type and region, and generate site type and region-specific fault data;

[0100] Step 4: By establishing a probability model, a site type directional dynamic fault data and a regional directional fault data are generated respectively, and multi-dimensional data fusion is performed. The start time and duration of the two data are selected, and the arithmetic mean is calculated respectively. The mean value is used as the value of the corresponding field of the newly generated fault data, which is expressed as:

[0101]

[0102] Among them, T1 and T2 are the starting time values ​​of site type directional dynamic fault data and regional directional fault data, respectively; D1 and D2 are the duration values ​​of site type directional dynamic fault data and regional directional fault data, respectively; T and D are the starting time and duration values ​​of a newly generated fault data;

[0103] It should be noted that the site type directional dynamic fault data and the region directional fault data contain an in-depth understanding of the failure modes of charging piles with different site types, different fault causes and regional distribution characteristics.

[0104] Step 5: According to the total amount of charging pile fault data generated by the user, perform a loop iteration until the user's needs are met. Each iteration repeats the steps of probability model construction and multi-dimensional data fusion, and records the new fault data records that meet the requirements;

[0105] It should be noted that this process enters the cyclic iteration stage and follows the demand clearly stated in the input parameters - that is, the total number of charging pile fault data expected to be generated, that is, the total amount of charging pile fault data that users need to generate. Among them, the total amount generated is generally not more than 2-5 times the existing data, but it needs to be adjusted according to actual needs; the total amount of existing data is 12,180 fault records (the amount of data used for data analysis). Each iteration repeats the above steps two to five, and continuously accumulates new fault data records that meet the standards until the predetermined data set size is reached, that is, the total amount of charging pile fault data that needs to be generated input by the user. This strategy not only ensures the scalability of the data set, but also enhances the diversity and authenticity of the data set through continuous iterative optimization, laying a solid foundation for subsequent fault prediction model training and performance evaluation. The overall process of data generation is as follows: Figure 8 shown.

[0106] Preferably, the specific steps of step three are as follows:

[0107] According to the site type or region and the fault cause of the randomly selected data, set the query conditions for retrieval to obtain data that matches the specified site type or specified region and fault cause;

[0108] Compare the amount of data obtained with the data volume threshold. If the amount of data obtained is less than the data volume threshold, expand the search range until the amount of data is greater than the data volume threshold. If it is greater than the data volume threshold, build a probability model and calculate the frequency of each fault cause. The calculated frequency value is used as the probability model, which is expressed as:

[0109] f i =n i / N

[0110] Among them, n irepresents the frequency of occurrence of the i-th fault cause, N represents the frequency of all fault causes, and f i represents the frequency of occurrence of the i-th fault cause;

[0111] The probability model is used to randomly generate fault data, and the fault cause of the next generated fault data is determined according to the probability of each fault cause.

[0112] If new data is added to the database, the frequency is recalculated and the probability model is updated.

[0113] Specifically, firstly, site type-oriented dynamic fault data generation and probability modeling are performed, including:

[0114] Parameter-driven preliminary search: Based on the site type in the input parameters and the fault causes of the 10 data extracted by stratified random sampling, a query condition is formed, and then all existing data is searched. For example, if the site type is "public charging pile", and the fault cause list includes "charging module failure" and "BMS communication abnormality", the query condition can be "site type = public charging pile AND (fault cause = charging module failure OR fault cause = BMS communication abnormality)", the purpose of this step is to find data that matches the specified site type and fault cause;

[0115] Data volume threshold judgment and processing: If the data entries in the retrieval result exceed the data volume threshold, where the data volume threshold is preferably 50, indicating that the sample size is sufficient, the frequency distribution of the fault cause will be used as the basis to build a probability model and calculate the frequency of occurrence of the fault cause; the fault cause frequency f will be used to generate new fault instances that meet the specific site type and fault cause to ensure the authenticity and representativeness of the generated data. In this way, the generated data will reflect the fault mode of the site and fault cause of this type; if the number of retrieved data is less than 50, the search scope needs to be expanded. Specifically, the site type is no longer limited, but data of all site types are retrieved based on the fault causes of the 10 data extracted in the previous step, aiming to accumulate richer data resources to enhance the reliability of the analysis.

[0116] Iterative retrieval and probability generation mechanism: If the retrieved data is greater than 50 at this time, the initially set probability generation rule is applied (that is, the frequency distribution of the fault cause is used as the probability to generate new fault data) to create new fault data points that reflect the target site type and the fault cause; conversely, if the amount of data is still less than the threshold of 50 even after extended retrieval, a more generalized strategy is adopted.

[0117] It should be noted that in this stage, a global probability model will be constructed based on the comprehensive frequency distribution of all site types and fault causes in the entire database. This model will be used to generate new fault instances to ensure that effective and balanced fault simulation can be achieved under limited data resources and to maintain the universal applicability and diversity of the model output.

[0118] Secondly, regional directional fault data generation and probability modeling are carried out, including:

[0119] Parameter-driven preliminary search: All existing data is searched based on the region in the input parameters and the failure causes of the 10 data extracted in the previous step. The purpose of this step is to find data that matches the specified region and failure cause.

[0120] Data volume threshold judgment and processing: If the search result data entries exceed 50, it indicates that the sample size is sufficient. At this time, the frequency distribution of the fault cause will be used as the basis to build a probability model, calculate the frequency of the fault cause, and use the calculated frequency value as the probability model. This model will be used to generate new fault instances that conform to specific regions and fault causes to ensure the authenticity and representativeness of the generated data. In this way, the generated data will reflect the fault mode of the region and the fault cause. If the number of retrieved data is less than 50, the search scope needs to be expanded. Specifically, the region is no longer limited, but the data of all regions are retrieved based on the fault causes of the 10 data extracted in the previous step, aiming to accumulate richer data resources to enhance the reliability of analysis.

[0121] Iterative retrieval and probability generation mechanism: If the retrieved data is greater than 50 at this time, the initially set probability generation rule is applied (that is, the frequency distribution of the fault cause is used as the probability to generate new fault data) to create new fault data points that reflect the region and fault cause; conversely, if the amount of data is still less than the threshold of 50 even after extended retrieval, a more generalized strategy is adopted.

[0122] It should be noted that at this stage, a global probability model will be constructed based on the comprehensive frequency distribution of all regions and fault causes in the entire database, and this model will be used to generate new fault instances to ensure that effective and balanced fault simulation can be achieved under limited data resources and to maintain the universal applicability and diversity of the model output.

[0123] The specific contents of the global probability model are as follows: collect and organize data on all site types and fault causes; calculate the frequency of occurrence of each fault cause in the entire database; use the probability model to calculate the frequency of each fault cause, and build a probability model based on the frequency of each fault cause, which means that each fault cause has a probability corresponding to its frequency of occurrence. This probability model can be a simple array or dictionary, where the key is the fault cause and the value is the corresponding probability; use this probability model to randomly generate fault data to ensure that the generated data follows the frequency distribution of fault causes in the entire database; the random number generator can be used to determine the fault cause of the next generated fault data based on the probability of each fault cause; if the generated data set is not large enough, or the diversity and quality of the data need to be further improved, the above process can be iterated until the predetermined data set size is met; if new data is added to the database, the frequency should be recalculated and the probability model should be updated to reflect the latest data distribution.

[0124] S104, storing the generated fault data, and performing visualization and quality assessment.

[0125] Preferably, the quality assessment includes:

[0126] Evaluate the quality of the generated data to determine whether there are missing values ​​in the data set, whether the range of values ​​is reasonable, and whether the records in the data set are consistent in format and units;

[0127] Compare the similarity between the generated data and the real data through statistical tests to confirm whether the distribution of the generated data is consistent with the distribution of the actual data set;

[0128] Compare the similarity between the variables in the generated data and the real data set to determine whether the difference is too large.

[0129] Among them, the similarity between data is evaluated by KL divergence, which is expressed as:

[0130]

[0131] Among them, p(x) represents the actual fault data of electric vehicle charging piles in a certain area, and q(x) represents the probability distribution of generated fault data. If the two distributions are closer, the KL divergence is smaller, and if the distributions are farther apart, the KL divergence is larger.

[0132] Specifically, select a suitable data format (such as CSV, JSON, etc.) and storage method (such as database, file system, etc.), write the generated fault data to disk, and evaluate the quality of the generated data set. This evaluation process requires knowledge of statistical analysis and visualization technology. Among them, statistical tests can be such as t-tests, chi-square tests, etc., to confirm whether the distribution of the generated data is consistent with the distribution of the actual data set; data diversity can also be verified to evaluate whether the data set covers all possible changes, including edge cases; among them, KL divergence is used to evaluate (Kullback-Leibler divergence), also known as relative entropy or relative information, which is an indicator used to measure the similarity or difference between two probability distributions. It can help us measure the information loss from one probability distribution (usually an estimated distribution) to another probability distribution (usually a true distribution). The core idea is that given two probability distributions P and Q, P is the true distribution and Q is the estimated distribution. KL divergence represents the minimum information loss from Q to P.

[0133] It should be noted that the present invention provides a method for generating fault data of electric vehicle charging piles, namely, a dynamic fault data synthesis algorithm (Dynamic Fault Data Synthesis Algorithm) referred to as DyFDS algorithm, which aims to solve the problem of missing fault data of electric vehicle charging piles. With the popularization of electric vehicles and the construction of charging infrastructure, the failure rate of charging piles has gradually increased, which has brought challenges to the safe operation of the charging system. Therefore, generating real and effective fault data of electric vehicle charging piles is of great significance to improving the accuracy and reliability of the fault prediction model of electric vehicle charging piles.

[0134] In the present invention, by deploying a series of charging pile sensors, these sensors will monitor the operating status of the electric vehicle charging pile in real time and collect the operating data parameters of the electric vehicle charging pile. It is possible to obtain data on the fault conditions of the charging pile under different environmental conditions and operating modes. These data are not only authentic, but also can reflect various fault conditions that the charging pile may encounter in actual operation.

[0135] Next, the present invention adopts a method based on data visualization analysis and statistics. By statistics and analysis of the existing electric vehicle charging pile fault data, feature extraction and classification of the existing electric vehicle charging pile fault data are performed, and fault type probability models of different dimensions are established. Then, based on the probability models of different dimensions, high-quality and diverse electric vehicle fault data are generated.

[0136] The advantage of the present invention is that by generating real and effective charging pile fault data and combining data visualization analysis and statistics for data enhancement, a large amount of high-quality fault data can be generated. These data can not only be used to train and optimize fault prediction models to improve prediction accuracy and reliability, but also provide strong data support for the operation and maintenance of charging piles. The method of the present invention can help charging pile managers better understand the fault characteristics and laws of charging piles, formulate targeted maintenance strategies, and improve the safety and reliability of charging systems.

[0137] The above is a schematic scheme of a charging pile fault data generation method of this embodiment. It should be noted that the technical scheme of the charging pile fault data generation system and the technical scheme of the above-mentioned charging pile fault data generation method belong to the same concept. For details not described in detail in the technical scheme of the charging pile fault data generation system in this embodiment, please refer to the description of the technical scheme of the above-mentioned charging pile fault data generation method.

[0138] In this embodiment, the charging pile fault data generation system includes:

[0139] The acquisition module is used to collect and record the fault data of the charging pile and pre-process the collected data;

[0140] The analysis module is used to analyze the pre-processed fault data through data visualization to obtain the distribution, characteristics and fault modes of the fault data in different dimensions;

[0141] A building module is used to determine the key features of the data based on the results of data visualization analysis, establish a fault data probability model, and generate electric vehicle charging pile fault data through a dynamic fault data synthesis algorithm;

[0142] The output module is used to store the generated fault data and perform visualization and quality assessment.

[0143] This embodiment further provides an electronic device, which is applicable to the case where charging pile fault data is generated, and includes:

[0144] Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the charging pile fault data generation method proposed in the above embodiment.

[0145] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for generating charging pile fault data as proposed in the above embodiment is implemented.

[0146] The storage medium proposed in this embodiment and the method for generating charging pile fault data proposed in the above embodiment belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0147] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0148] Example 2

[0149] Reference Figure 9-11 , which is an embodiment of the present invention, provides a charging pile fault data generation method. In order to verify its beneficial effects, the comparison results of two schemes are provided, and scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0150] The method of the present invention was implemented and verified by a simulation experiment, in which the operating environment of the experiment was a Windows 11 operating system and a Python programming language environment.

[0151] Specifically, Figure 8 It is a proportional distribution diagram of various fault causes in charging pile faults in a certain area, which clearly shows the relationship between the fault data generated by the DyFDS algorithm of the present invention and the actual fault data. The yellow bar graph represents the data generated by the DyFDS algorithm, while the blue bar graph corresponds to the actual collected real data. By observing and analyzing the heights and relative positions of these bar graphs, we can draw the following conclusions:

[0152] ① Overall trend: Overall, the heights of the two-color bar graphs are basically the same, which shows that there is a high similarity between the data generated by the DyFDS algorithm and the actual data, thus verifying the effectiveness and accuracy of the algorithm; ② Specific differences: Although the overall trend is similar, under certain specific fault causes, the yellow bar graph may be slightly higher or lower than the blue bar graph. This slight deviation may be due to the way the algorithm handles certain complex situations, but it does not affect its overall performance; ③ Fault distribution: By comparing the bar graphs corresponding to each fault cause, it can be found that some fault causes have a higher probability of occurrence, while others have a lower probability. This helps us understand the main sources of charging pile failures and take targeted preventive measures, further confirming the superior performance of the algorithm in generating charging pile fault data.

[0153] Fig. 9 This is a distribution diagram of the start time of charging pile failures in a certain district or county, which clearly shows the distribution of regional charging pile failures at different times. The yellow curve represents the data generated by the DyFDS algorithm of the present invention, while the blue curve corresponds to the actual data collected. By observing and analyzing the heights and relative positions of these bar graphs, we can draw the following conclusions:

[0154] ① Time distribution: The horizontal axis covers the time period from 0 to 24 hours, reflecting the occurrence of faults around the clock. It can be seen from the figure that the probability of faults increases significantly in certain time periods, while it is relatively stable in other time periods. This provides us with important information about the high incidence period of charging pile failures, which is conducive to optimizing maintenance strategies; ② Data fit: Although the two curves are not completely overlapping, their overall trends are very close, especially during the peak period of failures. This shows that the DyFDS algorithm can accurately capture the key features of the fault time distribution, and the generated data has a high correlation with the actual data; ③ Peak difference: At some peak moments, the yellow curve is slightly higher than the blue curve, while in some trough periods, the opposite is true. This slight deviation may be due to the algorithm's handling of extreme cases, but it does not significantly reduce its overall performance.

[0155] To quantify the performance of the DyFDS algorithm of the present invention, we use KL divergence to measure the difference between two probability distributions. Specifically, we compare the existence time distribution ratio of the electric vehicle charging pile failure data generated by the DyFDS algorithm in a specific county (see Fig.10 The yellow curve in Figure 2) is compared with the distribution ratio of the existence time of the actual electric vehicle charging pile failures recorded in the same area (see Fig.10The blue curve in the figure), in this process, we define the time distribution of the real fault data as the probability distribution P, and the time distribution of the fault data generated by the DyFDS algorithm is marked as Q; the calculated KL divergence value is 0.01619612552911, which reflects the slight deviation in the time distribution between the generated data and the real data and the high correlation between the generated data and the real data, thus verifying the remarkable achievements of the DyFDS algorithm in high-fidelity fault data synthesis and its good predictive performance, which helps to identify potential problems in advance and take preventive measures.

[0156] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for generating charging pile fault data, characterized in that: include: Collect and record the fault data of charging piles, and pre-process the collected data; Analyze the preprocessed fault data through data visualization to obtain the distribution, characteristics and fault modes of the fault data in different dimensions; Based on the results of the data visualization analysis, key features of the data are determined, a fault data probability model is established, and electric vehicle charging pile fault data is generated through a dynamic fault data synthesis algorithm; The generated fault data is stored, visualized and quality evaluated.

2. The charging pile fault data generation method according to claim 1, characterized in that: Collecting fault data and preprocessing it include: Monitor the working status and environmental conditions of the charging pile in real time, convert the analog signal of the sensor into a digital signal, and record fault data; The fault data includes the time, location, cause and manufacturer of the fault, and saves the relevant original data; Preprocessing the collected data, wherein the preprocessing includes data cleaning and data aggregation; The data cleaning includes removing duplicate, erroneous, incomplete and outlier data; The data aggregation includes grouping the data according to site type and urban area, counting the number of failures in each time period of each grouped data, and counting the number of failures occurring in each 10-minute time period in a day.

3. The charging pile fault data generating method according to claim 1 or 2, characterized in that: The data visualization analysis includes: Through data visualization of charts, the distribution and characteristics of fault data are displayed to obtain analysis results; Classify the fault data according to site type, and count the frequency ratio and cumulative time of different fault types; Classify the fault data according to regions, and count the frequency ratios and cumulative time of different fault types; The charging pile failure data is classified according to site classification and region classification, and the time distribution of charging pile failures is counted.

4. The charging pile fault data generation method according to claim 3, characterized in that: The generating of electric vehicle charging pile fault data comprises: Input parameters, including site type, region, and the number of charging pile fault data to be generated, to obtain the corresponding fault data set; According to the charging pile fault data set, random sampling of data is performed to select fault data corresponding to the input area, and the corresponding fault causes are extracted; Establish a fault data probability model for the fault data of site types and regions, and generate site type and regional directional fault data; By establishing the probability model, a site type directional dynamic fault data and a regional directional fault data are generated respectively, multi-dimensional data fusion is performed, the start time and duration of the two data are selected, the arithmetic mean is calculated respectively, and the mean value is used as the value of the corresponding field of the newly generated fault data; According to the total amount of charging pile fault data generated by the user, a loop iteration is performed until the user's needs are met. Each iteration repeats the steps of probability model construction and multi-dimensional data fusion to record the new fault data records that meet the requirements.

5. The charging pile fault data generation method according to claim 4, characterized in that: Establishing a probability model for fault data includes: According to the site type or region and the fault cause of the randomly selected data, set the query conditions for retrieval to obtain data that matches the specified site type or specified region and fault cause; The amount of data obtained is compared with the data volume threshold. If the amount of data obtained is less than the data volume threshold, the search range is expanded until the amount of data is greater than the data volume threshold. If it is greater than the data volume threshold, a probability model is constructed to calculate the frequency of each fault cause. The calculated frequency value is used as the probability model, which is expressed as: f i =n i / N Among them, n i represents the frequency of occurrence of the i-th fault cause, N represents the frequency of all fault causes, and f i represents the frequency of occurrence of the i-th fault cause; Randomly generate fault data through the probability module, and determine the fault cause of the next generated fault data according to the probability of each fault cause; If new data is added to the database, the frequency is recalculated and the probability model is updated.

6. The charging pile fault data generation method according to claim 5, characterized in that: The quality assessment includes: Evaluate the quality of the generated data to determine whether there are missing values ​​in the data set, whether the range of values ​​is reasonable, and whether the records in the data set are consistent in format and units; Compare the similarity between the generated data and the real data through statistical tests to confirm whether the distribution of the generated data is consistent with the distribution of the actual data set; Compare the similarity between the variables in the generated data and the real data set to determine whether the difference is too large.

7. The charging pile fault data generation method according to claim 6, characterized in that: The similarity calculation includes: The similarity between data is evaluated by KL divergence, which is expressed as: Among them, p(x) represents the actual failure data of electric vehicle charging piles in a certain area, and q(x) represents the probability distribution of generated failure data.

8. A charging pile fault data generation system, characterized in that: include: The acquisition module is used to collect and record the fault data of the charging pile and pre-process the collected data; The analysis module is used to analyze the pre-processed fault data through data visualization to obtain the distribution, characteristics and fault modes of the fault data in different dimensions; A construction module is used to determine the key features of the data based on the results of the data visualization analysis, establish a fault data probability model, and generate electric vehicle charging pile fault data through a dynamic fault data synthesis algorithm; The output module is used to store the generated fault data and perform visualization and quality assessment.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the charging pile fault data generation method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: It stores computer executable instructions, which, when executed by a processor, implement the steps of the charging pile fault data generating method described in any one of claims 1 to 7.

Images