Power supply fault identification method and device

Abstract

The embodiment of the invention discloses a power supply fault identification method and device. The method comprises the steps of obtaining a charging message reported by a target split type charging pile terminal, determining charging order information according to a charging order identifier, obtaining charging gun charging data according to the charging message and the charging order information, aggregating the charging gun charging data to obtain charging pile aggregated data, and sending the aggregated data to the target split type charging pile terminal. Sampling the aggregated data of the plurality of charging piles in the sampling time period to obtain feature data of the plurality of charging piles, and inputting the feature data of the plurality of charging piles in a predetermined number into a power supply fault recognition model to obtain corresponding power supply fault recognition results, if the power supply fault recognition result is that the ratio of the number of the power supply faults to the preset number is larger than the preset ratio, the charging pile is marked as the power supply fault, and the method for detecting the power supply faults through the charging message does not have requirements for parts of the charging pile and the power supply self-checking capacity and is suitable for charging piles of various models.

Description

Technical Field

[0001] This invention relates to the field of new energy technology, and more specifically, to a power supply fault identification method and apparatus. Background Technology

[0002] With the increasing popularity of electric vehicles, the demand for charging infrastructure is growing daily. Among them, split-type charging piles have been welcomed by the market due to their flexible installation methods and scalability. The design of split-type charging piles separates the power supply unit from the charging terminal. While this layout improves installation flexibility, it also increases the difficulty of identifying power supply faults. In addition, due to the complex working environment, low standardization, and diverse fault types, identifying power supply faults in split-type charging piles is difficult.

[0003] Current common methods for power supply fault identification include: installing sensors inside the charging pile's power supply unit, which increases the cost of the power supply unit and the additional sensor data can put significant pressure on the network bandwidth when the charging pile communicates with the platform server; or manually observing charging parameters to determine if a power supply unit fault has occurred, which involves a large amount of manual comparison and analysis, resulting in high labor costs, and the complex structure of the split-type charging pile makes it prone to misjudgment in abnormal scenarios. Therefore, to address the difficulty of fault identification in split-type charging pile power supply units, there is an urgent need to develop more intelligent and efficient fault detection technologies to improve the reliability of charging facilities and user experience. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a power supply fault identification method and apparatus to develop more intelligent and efficient power supply fault detection technology, realize automatic identification of faults in the power supply devices of split-type charging piles, thereby improving the reliability of charging facilities and user experience.

[0005] Firstly, a power supply fault identification method is provided, the method comprising:

[0006] The charging message is obtained from multiple terminals of the target split charging pile. The charging message is obtained from multiple terminals of the target split charging pile. The charging message includes at least a charging status message. The charging status message is used to record the charging status of the charging gun corresponding to the terminal. The charging status message includes a charging order identifier.

[0007] Based on the charging order identifier, determine the corresponding charging order information;

[0008] The charging gun charging data is obtained by filling the preset field template according to the charging message and the charging order information, and the charging gun charging data includes the reporting time of the charging status message.

[0009] According to the preset aggregation period and the reporting time of each charging status message, the charging data of multiple charging guns are aggregated to obtain the corresponding charging pile aggregated data.

[0010] Multiple charging pile aggregate data within a sampling time period are sampled to obtain multiple charging pile feature data, wherein the sampling time period is determined according to a preset sampling period;

[0011] A predetermined number of the charging pile feature data are input into a pre-trained power fault identification model to obtain the corresponding power fault identification results, which are either power faults or power normal.

[0012] If the ratio of the number of charging pile feature data whose power failure identification result is a power failure to the predetermined number is greater than a predetermined ratio, the target split-type charging pile is marked as having a power failure.

[0013] Secondly, a power fault identification device is provided, the device comprising:

[0014] The acquisition module is used to acquire charging messages reported by multiple terminals of the target split-type charging pile. The charging message includes at least a charging status message, which is used to record the charging status of the charging gun corresponding to the terminal. The charging status message includes a charging order identifier.

[0015] The determining module is used to determine the corresponding charging order information based on the charging order identifier;

[0016] The filling module is used to fill a preset field template according to the charging message and the charging order information to obtain the corresponding charging gun charging data, wherein the charging gun charging data includes the reporting time of the charging status message;

[0017] The aggregation module is used to aggregate the charging data of multiple charging guns according to a preset aggregation period and the reporting time of each charging status message to obtain the corresponding charging pile aggregated data.

[0018] The sampling module is used to sample aggregated data of multiple charging piles within a sampling time period to obtain multiple charging pile feature data. The sampling time period is determined according to a preset sampling period.

[0019] The identification module is used to input a predetermined number of the charging pile feature data into a pre-trained power fault identification model to obtain the corresponding power fault identification results, wherein the power fault identification results are power fault or power normal.

[0020] The marking module is used to mark the target split-type charging pile as having a power failure in response to a predetermined ratio of the number of characteristic data of charging piles with power failure identification results being greater than a predetermined ratio.

[0021] Thirdly, an electronic device is provided, including a memory and a processor, the memory being used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method as described in the first aspect above.

[0022] Fourthly, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when executed by a processor, the computer program implements the method described in the first aspect.

[0023] Fifthly, a computer program product is provided, comprising a computer program / instructions that, when executed by a processor, implement the method described in the first aspect above.

[0024] The method of this invention includes acquiring charging messages reported by multiple terminals of a target split-type charging pile, wherein each charging message includes at least a charging status message, which records the charging status of the charging gun corresponding to the terminal and includes a charging order identifier. Based on the charging order identifier, corresponding charging order information is determined. A preset field template is filled based on the charging messages and the charging order information to obtain corresponding charging gun charging data. The charging gun charging data includes the reporting time of the charging status message. Multiple charging gun charging data are aggregated according to a preset aggregation period and the reporting time of each charging status message to obtain corresponding charging pile aggregated data. Multiple charging pile aggregated data within a sampling time period are sampled to obtain multiple charging pile feature data. The sampling time period is determined according to a preset sampling period. A predetermined number of the charging pile feature data are input into a pre-trained power fault identification model to obtain corresponding power fault identification results, wherein the power fault identification result is either a power fault or a power normal result. In response to a ratio greater than a predetermined ratio between the number of charging pile feature data with a power fault identification result greater than a predetermined ratio, the target split-type charging pile is marked as having a power fault. The above method uses charging messages reported by the split-type charging pile terminal to detect power failures. It does not require the charging pile's components and power supply self-testing capabilities, can be adapted to various models of charging piles, has higher identification efficiency, and saves manpower and resources. Attached Figure Description

[0025] The above and other objects, features and advantages of the present invention will become clearer from the following description of embodiments of the invention with reference to the accompanying drawings, in which:

[0026] Figure 1This is a schematic diagram of a power fault identification system according to an embodiment of the present invention;

[0027] Figure 2 This is a schematic diagram of the power supply fault identification model training process according to an embodiment of the present invention;

[0028] Figure 3 This is a flowchart of a method for determining charging data for a charging gun according to an embodiment of the present invention;

[0029] Figure 4 This is a flowchart of the charging pile aggregation data determination method according to an embodiment of the present invention;

[0030] Figure 5 This is a schematic diagram of data aggregation according to an embodiment of the present invention;

[0031] Figure 6 This is a flowchart of the charging pile feature data determination method according to an embodiment of the present invention;

[0032] Figure 7 This is a schematic diagram of feature sampling according to an embodiment of the present invention;

[0033] Figure 8 This is a schematic diagram of the power supply fault identification model training method according to an embodiment of the present invention;

[0034] Figure 9 This is a schematic diagram of a power fault identification device according to an embodiment of the present invention;

[0035] Figure 10 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0036] The present application is described below based on embodiments, but it is not limited to these embodiments. In the detailed description of the present application below, certain specific details are described in detail. Those skilled in the art can fully understand the present application without these details. To avoid obscuring the substance of the present application, well-known methods, processes, flows, elements, and circuits are not described in detail.

[0037] Furthermore, those skilled in the art should understand that the accompanying drawings provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0038] Unless the context explicitly requires it, words such as "including" or "contains" throughout the application should be interpreted as including rather than exclusive or exhaustive; that is, meaning "including but not limited to".

[0039] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0040] The solutions described in this specification and embodiments, if involving the processing of personal information, will be processed only under the premise of having a legal basis (such as obtaining the consent of the personal information subject, or being necessary for the performance of a contract), and will only be processed within the scope stipulated or agreed upon. A user's refusal to process personal information beyond what is necessary for basic functions will not affect the user's use of basic functions.

[0041] Figure 1 This is a schematic diagram of a power fault identification system according to an embodiment of the present invention. Figure 1 As shown, the power fault identification system includes a target split-type charging pile 11 and a server 13.

[0042] In this embodiment, the target split-type charging pile 11 is a charging device that provides charging services for electric vehicles. Its key feature is that the main components of the charging pile are designed as independent modules. These modules can be installed in different locations and then connected via cables or wirelessly to jointly complete the charging task. A set of target split-type charging piles 11 consists of a main unit cabinet and several terminals 12. The charging guns are attached to and managed by each independent terminal 12, and each terminal 12 has its own communication unit to communicate with the server 13. The power modules are centrally installed in the main unit cabinet. The target split-type charging pile 11 flexibly allocates power modules to supply the charging guns on different terminals 12 through the control unit. The split-type charging pile allows for more flexible power expansion and distribution, adapting to changing charging needs. The server 13 is a general-purpose data processing device that provides computing or application services to the charging platform clients. It can be a single computer, a cluster of multiple computers, or a cloud server that can flexibly adjust computing resources through cloud technology.

[0043] In one possible implementation, the power fault identification system further includes a user terminal 14. The user terminal 14 is a terminal device held by a user, operator, or maintenance provider of the target split-type charging pile 11. That is, the user terminal 14 can be at least one of a terminal held by a charging user, a charging merchant, or a charging pile company using the target split-type charging pile 11. The user terminal 14 is bound to the target split-type charging pile 11. When the user terminal 14 is a terminal held by a charging user, the user terminal 14 and the target split-type charging pile 11 are temporarily bound based on the charging order during the time period when the charging user uses any terminal 12 of the target split-type charging pile 11 for charging. Specifically, the user terminal 14 is a general-purpose terminal capable of running charging platform client applications or mini-programs, such as a mobile phone, computer, or tablet computer. The user terminal 14 is used to receive prompt information to alert the user holding the user terminal 14 to the power fault identification result.

[0044] In this embodiment, the server 13 interacts with the terminals 12 of multiple target split-type charging piles 11 via the network to obtain the charging messages generated by the multiple terminals 12 corresponding to each target split-type charging pile 11 during the charging process, and realizes the identification of charging pile power supply faults based on the charging messages.

[0045] Specifically, after establishing a network connection with server 13, terminal 12 will upload various types of charging messages to server 13. These include charging check-in messages, charging status messages, charging alarm messages, charging order messages, charging user authentication messages, charging configuration request messages, firmware update request messages, command response messages, heartbeat messages, and charging log messages, etc. Different types of charging messages contain different information and are reported at different times.

[0046] Among them, the charging check-in message is a message reported by the terminal 12 when establishing a network connection with the server 13. It is used to report the equipment information of the target split charging pile 11. The equipment information may include the number of power supply devices, the power supply device power, the output voltage range, the charging pile model, the charging pile manufacturer, the charging pile identification, the terminal identification attached to the current charging pile, the charging gun identification corresponding to each terminal, the maximum output current that the charging gun line can withstand, the maximum output power of the charging gun, and the average output power of the charging pile to each gun.

[0047] The charging status message is a message sent by terminal 12 to server 13 at fixed time intervals (i.e., the first time period) after confirming a successful network connection. The charging status message is reported on a per-charging-gun basis; that is, terminal 12 records the charging status of each charging gun on its terminal, generates a corresponding charging status message for each gun, and then reports the charging status message for each gun to server 13. The reporting interval (i.e., the first time period) of the charging status message for each charging gun is the same, but the reporting time points are not necessarily the same. For example, terminal 12 attempts to establish a network connection with server 13. After the network connection is successful, starting from the same time point t, it reports the charging status messages of each charging gun at the same time interval T (T>0). At time point t+a (a>0), one of the charging guns malfunctions and stops working. Then terminal 12 also stops reporting the charging status of that charging gun. At time point t+a+b (b>0, and a+b≠n*T, where n is a positive integer), it resumes normal operation. Then that charging gun restarts reporting the charging status message from time point t+a+b at the time interval T. At this time, the reporting time point t+a+b+x*T (x is a positive integer) of that charging gun is different from the reporting time point t+y*T (y is a positive integer) of other charging guns. As the target split-type charging pile 11 operates continuously for longer periods, the more likely the charging status messages for the multiple charging guns of the terminal 12 will be reported at different times. The charging status messages include information such as the charging order identifier, output current, output voltage, desired current and voltage of the vehicle being charged, and the charging gun identifier.

[0048] The charging alarm message is a message reported by terminal 12 when it detects any abnormality or malfunction in the main cabinet of the target split charging pile 11 or by the terminal itself. It reports fault information determined by sensors built into the target split charging pile 11, including fault identifier, fault occurrence time, and fault description. The fault identifier (also called a fault code) indicates the fault type. In addition, the charging alarm message also includes a charging order identifier to associate with the corresponding charging order.

[0049] The charging order message is a message reported by terminal 12 to server 13 when providing charging services. It is used to record charging order-related information for billing and statistical analysis after the order is completed. The charging order message includes information such as the start time, end time, charging amount, and charging cost of the charging order.

[0050] The charging log message is a message reported by terminal 12 in response to a log retrieval command sent by server 13. During operation, the charging terminal 12 stores various data generated during the process on its local hard drive to prevent data loss due to network connection failures. Server 13 can retrieve this log by issuing a log retrieval command. The charging log message contains all messages and other data generated by terminal 12 within a certain period.

[0051] The above messages are common in the field of electric vehicle charging and will not be explained in detail here. The charging messages used by this power fault identification system include at least charging status messages, and may also include any one or more of the above types of messages. The more message types used, the more accurate the power fault identification results obtained from the charging messages may be, but the corresponding computational load will also increase. Therefore, the types of charging messages to be used can be determined and adjusted according to the actual situation, such as the amount of available computing resources, the amount of data to be calculated, and the required identification accuracy.

[0052] After the target split-type charging pile 11 is activated, multiple terminals 12 of the target split-type charging pile 11 begin to report corresponding charging messages to the server 13. Then, the server 13 begins to determine whether there is a power failure in the target split-type charging pile 11 based on the charging messages. However, before starting the power failure identification, the server 13 needs to train a power failure identification model.

[0053] Figure 2 This is a schematic diagram illustrating the training process of the power supply fault identification model according to an embodiment of the present invention. Figure 2 As shown, the training process of the power supply fault identification model may include three stages: model building stage, model verification stage, and model grayscale stage.

[0054] The model building phase includes the training sample dataset construction sub-phase, the charging feature data sampling sub-phase, and the power supply fault identification model building sub-phase.

[0055] In the training sample dataset construction phase, historical charging messages and historical power failure information are acquired. Based on the historical fault information, historical charging messages corresponding to the fault information are selected as positive samples, and historical charging messages without faults are selected as negative samples. This constitutes a training sample dataset with a mixture of positive and negative samples and reasonable labeling. It is worth noting that charging messages with a current of 0 should be filtered out when acquiring historical charging messages, because a current of 0 indicates that the charging gun is not in a charging state. Power failure identification cannot be performed when the charging gun is not in a charging state. Therefore, charging messages that are not in a charging state are invalid data. Removing invalid data can greatly reduce the amount of data processing and speed up the processing speed.

[0056] In the charging feature data sampling sub-stage, the training sample data in the training sample dataset is processed to obtain charging pile feature sample data. It is worth noting that charging feature data sampling is required in the model building stage, model validation stage, model gray-scale stage, and subsequent model usage stage, only the data processed differs. Specifically, the data processed in the model building stage is training sample data, the data processed in the model validation stage is validation sample data, the data processed in the model gray-scale stage is gray-scale sample data, and after the model is put into normal use, the data used is data reported by terminal 12 a certain time before the current time, such as data reported the previous day. Although the data collection time and method differ, the information type and data processing process are the same. For ease of description, the following data processing process will no longer use terms such as "historical," "sample," or "validation" to characterize data collection time or method, but it should be understood that the data corresponding to the stage should be used in different stages. For example, the data in the model building stage should be training sample data.

[0057] First, the charging messages are integrated to obtain the charging gun charging data. Since the charging status message is the main characteristic data, the integration is based on the charging pile characteristic data. The following explanation uses charging messages, including charging status messages, charging alarm messages, and charging signature messages, as an example.

[0058] Figure 3 This is a flowchart illustrating the charging data determination method for a charging gun according to an embodiment of the present invention. Figure 3 As shown, the method for determining charging data of the charging gun includes the following steps:

[0059] Step S301: Determine the corresponding charging order information based on the charging order identifier.

[0060] In one possible implementation, the charging message also includes a charging order message. The corresponding charging order message can be matched with the charging order identifier in the charging status message to obtain the corresponding charging order information.

[0061] In one possible implementation, if the charging message does not include a charging order message, then the charging order information can be requested from the charging platform server based on the charging order identifier.

[0062] Step S302: Based on the charging order identifier, determine the charging check-in message and charging alarm message associated with the charging status message.

[0063] Step S303: Fill the preset field template with the charging status message and the charging check-in message, charging alarm message and charging order information associated with the charging status message to obtain the corresponding charging gun charging data.

[0064] For example, the preset field templates are as follows: [Charging status message: charging order identifier of the charging gun, output current, output voltage, expected current and expected voltage of the charging vehicle, and charging gun identifier, reporting time; Charging order information: charging order identifier, charging user identifier, and charging vehicle model; Charging alarm message: fault identifier, fault occurrence time, and fault description information; Charging check-in message: charging pile model, charging pile rated power, charging gun maximum current, and charging gun maximum power].

[0065] After obtaining the charging message and charging order information, fill in the corresponding parameters into the preset field template to generate the corresponding charging gun charging data.

[0066] pass Figure 3 The method shown can determine the charging data of all charging guns on all terminals 12 of the target split-type charging pile 11. Since the power supply of the split-type charging pile is uniformly installed in the main cabinet and dynamically allocated to the charging guns according to charging demand, it is difficult to determine whether the power supply device or power module is faulty based on the charging data of a single charging gun. It is necessary to comprehensively consider the charging data of the entire target split-type charging pile 11 to perform fault detection on the uniformly installed power supply device. Therefore, it is also necessary to aggregate the charging data of multiple charging guns according to a preset aggregation period and the reporting time of each charging status message to obtain the corresponding charging pile aggregated data.

[0067] Figure 4 This is a flowchart of the charging pile aggregation data determination method according to an embodiment of the present invention. Figure 5 This is a schematic diagram of data aggregation according to an embodiment of the present invention. The following is in conjunction with... Figure 5 right Figure 4 The method for determining the aggregated data of the charging piles shown is explained.

[0068] like Figure 4 As shown, the method for determining the aggregated data of charging piles includes the following steps:

[0069] Step S401: Obtain the corresponding aggregation time period according to the preset aggregation cycle.

[0070] Step S402: Determine multiple charging gun charging data points whose reporting time falls within the aggregation time period.

[0071] Step S403: Aggregate the multiple charging gun charging data to obtain the corresponding charging pile aggregated data.

[0072] As mentioned above, the time interval for the terminal 12 to report the charging status information of each charging gun is the same, but the reporting time points may be different. Furthermore, the charging reporting time of multiple charging guns located on different terminals 12 may also be different. Therefore, it is difficult to aggregate data by time point. It is necessary to set a preset aggregation period to aggregate the charging data of all charging guns on the target split charging pile 11 in the same first time period.

[0073] The preset aggregation period can be the same as the first time period, so as to ensure that each preset aggregation period contains charging data of all the charging guns that are running on the target split charging pile 11. The preset aggregation period can be set according to actual needs, such as 30 seconds, 1 minute, 5 minutes, etc.

[0074] In one possible implementation, the preset aggregation period can also be greater than the first time period; for example, the preset aggregation period can be a multiple of the first time period.

[0075] like Figure 5 As shown, charging gun 1 has two charging data points before and after a first time period, falling into aggregation time period 1 and aggregation time period 2 respectively. For example, if the preset aggregation time period is 30 seconds, then aggregation time period 1 is [0 seconds, 30 seconds), and aggregation time period 2 is [30 seconds, 60 seconds). The charging status report time for the first charging data point of charging gun 1 is 5 seconds, and the charging status report time for the second charging data point of charging gun 1 is 35 seconds. All charging data points falling within aggregation time period 1 are aggregated into a single data point according to the aggregation rules to obtain the charging pile aggregate data. The report time of this charging pile aggregate data is set to any time within aggregation time period 1, such as the start or end time of aggregation time period 1. Figure 5 This shows that the reporting time is set as the start time of the aggregated time period.

[0076] The aggregation rules specify the aggregation methods for various data in the charging gun's charging data. For example, the aggregation method for data such as output current, output power, expected current, and expected power in the charging status message is summation; the aggregation method for data such as output voltage and expected voltage is taking the maximum value; the aggregation method for data such as charging pile model, charging pile rated power, charging gun maximum current, and charging gun maximum power in the charging check-in message is multiple-choice, because all charging guns belong to the same charging pile, so the corresponding device information is the same, and any device information from any charging gun's charging data can be selected; the aggregation method for data such as charging order identifier, charging user identifier, and charging vehicle model in the charging order information is merging. Each charging gun participating in data aggregation will have a charging order, so all charging data should be merged, for example, in the form of a JSON object.

[0077] pass Figure 4 The method shown can acquire aggregated charging pile data within multiple aggregated time periods, and then sample the aggregated charging pile data within the sampling time period to obtain multiple charging pile feature data. The sampling time period is determined according to a preset sampling period.

[0078] Figure 6 This is a flowchart of a method for determining the characteristic data of a charging pile according to an embodiment of the present invention. Figure 7 This is a schematic diagram of feature sampling according to an embodiment of the present invention. The following is in conjunction with... Figure 7 right Figure 6 The method for determining the characteristic data of the charging pile shown is explained.

[0079] like Figure 6 As shown, the method for determining the characteristic data of the charging pile includes the following steps:

[0080] Step S601: Determine the corresponding sampling time period according to the preset sampling period.

[0081] The preset sampling period is much larger than the first time period and the aggregation time period. For example, the preset sampling period can be 8 hours, 12 hours, 24 hours, etc.

[0082] Step S602: Determine the aggregated data of multiple charging piles within a preset time period.

[0083] Step S603: Sort the aggregated data of the multiple charging piles according to the time sequence of the reporting time of the charging status message.

[0084] Step S604: Determine the sampling starting point according to the preset sampling rules.

[0085] Step S605: Perform window sliding sampling according to the preset window length and preset sliding step size to obtain multiple charging pile feature data.

[0086] like Figure 7 As shown, all the aggregated data of charging piles within the sampling time period are sorted in chronological order. A charging pile aggregated data is randomly selected as the starting point, and window sliding sampling is performed according to the preset window length and preset sliding step size to obtain multiple charging pile feature data. For example, the preset window length is 6 and the preset sliding step size is 3.

[0087] through Figures 3-7 The method shown can be used to obtain the characteristic data of the charging pile.

[0088] If the charging feature data sampling sub-stage is a sub-stage of the model building stage or the model validation stage, then the charging feature data should also be labeled. If it is positive sample data, the corresponding label for the charging feature data should be power failure; if it is negative sample data, the corresponding label for the charging feature data should be power normal.

[0089] In the power supply fault identification model building sub-stage, the initial version of the power supply fault identification model can be obtained by training the model with the training sample dataset obtained in the charging feature data sampling sub-stage. The power supply fault identification model is a binary classification model, such as the extreme gradient boosting model (XGBoost), the lightweight gradient boosting machine (LightGBM), and the gradient boosting decision tree (GBDT).

[0090] The model validation phase includes the validation sample dataset construction sub-phase, the charging feature data sampling sub-phase, and the power supply fault identification model validation sub-phase.

[0091] The verification sample dataset construction sub-stage is the same as the training sample dataset construction sub-stage, both based on historical charging messages and historical power failure information. The specific construction method is as described above and will not be repeated here.

[0092] The data processing procedure for the charging characteristic data sampling sub-stage is as described above and will not be repeated here.

[0093] In the power fault identification model verification sub-stage, the training samples of the charging pile aggregate data in the training sample dataset are input into the power fault identification model. Then, the output power fault identification result is compared with the pre-labeled label to determine the identification accuracy. If the identification accuracy is higher than the preset accuracy, it directly enters the model grayscale stage. If it is not higher than the preset accuracy, it returns to the model building stage to retrain the model until the identification accuracy is higher than the preset accuracy.

[0094] The grayscale stage of the model includes the grayscale sample dataset construction sub-stage, the charging feature data sampling sub-stage, and the power supply fault identification model rollback sub-stage.

[0095] In the grayscale sample dataset construction sub-stage, charging messages from multiple split-type charging piles in a pre-selected area are obtained. The models and specifications of the split-type charging piles should be diverse.

[0096] The data processing procedure for the charging feature data sampling sub-stage is as described above and will not be repeated here. However, it is worth noting that the model input data in the grayscale stage does not need to be labeled.

[0097] During the rollback phase of the power fault identification model, feature data from multiple charging piles are input into the pre-trained power fault identification model to obtain corresponding power fault identification results. Then, based on the number of power faults identified in the power fault identification results, the labeling result for the corresponding split-type charging pile is determined.

[0098] In one possible implementation, the ratio of the number of power failures identified in the power failure identification results to the number of charging pile feature data input into the model is compared. If the ratio is greater than a predetermined ratio, the split-type charging pile is marked as having a power failure. The predetermined ratio can be 0.5, 0.6, or 0.7, etc.

[0099] In one possible implementation, the ratio of the number of data points identified as power failures to the number of data points identified as power failures is compared. If the ratio is greater than a predetermined ratio, the split-type charging station is marked as having a power failure. The predetermined ratio can be 1, 2, or 3, etc.

[0100] After determining the marking results of the split-type charging piles, compare them with the actual offline power failure situation of the split-type charging piles to determine the model's recognition accuracy, and then determine whether it passes the gray-scale test. If it passes, the power failure recognition model can be put into use. If it fails, return to the model building stage to retrain the model until it passes the gray-scale test.

[0101] through Figure 2 After the power fault identification model training process shown is completed, the server 13 can perform power fault detection and identification on the target split-type charging pile 11 to which the terminal 12 belongs based on the charging message reported by the terminal 12.

[0102] Figure 8 This is a schematic diagram of the power supply fault identification model training method according to an embodiment of the present invention. Figure 8 As shown, the power supply fault identification model training method includes the following steps:

[0103] Step S801: Obtain the charging messages reported by multiple terminals of the target split-type charging pile.

[0104] Step S802: Determine the corresponding charging order information based on the charging order identifier.

[0105] Step S803: Fill the preset field template according to the charging message and the charging order information to obtain the corresponding charging gun charging data. The charging gun charging data includes the reporting time of the charging status message.

[0106] Step S804: According to the preset aggregation period and the reporting time of each charging status message, the charging data of multiple charging guns are aggregated to obtain the corresponding charging pile aggregated data.

[0107] Step S805: Sample the aggregated data of multiple charging piles within the sampling time period to obtain multiple charging pile feature data. The sampling time period is determined according to a preset sampling period.

[0108] Step S806: Input a predetermined number of the charging pile feature data into the pre-trained power fault identification model to obtain the corresponding power fault identification results, wherein the power fault identification results are power fault or power normal.

[0109] Step S807: In response to the fact that the ratio of the number of charging pile feature data with power failure identification result being greater than the predetermined number is greater than the predetermined ratio, the target split charging pile is marked as having a power failure.

[0110] Figure 8 The method shown is the same as the one described above. Figures 2-7 The implementation process of the methods shown is similar, and will not be described in detail here.

[0111] After the server marks the target split-type charging pile as having a power failure, it can generate various prompt messages, which are sent to different user terminals 14 to prompt the corresponding users.

[0112] In one possible implementation, in response to the target split-type charging pile being marked as having a power failure, a corresponding charging user prompt message is generated and sent to the terminal corresponding to the charging user using the target split-type charging pile to prompt the charging user to replace the charging pile.

[0113] Specifically, the charging user notification message does not need to include detailed information about the power failure, such as abnormal current curves or abnormal voltage curves; it only needs to include the power failure determination result. Furthermore, it can also provide reference information for replacing the charging station based on the charging status of other charging stations.

[0114] In one possible implementation, in response to the target split-type charging pile being marked as having a power failure, corresponding charging merchant prompt information and charging pile company prompt information are generated. The charging merchant prompt information includes power failure parameters. The charging merchant prompt information is sent to the terminal corresponding to the charging merchant so that the charging merchant can determine whether to authorize sending the charging pile company prompt information to the charging pile company corresponding to the target split-type charging pile based on the power failure parameters. In response to determining that the terminal corresponding to the charging merchant agrees to the authorization, the charging pile company prompt information is sent to the terminal corresponding to the charging pile company to prompt the charging pile company's maintenance personnel to carry out maintenance.

[0115] Specifically, a notification message is sent to the charging merchant's corresponding terminal, informing them that data from the past period indicates a power failure within the charging unit, requiring confirmation and contact with the charging station operator for on-site repair. Furthermore, the message can include details of the power failure, such as power and current curves from multiple charging guns, indicating which parts of the curves show abnormal charging, the number of charging guns during each abnormal charging period, the output of each charging gun, alarm status, and the rated power of the main unit cabinet, allowing the charging merchant to compare and confirm if a problem exists. After the charging merchant confirms and authorizes the charging station operator to perform power repairs, the notification message is sent to the charging station operator's corresponding terminal to notify the operator's maintenance personnel to expedite repairs.

[0116] The method of this invention includes acquiring charging messages reported by a target split-type charging pile terminal, determining charging order information based on a charging order identifier, obtaining charging gun charging data based on the charging messages and charging order information, aggregating multiple charging gun charging data to obtain charging pile aggregate data, sampling multiple charging pile aggregate data within a sampling time period to obtain multiple charging pile feature data, inputting a predetermined number of charging pile feature data into a power fault identification model to obtain corresponding power fault identification results, wherein the power fault identification results include power faults, and if the ratio of the number of power faults to the predetermined number is greater than a predetermined ratio, the charging pile is marked as having a power fault. This method of using charging messages reported by a split-type charging pile terminal for power fault detection has no requirements on the charging pile's components and power supply self-testing capabilities, can adapt to various models of charging piles, has higher identification efficiency, and saves manpower and resources.

[0117] Figure 9 This is a schematic diagram of a power fault identification device according to an embodiment of the present invention. Figure 9 As shown, the power fault identification device includes:

[0118] The acquisition module 901 is used to acquire charging messages reported by multiple terminals of the target split-type charging pile. The charging message includes at least a charging status message, which is used to record the charging status of the charging gun corresponding to the terminal. The charging status message includes a charging order identifier.

[0119] The determining module 902 is used to determine the corresponding charging order information based on the charging order identifier.

[0120] The filling module 903 is used to fill a preset field template according to the charging message and the charging order information to obtain the corresponding charging gun charging data, wherein the charging gun charging data includes the reporting time of the charging status message.

[0121] The aggregation module 904 is used to aggregate the charging data of multiple charging guns according to a preset aggregation period and the reporting time of each charging status message to obtain the corresponding charging pile aggregated data.

[0122] The sampling module 905 is used to sample aggregated data of multiple charging piles within a sampling time period to obtain multiple charging pile feature data. The sampling time period is determined according to a preset sampling period.

[0123] The identification module 906 is used to input a predetermined number of the charging pile feature data into a pre-trained power fault identification model to obtain the corresponding power fault identification results, wherein the power fault identification results are power fault or power normal.

[0124] The marking module 907 is used to mark the target split-type charging pile as having a power failure in response to the ratio of the number of charging pile feature data whose power failure identification result is a power failure to the predetermined number being greater than a predetermined ratio.

[0125] The apparatus described in this invention is used in a power fault identification method and apparatus disclosed in this embodiment. The method includes acquiring charging messages reported by a target split-type charging pile terminal, determining charging order information based on a charging order identifier, obtaining charging gun charging data based on the charging messages and charging order information, aggregating multiple charging gun charging data to obtain charging pile aggregate data, sampling multiple charging pile aggregate data within a sampling time period to obtain multiple charging pile feature data, inputting a predetermined number of charging pile feature data into a power fault identification model to obtain corresponding power fault identification results, whereby the power fault identification results include power faults. If the ratio of the number of power faults identified in the power fault identification results to the predetermined number is greater than a predetermined ratio, the charging pile is marked as having a power fault. Using charging messages reported by split-type charging pile terminals for power fault detection does not require specific components or power supply self-testing capabilities from the charging pile, is compatible with various models of charging piles, has higher identification efficiency, and saves manpower and resources.

[0126] Figure 10 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Figure 10 As shown, Figure 10The illustrated electronic device is a general address lookup device, comprising a general computer hardware architecture, including at least a processor 1001 and a memory 1002. The processor 1001 and memory 1002 are connected via a bus 1003. The memory 1002 is adapted to store instructions or programs executable by the processor 1001. The processor 1001 can be a standalone microprocessor or a collection of one or more microprocessors. Thus, the processor 1001 executes the instructions stored in the memory 1002, thereby performing the method flow of the embodiments of the present invention as described above to process data and control other devices. The bus 1003 connects the aforementioned components together, and also connects these components to a display controller 1004, a display device, and an input / output (I / O) device 1005. The input / output (I / O) device 1005 can be a mouse, keyboard, modem, network interface, touch input device, motion-sensing input device, printer, and other devices known in the art. Typically, the input / output device 1005 is connected to the system via an input / output (I / O) controller 1006.

[0127] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus (devices), or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0128] This application is described with reference to flowchart illustrations of methods, apparatus (devices), and computer program products according to embodiments of this application. It should be understood that each step in the flowchart can be implemented by computer program instructions.

[0129] These computer program instructions may be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction means, the implementation process of which is described in the instruction means. Figure 1 The function specified in one or more processes.

[0130] These computer program instructions may also be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, produce instructions for implementing processes. Figure 1 A device for a function specified in one or more processes.

[0131] The method of this invention includes acquiring charging messages reported by the target split-type charging pile terminal, determining charging order information based on the charging order identifier, obtaining charging gun charging data based on the charging messages and charging order information, aggregating multiple charging gun charging data to obtain charging pile aggregate data, sampling multiple charging pile aggregate data within a sampling time period to obtain multiple charging pile feature data, inputting a predetermined number of charging pile feature data into a power fault identification model to obtain corresponding power fault identification results, the power fault identification results including power faults, and if the ratio of the number of power faults to the predetermined number is greater than a predetermined ratio, the charging pile is marked as having a power fault. Using charging messages reported by the split-type charging pile terminal for power fault detection does not require specific components or power supply self-testing capabilities of the charging pile, can adapt to various models of charging piles, has higher identification efficiency, and saves manpower and resources.

[0132] Another embodiment of the present invention relates to a non-volatile storage medium for storing a computer-readable program for use by a computer to execute some or all of the above-described method embodiments.

[0133] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program specifying the relevant hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0134] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A power supply fault identification method, characterized in that, The method includes: The charging messages reported by multiple terminals of the target split-type charging pile are obtained. The charging messages include at least a charging status message, which is used to record the charging status of the charging gun corresponding to the terminal. The charging status message includes a charging order identifier. Based on the charging order identifier, determine the corresponding charging order information; The charging gun charging data is obtained by filling the preset field template according to the charging message and the charging order information, and the charging gun charging data includes the reporting time of the charging status message. According to the preset aggregation period and the reporting time of each charging status message, the charging data of multiple charging guns are aggregated to obtain the corresponding charging pile aggregated data. Multiple charging pile aggregate data within a sampling time period are sampled to obtain multiple charging pile feature data, wherein the sampling time period is determined according to a preset sampling period; A predetermined number of the charging pile feature data are input into a pre-trained power fault identification model to obtain the corresponding power fault identification results, which are either power faults or power normal. If the ratio of the number of charging pile feature data whose power failure identification result is a power failure to the predetermined number is greater than a predetermined ratio, the target split-type charging pile is marked as having a power failure.

2. The method according to claim 1, characterized in that, Each charging gun corresponding to the terminal generates a corresponding charging status message according to the first time period, and the terminal reports it. The charging status message also includes output current, output voltage, desired current and desired voltage of the charging vehicle, and charging gun identification. The charging message also includes a charging check-in message and / or a charging alarm message. The charging check-in message is a message reported by the terminal when establishing a network connection with the charging monitoring server. The charging pile check-in message includes the device information of the target split-type charging pile, and the device information includes the charging gun identifier. The charging alarm message is a message reported by the terminal when it detects a device abnormality. The charging pile alarm message includes the charging order identifier, fault identifier, and fault description information.

3. The method according to claim 1, characterized in that, The step of determining the corresponding charging order information based on the charging order identifier includes: Based on the charging order identifier, request the corresponding charging order information from the charging platform server.

4. The method according to claim 2, characterized in that, The step of filling a preset field template with the charging message and the charging order information to obtain the corresponding charging gun charging data includes: Based on the charging order identifier, determine the charging check-in message and charging alarm message associated with the charging status message; Based on the charging status message and the charging check-in message, charging alarm message, and charging order information associated with the charging status message, the preset field template is filled to obtain the corresponding charging gun charging data.

5. The method according to claim 4, characterized in that, The charging data of multiple charging guns is aggregated according to a preset aggregation period and the reporting time of each charging status message to obtain corresponding charging pile aggregated data, including: The corresponding aggregation time period is obtained according to the preset aggregation period; Identify multiple charging gun charging data entries whose reporting time falls within the aggregated time period; The charging data from the multiple charging guns are aggregated to obtain the corresponding charging pile aggregate data.

6. The method according to claim 1, characterized in that, The process involves sampling aggregated data from multiple charging piles within a sampling time period to obtain multiple charging pile feature data, including: The corresponding sampling time period is determined according to the preset sampling period; Determine aggregated data from multiple charging piles within a preset time period; The aggregated data of the multiple charging piles are sorted according to the time sequence of the reporting time of the charging status messages; The sampling starting point is determined according to the preset sampling rules; Window sliding sampling is performed based on preset window length and preset sliding step size to obtain multiple charging pile feature data.

7. The method according to claim 1, characterized in that, The method further includes: Based on the power fault identification results, generate corresponding charging user prompt information; Send the charging user prompt information to the terminal corresponding to the charging user using the target physical charging pile to prompt the charging user to change the charging pile.

8. The method according to claim 1, characterized in that, The method further includes: Based on the power failure identification results, corresponding charging merchant prompt information and charging pile enterprise prompt information are generated, and the charging merchant prompt information includes power failure parameters; Send the charging merchant prompt information to the terminal corresponding to the charging merchant, so that the charging merchant can determine whether to agree to authorize the sending of the charging enterprise prompt information to the charging enterprise corresponding to the target split charging pile based on the power failure parameters; In response to determining that the terminal corresponding to the charging merchant has agreed to the authorization, a charging pile company prompt message is sent to the terminal corresponding to the charging pile company to prompt the charging pile company's maintenance personnel to carry out maintenance.

9. The method according to claim 1, characterized in that, The power fault identification model is trained through the following steps: Obtain historical charging messages and historical power failure information; Based on the historical charging messages and the historical power failure information, a labeled training sample dataset and a validation sample dataset are obtained. The power supply fault identification model was trained using the aforementioned training sample dataset. The trained power fault identification model is validated using the validation sample dataset until the recognition accuracy of the power fault identification model is higher than the preset accuracy.

10. A power supply fault identification device, characterized in that, The device includes: The acquisition module is used to acquire charging messages reported by multiple terminals of the target split-type charging pile. The charging message includes at least a charging status message, which is used to record the charging status of the charging gun corresponding to the terminal. The charging status message includes a charging order identifier. The determining module is used to determine the corresponding charging order information based on the charging order identifier; The filling module is used to fill a preset field template according to the charging message and the charging order information to obtain the corresponding charging gun charging data, wherein the charging gun charging data includes the reporting time of the charging status message; The aggregation module is used to aggregate the charging data of multiple charging guns according to a preset aggregation period and the reporting time of each charging status message to obtain the corresponding charging pile aggregated data. The sampling module is used to sample aggregated data of multiple charging piles within a sampling time period to obtain multiple charging pile feature data. The sampling time period is determined according to a preset sampling period. The identification module is used to input a predetermined number of the charging pile feature data into a pre-trained power fault identification model to obtain the corresponding power fault identification results, wherein the power fault identification results are power fault or power normal. The marking module is used to mark the target split-type charging pile as having a power failure in response to a predetermined ratio of the number of characteristic data of charging piles with power failure identification results being greater than a predetermined ratio.

11. An electronic device comprising a memory and a processor, characterized in that, The memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method as described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1-9.

13. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method as described in any one of claims 1-9.

Images