Power battery data packet method and device, electronic equipment and computer readable storage medium

Abstract

The application relates to power battery technology and provides a power battery data grouping method and device, electronic equipment and a computer readable storage medium, the method comprises the following steps: acquiring a plurality of battery data collected by a target power battery in a plurality of sampling frames, each battery data comprises a collection time, a charge state and a vehicle state of a corresponding sampling frame; determining an event type of each sampling frame according to the charge state and the vehicle state of the battery data of each sampling frame, the event type representing an operation condition of the target power battery in the corresponding sampling frame; and dividing the plurality of battery data into at least one block group according to the collection time and the event type of each sampling frame. The application can improve the rationality and the consistency of the operation condition of the power battery data grouping.

Description

Technical Field

[0001] This invention relates to the field of power battery technology, and more specifically, to a power battery data grouping method, apparatus, electronic device, and computer-readable storage medium. Background Technology

[0002] With the rapid growth of new energy vehicles, the operational safety of power batteries, as a core energy source, throughout their entire lifecycle has become a major concern in the industry. During long-term, complex use, power batteries inevitably experience various physical and electrochemical anomalies. For example, insulation cracking can lead to leakage and short circuit hazards, excessive internal resistance can cause severe localized heating, and abnormal states such as overvoltage and overtemperature during charging and discharging can directly accelerate battery aging and optimize electrochemical side reactions. If these localized hazards are not detected in time, they can easily evolve into irreversible battery thermal runaway, potentially leading to vehicle fires. Therefore, using cloud-based big data for comprehensive safety early warning of power batteries has significant safety value.

[0003] The grouping of power battery data is crucial for the accuracy and reliability of safety warnings. How to rationally group power battery data is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The present invention aims to provide a method, apparatus, electronic device and computer-readable storage medium for grouping power battery data.

[0005] The embodiments of the present invention can be implemented as follows: In a first aspect, the present invention provides a method for grouping power battery data, the method comprising: Acquire multiple battery data points of the target power battery collected in multiple sampling frames, each of which includes the acquisition time, charging status, and vehicle status of the corresponding sampling frame; Based on the charging status and vehicle status of the battery data in each sampling frame, the event type of each sampling frame is determined, and the event type characterizes the operating condition of the target power battery in the corresponding sampling frame. Based on the acquisition time and event type of each sampling frame, the plurality of battery data are divided into at least one block group, and the sampling frames to which the battery data in the same block group belong have the same event type.

[0006] In an optional implementation, the step of dividing the plurality of battery data into at least one block group according to the acquisition time and event type of each sampling frame includes: All the sampled frames are sorted in chronological order of their acquisition time to obtain a sampled frame sequence; For each of the sampling frame sequences, each sampling frame in the sampling frame sequence is traversed sequentially. If the current frame is the first frame in the sampling frame sequence or the event type of the current frame is different from that of the previous frame, the battery data of the current frame is taken as the first battery data of the new block group. Otherwise, the block group to which the battery data of the current frame belongs is determined based on the current frame and the previous frame, and finally all block groups of the multiple battery data are obtained.

[0007] In an optional implementation, the step of determining the block group to which the battery data of the current frame belongs based on the current frame and the previous frame includes: The sampling frame containing the first battery data in the block group to which the battery data of the previous frame belongs is taken as the first frame of the block; If the event type is not a preset static event, then the block group to which the battery data of the current frame belongs is determined based on the first frame of the block, the previous frame, and the current frame. If the event type is the preset static event, then the block group to which the battery data of the current frame belongs is determined based on the acquisition time of the previous frame and the acquisition time of the current frame.

[0008] In an optional implementation, the battery data further includes the state of charge and current value, wherein the state of charge represents the percentage of the remaining charge of the target power battery relative to a preset available capacity; The step of determining the block group to which the battery data of the current frame belongs based on the first frame of the block, the previous frame, and the current frame includes: The first state of charge of the battery data in the first frame of the block, the second state of charge of the battery data in the previous frame, the current state of charge of the battery data in the current frame, the acquisition time of the first frame of the block, and the acquisition time of the previous frame are obtained. If the event type of the current frame is a preset charging event, then obtain the reference current value of the battery data of the previous frame and the current current value of the battery data of the current frame. If the difference between the first state of charge and the current state of charge is greater than a first preset state of charge threshold, or the current state of charge and the second state of charge meet preset conditions, or the time difference between the acquisition time of the current frame and the acquisition time of the first frame of the block is greater than a first preset time interval, or the event type of the current frame is a preset charging event, and the current difference between the current current value and the reference current value is greater than a jump threshold, then the battery data of the current frame is used as the first battery data of a new block group; otherwise, the battery data of the current frame is assigned to the block group to which the previous frame belongs. The jump threshold is determined based on the preset current range to which the current value of the battery data of the current frame belongs.

[0009] In an optional implementation, the step of determining the block group to which the battery data of the current frame belongs based on the first frame of the block, the previous frame, and the current frame further includes: If the event type of the current frame is a preset discharge event, and the difference between the current state of charge and the second state of charge is greater than the second preset state of charge threshold, then it is determined that the current state of charge and the second state of charge meet the preset conditions. If the event type of the current frame is a preset charging event, and the current state of charge is less than the second state of charge, and the difference between the two is less than a third preset state of charge threshold, then it is determined that the current state of charge and the second state of charge meet the preset conditions.

[0010] In an optional implementation, the step of determining the block group to which the battery data of the current frame belongs based on the acquisition time of the previous frame and the acquisition time of the current frame includes: If the time difference between the acquisition time of the current frame and the acquisition time of the previous frame is greater than the second preset interval threshold, then the battery data of the current frame is used as the first battery data of a new block group; otherwise, the battery data of the current frame is assigned to the block group to which the battery data of the previous frame belongs.

[0011] In an optional implementation, the charging state includes a charging state, a non-charging state, and a charging completed state, and the vehicle state includes a started state, a turned-off state, and other states. The step of determining the event type of each sampling frame based on the charging state and vehicle state of the battery data in each sampling frame includes: If the charging state is the charging state, then the event type is determined to be a preset charging event; otherwise, if the vehicle state is the starting state or the other state, then the event type is determined to be a preset discharging event. If the charging state is not the charging state and the vehicle state is the off state, then the event type is determined to be a preset static event.

[0012] In a second aspect, the present invention provides a power battery data grouping device, the device comprising: The acquisition module is used to acquire multiple battery data collected from the target power battery in multiple sampling frames. Each battery data includes the acquisition time, charging status, and vehicle status of the corresponding sampling frame. The determination module is used to determine the event type of each sampling frame based on the charging state and vehicle state of the battery data in each sampling frame. The event type represents the operating condition of the target power battery in the corresponding sampling frame. The partitioning module is used to divide the multiple battery data into at least one block group according to the acquisition time and event type of each sampling frame, wherein the battery data in the same block group belong to the same event type of the sampling frame.

[0013] Thirdly, the present invention provides an electronic device including a processor and a memory, the memory being used to store a program, and the processor being used to implement the power battery data grouping method as described in the first aspect when executing the program.

[0014] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the power battery data grouping method as described in the first aspect.

[0015] Compared with the prior art, the present invention has the following beneficial effects: Based on the charging state and vehicle state of the acquired target power battery data, this invention determines the event type characterizing the operating condition of the target power battery in the corresponding sampling frame. According to the acquisition time and event type of each sampling frame, multiple battery data are divided into at least one block group. This makes the operating conditions of battery data belonging to the same block group uniform, stable and homogeneous, and makes the grouping granularity match the actual physical process of battery charging and discharging, thus improving the rationality of power battery data grouping. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is an example diagram illustrating the processing procedure of the power battery safety early warning system provided in this embodiment.

[0018] Figure 2 This is a flowchart illustrating the power battery data grouping method provided in this embodiment.

[0019] Figure 3 This is a block diagram illustrating the power battery data grouping device provided in this embodiment.

[0020] Figure 4 This is a block diagram of the electronic device provided in this embodiment.

[0021] Icons: 10-Electronic device; 11-Processor; 12-Memory; 13-Bus; 100-Power battery data grouping device; 110-Acquisition module; 120-Determination module; 130-Division module. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0024] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0025] In the description of this invention, it should be noted that if terms such as "upper," "lower," "inner," or "outer" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of this invention is usually placed, they are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0026] Furthermore, the terms "first" and "second" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0027] It should be noted that, where there is no conflict, the features in the embodiments of the present invention can be combined with each other.

[0028] Please refer to Figure 1 , Figure 1 This is an example diagram illustrating the processing procedure of the power battery safety warning system provided in this embodiment. The processing procedure includes the following stages: (1) Data acquisition: Battery operation data, including total battery voltage, current, SOC (State of Charge), cell voltage, temperature probe data, charging status, vehicle status, etc., are collected in real time through the battery management system (BMS) and uploaded to the cloud big data platform through the vehicle terminal and vehicle network service platform.

[0029] (2) Data cleaning: missing value processing, duplicate value removal, outlier identification and filtering are performed on the received raw data to ensure the input quality of subsequent analysis.

[0030] (3) Data grouping: The continuous time-series sampled data is divided into several analysis units with business semantics (such as a charging process, a sliding window, a SOC interval, etc.) as the basic granularity for feature calculation and anomaly detection.

[0031] (4) Feature extraction or model calculation: Based on the grouping results, some schemes use feature engineering to extract statistical features, incremental features, consistency features, etc.; some schemes use machine learning or deep learning models to perform anomaly scoring and fault probability estimation.

[0032] (5) Alarm triggering: When the feature value or model output exceeds the preset threshold, a safety warning is triggered.

[0033] In the above processing, data grouping is at the pivotal position connecting data cleaning and early warning algorithms. Whether the grouping method is reasonable determines whether the working conditions within the group are stable, whether the data have similarity, and whether the features are comparable, thus affecting the accuracy and reliability of the early warning. If the grouping granularity does not match the real physical working conditions, it will lead to the dilution of features, the incorrect cutting of boundaries, or the omission of key working condition changes, ultimately causing missed or false early warnings. It can be seen that the main technical problems in data grouping are: (1) how to make the data grouping as close as possible to the actual working conditions, ensure that the working conditions within the group are uniform, stable and homogeneous, and match the grouping granularity with the real physical process of battery charging and discharging; (2) how to generate grouping labels at the data cleaning layer in the case of large data volume, as a unified input for multiple downstream early warning algorithms, to avoid repeated splitting and repeated calculation by each algorithm.

[0034] To address the aforementioned technical issues, the relevant data grouping methods mainly include the following: (1) Grouping by the whole event This method treats a complete charging process, discharging process, or resting process as a grouping unit and calculates features uniformly within the entire event.

[0035] This method has the following drawbacks: (a) Uneven operating conditions within the group, which do not correspond to the actual physical process. A complete charging process usually includes a constant current stage, a constant voltage stage, and possible sub-processes such as pause, recovery, and current step; the current amplitude, temperature rise rate, etc., vary in different sub-stages. Treating the entire charging process as a single group will mix the sampling frames of different operating conditions for calculation, resulting in the "averaging" of features such as voltage increment and internal resistance estimation, which will mask local anomalies within the segment; (b) Insufficient applicability to discharge and resting. During discharge, the SOC range is large and the load changes frequently; during resting, although the current is close to zero, long-term resting may still result in self-discharge anomalies, insulation anomalies, etc., and the entire group cannot distinguish the risk characteristics of different resting stages.

[0036] (2) Grouping by fixed windows This method first identifies the boundaries of events such as charging and discharging. Within an event, the data is sorted according to the sampling time. Then, the event is divided into equal-length windows with a fixed number of frames, and feature statistics are performed on a window-by-window basis.

[0037] This approach has the following drawbacks: (a) When using fixed sliding window grouping within a single event, the window boundaries are difficult to align with the actual sub-conditions. A single event often encompasses multiple sub-processes such as constant current / constant voltage switching, load mutation, and rapid SOC changes. However, the cut-off point of a fixed window length is arbitrary, easily resulting in a mixture of multiple sub-conditions within the same window. The features within the window cannot correspond to a single, complete actual condition. This leads to the risk of missed alarms: if an abnormal signal that should be triggered is cut off by the window boundary, its key features (such as voltage extreme values) may be scattered in adjacent windows, and the features of each window are incomplete and cannot exceed the alarm threshold, thus causing missed alarms; (b) In scenarios with extremely large data volumes, repetitive calculations are severe, resulting in high computational pressure. In cloud-based big data platforms with massive vehicle-side operational data (millions of vehicles, second-level sampling), the daily data volume can reach TB levels. In early warning systems, multiple alarm strategies often run in parallel (such as internal resistance anomalies, temperature rise anomalies, individual outliers, insulation anomalies, etc.). If each alarm algorithm needs to independently perform fixed sliding window segmentation and feature calculation within each charging and discharging event, it will result in: time series data within the same event being repeatedly segmented and traversed; and a large amount of computing resources being consumed on window division and basic statistics within the event.

[0038] (3) According to the fixed SOC interval division method This method divides historical data into segments according to preset SOC intervals (e.g., each interval is 10%), and extracts feature values ​​within each SOC interval.

[0039] This method has the following drawbacks: (a) It cannot characterize dynamic dwell characteristics. The dwell time of a vehicle within the same SOC range varies significantly—from as short as a few seconds to as long as several hours; a fixed range cannot comprehensively characterize the operation process by combining dwell time, current level, and event type, resulting in a coarse characterization of dynamic behavior; (b) It is prone to missed detection near the boundary. If an anomaly occurs near the boundary of the range, its characteristics are often dispersed to adjacent ranges, and the anomaly signal within a single range is incomplete and diluted in intensity, thereby weakening the detection capability and leading to the risk of missed warnings.

[0040] Based on an in-depth analysis of the aforementioned implementation methods, the inventors discovered that the power battery safety early warning system should have the following essential requirements for data grouping: (1) To be as close as possible to actual working conditions. The group boundaries should be dynamically determined according to the physical processes such as the charging and discharging state of the battery, current changes, SOC changes, and duration, so that the sampling frames in each group are under similar working conditions, rather than artificially fixed frame numbers, SOC scales, or entire events.

[0041] (2) The operating conditions within the group should be kept consistent and stable. The current level, SOC change trend and event type within the same group should be as consistent as possible to ensure that the characteristics within the group (such as voltage increment, internal resistance, temperature rise rate and consistency index) are comparable and statistically significant, and reduce false alarms caused by mixed operating conditions.

[0042] (3) The boundaries between groups are physically interpretable. The group splitting point should correspond to the current jump and SOC.

[0043] (4) Explainable physical or behavioral changes such as callbacks / rises, event type switching, duration exceeding limits, etc., rather than arbitrary equal-length splits.

[0044] Furthermore, the inventors discovered that in cloud-based scenarios involving massive amounts of data, historical data from the same battery pack is often repeatedly accessed by multiple early warning algorithms. If each downstream module executes its grouping logic independently, the following problems will arise: (1) Inconsistent grouping results. The same sampled frame may be classified into different groups in different alarm modules, resulting in inconsistent feature calculation methods and difficulty in cross-validating the warning conclusions.

[0045] (2) Repetitive consumption of computing resources. Each alarm algorithm performs grouping and preprocessing independently, resulting in a large amount of repetitive calculations under extremely large data volumes, which significantly increases cloud computing power consumption and processing latency.

[0046] In view of this, this embodiment provides a power battery data grouping method, apparatus, electronic device, and computer-readable storage medium to solve the above problems, which will be described in detail below.

[0047] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the power battery data grouping method provided in this embodiment. The method includes the following steps: Step S101: Acquire multiple battery data of the target power battery collected in multiple sampling frames. Each battery data includes the collection time, charging status and vehicle status of the corresponding sampling frame.

[0048] In this embodiment, multiple raw data points collected from multiple power batteries within a preset time period and at a preset cycle in each sampling frame can be read from the server. Preset fields are extracted from each raw data point. These preset fields include, but are not limited to, battery code, sampling frame acquisition time, total voltage, total current, SOC, charging state, vehicle state, individual cell voltage array, temperature probe array, and insulation resistance. The sampling frame acquisition time can be a sampling timestamp. Basic filtering is performed on the extracted data to remove invalid data. Invalid data can be data with an empty battery code or an empty acquisition time. Other invalid rules can also be set as needed to identify invalid data. For example, invalid rules can include a field value greater than a preset value or a field value that does not conform to the data type of that field.

[0049] In this embodiment, since the data read from the server includes data from multiple power batteries, and the battery code is used to uniquely identify the power battery, it is possible to first make a preliminary division according to the battery code, group the data with the same battery code into the same processing unit, and then group the power battery corresponding to each battery code as the target power battery one by one, using the power battery data grouping method provided in this embodiment.

[0050] It should be noted that, for the convenience of subsequent processing, the sampling frames of the same processing unit can be sorted according to the order of collection time. When sorting, for multiple sampling frames with the same collection time, deduplication can be performed, and only one sampling frame needs to be kept. You can randomly select any sampling frame from multiple repeated sampling frames to keep, or you can keep the first sampling frame in the multiple sampling frames to avoid distortion of subsequent group counting due to repeated reporting.

[0051] Step S102: Based on the charging status and vehicle status of the battery data in each sampling frame, determine the event type of each sampling frame. The event type represents the operating condition of the target power battery in the corresponding sampling frame.

[0052] In this embodiment, the charging state includes a charging state and a non-charging state, with the charging state including charging while parked and charging while driving. Vehicle states include vehicle started, vehicle off, and other states. Event types include preset idle events, preset charging events, and preset discharging events.

[0053] Step S103: Based on the acquisition time and event type of each sampling frame, divide multiple battery data into at least one block group, and the sampling frames to which the battery data in the same block group belong have the same event type.

[0054] In this embodiment, the battery data within the same block group belongs to the same event type. For example, block group 1 includes 100 battery data points, and the event type of the sampling frames to which these 100 battery data points belong is a preset charging event. Multiple block groups can have the same event type. In this case, battery data of the same event type can be divided into multiple block groups. For example, the event type of the sampling frames to which the battery data in block group 1 and block group 2 belong is a preset discharging event.

[0055] The method provided in this embodiment determines the event type characterizing the operating condition of the target power battery in the corresponding sampling frame based on the charging state and vehicle state of the acquired target power battery data. According to the acquisition time and event type of each sampling frame, the battery data of the sampling frames of the same event type are divided into at least one block group. This makes the operating conditions of the battery data belonging to the same block group uniform, stable and homogeneous, and makes the grouping granularity match the actual physical process of battery charging and discharging, thereby improving the rationality of power battery data grouping.

[0056] In an optional implementation, to ensure that the same physical process of the target power battery can always be mapped to a consistent event label under different vehicles, operating conditions, and data quality conditions, and to provide a stable, reusable, and unambiguous top-level classification basis for subsequent dynamic grouping, this embodiment provides an implementation method for determining event types: If the charging status is "charging in progress", the event type is determined to be a preset charging event; otherwise, if the vehicle status is "starting" or other status, the event type is determined to be a preset discharging event. If the charging status is not in the charging state and the vehicle status is off, the event type is determined to be a preset static event.

[0057] In this embodiment, both the charging status and vehicle status are derived from national standard data uploaded by the vehicle. The charging status refers to whether the vehicle is currently in an energy input state, while the vehicle status represents the operating status of the vehicle's powertrain system. The specific values ​​for both conform to unified coding rules. For example, for the charging status, 1 represents charging while parked, 2 represents charging while driving, 3 represents not charging, and 4 represents charging complete. Parking and driving charging are both considered charging states. For the vehicle status, 1 represents the started state, 2 represents the off state, and 3 represents other states, which are states other than the started and off states.

[0058] As one implementation method, the judgment logic for determining the event type can be as follows: First, check whether the charging state is 1 or 2. If so, mark it directly as a preset charging event. If not, further check whether the vehicle state is 1 or 3. If so, mark it as a preset discharging event. Only when the charging state is neither 1 nor 2 and the vehicle state is exactly 2, is it marked as a preset stationary event.

[0059] In this embodiment, to ensure that the grouping logic is based on the real-time evolution relationship and to avoid misjudgment of working condition boundaries due to data reporting delays, retransmissions, or out-of-order delivery, this embodiment provides an implementation method for determining block grouping frame by frame according to the chronological order of data acquisition: First, sort all the sampled frames in chronological order of their acquisition time to obtain the sampled frame sequence; Secondly, for each sampling frame sequence, each sampling frame in the sampling frame sequence is traversed in turn. If the current frame is the first frame in the sampling frame sequence or the event type of the current frame is different from that of the previous frame, the battery data of the current frame is taken as the first battery data of the new block group. Otherwise, the block group to which the battery data of the current frame belongs is determined according to the current frame and the previous frame, and finally all block groups of multiple battery data are obtained.

[0060] In this embodiment, the current frame is the currently traversed sampling frame, and the previous frame is the sampling frame adjacent to and preceding the current frame in the sampling frame sequence. If the current frame is the first frame in the sampling frame sequence, then the battery data of the current frame is the first battery data in the new block group. If the current frame is not the first frame, and the event type of the current frame is inconsistent with the event type of the previous frame, in order to avoid grouping the battery data of sampling frames with different event types into the same block group, the battery data of the current frame is taken as the first battery data in the new block group. If none of the above conditions are met, then the block grouping needs to be performed according to the actual situation of the current frame and the previous frame.

[0061] It should be noted that while using the battery data of the current frame as the first battery data in a new block group, to facilitate the division of subsequent block groups, the new block group can also undergo first-frame initialization processing. This includes: using the acquisition time of the current frame as the block group identifier (block_group_tag), and setting the intra-group sequence number (event_group_tag_rank) of the battery data in the current frame to 1, with the intra-group sequence number of subsequent battery data belonging to this block group increasing sequentially. First-frame metrics are recorded according to their event type. If the event type is a preset static event, the acquisition time of the current frame is recorded as the first-frame static time. If the event type is a preset charging event, the charging SOC of the current frame is recorded as the first-frame SOC, and the acquisition time of the current frame is recorded as the first-frame charging time. If the event type is a preset discharging event, the discharging SOC of the current frame is recorded as the first-frame SOC, and the acquisition time of the current frame is recorded as the first-frame discharging time.

[0062] It should also be noted that, in order to provide high-quality and consistent input for the security warning algorithm located downstream of the group, an event group identifier (event_group_tag) can be generated for each event type, and a block group identifier can be set for a block group. Each block group can be identified by combining the event group identifier and the block group identifier.

[0063] In an optional implementation, to enable block grouping to respond to time drift in static scenarios and capture multi-dimensional operational condition jumps during charging and discharging, this embodiment adopts different classification and processing strategies based on the differences in the physical characteristics of event types. One implementation method is as follows: First, the sampling frame containing the first battery data in the block group to which the battery data of the previous frame belongs is taken as the first frame of the block; Secondly, if the event type is not a preset static event, the block group to which the battery data of the current frame belongs is determined based on the first frame of the block, the previous frame, and the current frame. In this embodiment, for non-preset static events, the judgment needs to be made by comprehensively considering the parameter changes among the first frame, the previous frame, and the current frame. For example, for charging events, the state of charge (SOC) in the battery data of the current frame and the first frame of the block, as well as the SOC change trends of the current frame and the previous frame, need to be considered simultaneously to ensure that the operating conditions within the block group remain consistent and stable. The SOC represents the percentage of the remaining charge of the target power battery relative to the preset available capacity.

[0064] Finally, if the event type is a preset static event, the block group to which the battery data of the current frame belongs is determined based on the acquisition time of the previous frame and the acquisition time of the current frame.

[0065] In this embodiment, for the preset static event, since the current approaches zero and the SOC remains basically unchanged under this condition, the core change is most significantly affected by time. Therefore, it is only necessary to determine whether to segment based on the acquisition time of the current frame and the previous frame, and finally determine the block group to which the battery data of the current frame belongs.

[0066] In an optional implementation, to ensure that the block grouping boundary can be naturally anchored to the substantial turning point of the charging and discharging process, rather than a fixed scale or average trend set artificially, this embodiment sets differentiated triggering conditions through multiple parameters such as the first frame, the previous frame, and the current frame, and comprehensively judges whether the current frame still belongs to the same stable operating range. One implementation method is as follows: First, acquire the first state of charge of the battery data in the first frame of the block, the second state of charge of the battery data in the previous frame, the current state of charge of the battery data in the current frame, and the acquisition time of the first frame and the acquisition time of the previous frame. Secondly, if the event type of the current frame is a preset charging event, then obtain the reference current value of the battery data in the previous frame and the current current value of the battery data in the current frame. In this embodiment, for the preset charging event, when the constant current to constant voltage switch, the charging pile output power is adjusted, or the charging strategy is switched, it means that the charging is in a new stage and a significant current change occurs, resulting in a step change in the current. At this time, it is also necessary to perform block grouping.

[0067] Finally, if the difference between the first state of charge and the current state of charge is greater than the first preset state of charge threshold, or the current state of charge and the second state of charge meet preset conditions, or the time difference between the acquisition time of the current frame and the acquisition time of the first frame of the block is greater than the first preset time interval, or the event type of the current frame is a preset charging event, and the current difference between the current current value and the reference current value is greater than the jump threshold, then the battery data of the current frame is used as the first battery data of the new block group; otherwise, the battery data of the current frame is assigned to the block group to which the previous frame belongs. The jump threshold is determined based on the preset current range to which the current value of the battery data of the current frame belongs.

[0068] In this embodiment, the first state of charge serves as the initial energy reference for the block group, the second state of charge reflects the energy level at a recent moment, and the current state of charge represents the remaining charge at the latest sampling point.

[0069] As one implementation method, for a preset discharge event, the difference between the first state of charge and the current state of charge = first state of charge - current state of charge. The first preset charge threshold can be preset, for example, 5%. This indicates that the discharge process has reached a certain discharge depth and needs to be divided into new block groups. For a preset charging event, the difference between the first state of charge and the current state of charge = current state of charge - first state of charge. This indicates that the charging process has reached a certain depth range and needs to be divided into new block groups. It should be noted that the first preset charge threshold used for the preset discharge event and the preset charging event can be the same, for example, both 5%, or it can be set differently according to their actual needs.

[0070] In this embodiment, whether the preset conditions are met depends on comparing the difference between the current state of charge and the second state of charge to determine whether a sudden change in operating conditions has occurred.

[0071] In this embodiment, a time difference between the acquisition time of the current frame and the acquisition time of the first frame of the block being greater than a first preset time interval means that the discharge or charging process has been ongoing for a long time, necessitating the segmentation of a new block group. It should be noted that the first preset time interval used for the preset discharge event and the preset charging event can be the same, for example, both set to 3600s, or it can be set differently according to their respective actual needs.

[0072] In this embodiment, the event type of the current frame is a preset charging event, and the current difference between the current value and the reference current value is greater than the jump threshold, which means that a step change in current has occurred during the charging process. This is a case of sudden current change, and new block grouping is required.

[0073] It should be noted that the threshold can be a variable, with different thresholds set according to the absolute value of the current. For example, if the absolute value of the current is in the range (0,10], the threshold is 5; if the absolute value of the current is in the range (10,50], the threshold is 10; if the absolute value of the current is in the range (50,300], the threshold is 0.2 * the absolute value of the current; and if the absolute value of the current is in other ranges, the threshold is 0.3 * the absolute value of the current.

[0074] The method provided in this embodiment covers the typical operating condition transitions of different categories during the charging and discharging process when new block grouping is required: the first state of charge and the current state of charge characterize the SOC span and reflect the SOC depth change; the current state of charge and the second state of charge can determine the SOC rollback and reflect abnormal disturbances; the time difference between the acquisition time of the current frame and the acquisition time of the first frame of the block reflects the steady state continuity; and the current value jump indicates the switching of the charging stage. This avoids mixing real sub-processes such as constant current to constant voltage, sudden load increase, and brief feedback into the same block group.

[0075] In an optional implementation, in order to refine the state of charge of battery data in two adjacent sampling frames during a preset discharge event and a preset charge event, and to give the block grouping a more granular physical meaning, this embodiment provides a method for determining whether the current state of charge and the second state of charge meet preset conditions: For a preset discharge event: if the difference between the current state of charge and the second state of charge is greater than the second preset state of charge threshold, then the current state of charge and the second state of charge are determined to meet the preset conditions.

[0076] In this embodiment, the difference between the current state of charge and the second state of charge is equal to the current state of charge minus the second state of charge. The second preset state of charge threshold can be preset, for example, it can be preset to 1%. If the difference is greater than the second preset state of charge threshold, it means that the SOC has rebounded. This corresponds to conditions such as energy feedback, short-term charging disturbance after discharge, and sudden load change. It indicates that the discharge condition has changed and the preset condition is met. Therefore, a new block grouping is required.

[0077] For a preset charging event: if the current state of charge is less than the second state of charge and the difference between the two is less than the third preset state of charge threshold, then the current state of charge and the second state of charge are determined to meet the preset conditions.

[0078] In this embodiment, the difference between the current state of charge (SOC) and the second state of charge (SOC) is equal to the difference between the current SOC and the second SOC. Under normal circumstances, the SOC should gradually increase, and the difference should be non-negative. If the difference is negative and less than the third preset charge threshold, it means that the SOC decreases instead of increasing during charging, indicating an abnormal rollback. This is deemed to meet the preset conditions, and a new block grouping is required. The third preset charge threshold can be preset, for example, it can be -1%.

[0079] In an optional implementation, for a preset static event, one way to determine the block group to which the battery data of the current frame belongs is as follows: If the time difference between the acquisition time of the current frame and the acquisition time of the previous frame is greater than the second preset interval threshold, the battery data of the current frame will be used as the first battery data of the new block group; otherwise, the battery data of the current frame will be assigned to the block group to which the battery data of the previous frame belong.

[0080] In this embodiment, the second preset interval threshold can be set as needed, for example, the second preset interval can be set to 600s.

[0081] To perform the corresponding steps in the above embodiments and various possible implementations, an implementation method of the power battery data grouping device 100 is given below. Please refer to... Figure 3 , Figure 3This is a block diagram of the power battery data grouping device provided in this embodiment. It should be noted that the basic principle and technical effects of the power battery data grouping device 100 provided by the present invention are the same as those of the corresponding embodiments described above. For the sake of brevity, some parts of this embodiment are not mentioned.

[0082] The power battery data grouping device 100 includes an acquisition module 110, a determination module 120, and a division module 130.

[0083] The acquisition module 110 is used to acquire multiple battery data collected from the target power battery in multiple sampling frames. Each battery data includes the acquisition time, charging status, and vehicle status of the corresponding sampling frame.

[0084] The determination module 120 is used to determine the event type of each sampling frame based on the charging state and vehicle state of the battery data in each sampling frame. The event type represents the operating condition of the target power battery in the corresponding sampling frame.

[0085] The partitioning module 130 is used to divide multiple battery data into at least one block group according to the acquisition time and event type of each sampling frame, wherein the battery data in the same block group belong to the same event type of the sampling frame.

[0086] In an optional implementation, the partitioning module 130 is specifically used for: All sampled frames are sorted in chronological order of acquisition time to obtain a sampled frame sequence; For each sampling frame sequence, each sampling frame in the sampling frame sequence is traversed in turn. If the current frame is the first frame in the sampling frame sequence or the event type of the current frame is different from that of the previous frame, the battery data of the current frame is taken as the first battery data of the new block group. Otherwise, the block group to which the battery data of the current frame belongs is determined according to the current frame and the previous frame, and finally all block groups of multiple battery data are obtained.

[0087] In an optional implementation, the partitioning module 130, when specifically used to determine the block group to which the battery data of the current frame belongs based on the current frame and the previous frame, is further used to: The sampling frame containing the first battery data in the block group containing the battery data of the previous frame is taken as the first frame of the block; If the event type is not a preset static event, the block group to which the battery data of the current frame belongs is determined based on the first frame of the block, the previous frame, and the current frame. If the event type is the preset static event, then the block group to which the battery data of the current frame belongs is determined based on the acquisition time of the previous frame and the acquisition time of the current frame.

[0088] In an optional implementation, the battery data also includes state of charge and current value, whereby the state of charge characterizes the percentage of the remaining charge of the target power battery relative to a preset available capacity. In its specific function of determining the block group to which the battery data of the current frame belongs based on the first frame, the previous frame, and the current frame, the partitioning module 130 is further used for: Acquire the first state of charge of the battery data in the first frame of the block, the second state of charge of the battery data in the previous frame, the current state of charge of the battery data in the current frame, and the acquisition time of the first frame and the acquisition time of the previous frame. If the event type of the current frame is a preset charging event, then obtain the reference current value of the battery data in the previous frame and the current current value of the battery data in the current frame. If the difference between the first state of charge and the current state of charge is greater than the first preset state of charge threshold, or the current state of charge and the second state of charge meet preset conditions, or the time difference between the acquisition time of the current frame and the acquisition time of the first frame of the block is greater than the first preset time interval, or the event type of the current frame is a preset charging event, and the current difference between the current current value and the reference current value is greater than the jump threshold, then the battery data of the current frame is used as the first battery data of the new block group; otherwise, the battery data of the current frame is assigned to the block group to which the previous frame belongs. The jump threshold is determined based on the preset current range to which the current value of the battery data of the current frame belongs.

[0089] In an optional implementation, when the partitioning module 130 is specifically used to determine the block group to which the battery data of the current frame belongs based on the block start frame, the previous frame, and the current frame, it is further used to: If the event type of the current frame is a preset discharge event, and the difference between the current charge state and the second charge state is greater than the second preset charge threshold, then it is determined that the current charge state and the second charge state meet the preset conditions. If the event type of the current frame is a preset charging event, and the current state of charge is less than the second state of charge, and the difference between the two is less than the third preset state of charge threshold, then the current state of charge and the second state of charge are determined to meet the preset conditions.

[0090] In an optional implementation, when the partitioning module 130 is specifically used to determine the block group to which the battery data of the current frame belongs based on the acquisition time of the previous frame and the acquisition time of the current frame, it is further used to: If the time difference between the acquisition time of the current frame and the acquisition time of the previous frame is greater than the second preset interval threshold, the battery data of the current frame will be used as the first battery data of the new block group; otherwise, the battery data of the current frame will be assigned to the block group to which the battery data of the previous frame belong.

[0091] In an optional implementation, the charging state includes a charging state, a non-charging state, and a charging completed state, and the vehicle state includes a started state, a turned-off state, and other states. Module 120 is specifically used for: If the charging status is "charging in progress", the event type is determined to be a preset charging event; otherwise, if the vehicle status is "starting" or other status, the event type is determined to be a preset discharging event. If the charging status is not in the charging state and the vehicle status is off, the event type is determined to be a preset static event.

[0092] This invention also provides a block diagram of an electronic device 10, which implements the power battery data grouping method described in the aforementioned embodiments. Please refer to... Figure 4 , Figure 4 This is a block diagram of the electronic device 10 provided in this embodiment. The electronic device 10 includes a processor 11, a memory 12 and a bus 13. The processor 11 and the memory 12 are connected through the bus 13.

[0093] The processor 11 can be an integrated circuit chip with signal processing capabilities. In implementation, each step of the power battery data grouping method described above can be completed by the integrated logic circuitry in the processor 11 or by software instructions. The processor 11 can be a general-purpose processor, including a CPU (Central Processing Unit), NP (Network Processor), GPU (Graphics Processing Unit), etc.; it can also be a DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Logic Gate Array), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0094] The memory 12 is used to store a program for implementing the power battery data grouping method. This program can be a software function module stored in the memory 12 in the form of software or firmware or embedded in the OS (Operating System) of the electronic device 10.

[0095] After receiving the execution instruction, the processor 11 executes the program to implement the power battery data grouping method of the aforementioned embodiment.

[0096] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the power battery data grouping method as described in this embodiment.

[0097] In summary, embodiments of the present invention provide a method, apparatus, electronic device, and computer-readable storage medium for grouping power battery data. The method includes: acquiring multiple battery data of a target power battery, each battery data including acquisition time, charging status, and vehicle status; determining the event type of each battery data based on the charging status and vehicle status, the event type representing the operating condition of the target power battery; and dividing the battery data of the same event type into at least one block group based on the acquisition time and event type of each battery data.Compared with the prior art, this embodiment has at least the following advantages: (1) Based on the charging state and vehicle state of the acquired target power battery data, the event type characterizing the operating condition of the target power battery in the corresponding sampling frame is determined. Based on the acquisition time and event type of each sampling frame, the battery data of the sampling frames of the same event type are divided into at least one block group. Thus, the battery data operating conditions belonging to the same block group are unified, stable and homogeneous, and the grouping granularity matches the real physical process of battery charging and discharging, improving the rationality of power battery data grouping; (2) In view of the defects in the prior art that a complete charge / discharge / rest is taken as a single grouping unit, ignoring the differences in internal sub-operating conditions, resulting in heterogeneous physical states within the group, averaged features and masked local anomalies, this embodiment determines the event type based on the charging state and vehicle state. And according to the event type differentiation segmentation logic, different event types adopt different grouping strategies to exclude cross-event mixing, so that the entire charging process or the entire discharging process is decomposed into multiple physical semantically clear sub-blocks (such as: constant current initial segment, constant current middle segment, constant voltage final segment, abnormal interruption segment), and the current, SOC slope, and temperature rise rate within each block are highly consistent; (3) In view of the defects of the existing technology using a fixed frame number sliding window (such as S=100 frames), the window boundary is not related to the physical process, and it is easy to cut key anomalies (such as the voltage extreme point is exactly on both sides of the window boundary), resulting in feature dispersion, unreachable threshold, and missed warning. This embodiment relies on state change events (such as current jump, SOC inflection point, time over-limit) to determine the grouping segmentation conditions. The segmentation boundary is the time when an interpretable physical event occurs, and the data in the same group has strong temporal homogeneity. (4) In response to the shortcomings of existing technologies that divide according to fixed SOC scale and ignore the huge differences in dwell time, current level and event type within the same SOC interval, resulting in the lack of distinguishability of features and the failure to detect the boundary, this embodiment uses SOC-related criteria as relative change quantities (e.g., the first frame and the current frame or the previous frame and the current frame) when the block grouping is triggered, which reflects the dynamics of the process, so that the data in the same group meet the same event type, approximate SOC change rate, controllable time span, and current stability under charging scenario, so that the grouping result naturally carries the timing dynamics information of battery operation; (5) In response to the shortcomings of existing technologies that each alarm module implements its own grouping logic, resulting in repeated grouping and inconsistent results of multiple early warning algorithms in big data scenarios, this embodiment uses event identifiers and group identifiers as data tags, which are persistently stored along with the original data. The data tags are calculated once and reused permanently. All subsequent downstream algorithms (no matter how many new ones are added) can directly read the data tags to obtain complete and consistent analysis units, eliminating the source of ambiguity caused by the difference in grouping logic on the algorithm side.

[0098] The above descriptions are merely various embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for grouping power battery data, characterized in that, The method includes: Acquire multiple battery data points of the target power battery collected in multiple sampling frames, each of which includes the acquisition time, charging status, and vehicle status of the corresponding sampling frame; Based on the charging status and vehicle status of the battery data in each sampling frame, the event type of each sampling frame is determined, and the event type characterizes the operating condition of the target power battery in the corresponding sampling frame. Based on the acquisition time and event type of each sampling frame, the plurality of battery data are divided into at least one block group, and the sampling frames to which the battery data in the same block group belong have the same event type.

2. The method according to claim 1, characterized in that, The step of dividing the plurality of battery data into at least one block group according to the acquisition time and event type of each sampling frame includes: All the sampled frames are sorted in chronological order of their acquisition time to obtain a sampled frame sequence; For each of the sampling frame sequences, each sampling frame in the sampling frame sequence is traversed sequentially. If the current frame is the first frame in the sampling frame sequence or the event type of the current frame is different from that of the previous frame, the battery data of the current frame is taken as the first battery data of the new block group. Otherwise, the block group to which the battery data of the current frame belongs is determined based on the current frame and the previous frame, and finally all block groups of the multiple battery data are obtained.

3. The method according to claim 2, characterized in that, The step of determining the block group to which the battery data of the current frame belongs based on the current frame and the previous frame includes: The sampling frame containing the first battery data in the block group to which the battery data of the previous frame belongs is taken as the first frame of the block; If the event type is not a preset static event, then the block group to which the battery data of the current frame belongs is determined based on the first frame of the block, the previous frame, and the current frame. If the event type is the preset static event, then the block group to which the battery data of the current frame belongs is determined based on the acquisition time of the previous frame and the acquisition time of the current frame.

4. The method according to claim 3, characterized in that, The battery data also includes the state of charge and current value, wherein the state of charge represents the percentage of the remaining charge of the target power battery relative to the preset available capacity; The step of determining the block group to which the battery data of the current frame belongs based on the first frame of the block, the previous frame, and the current frame includes: The first state of charge of the battery data in the first frame of the block, the second state of charge of the battery data in the previous frame, the current state of charge of the battery data in the current frame, the acquisition time of the first frame of the block, and the acquisition time of the previous frame are obtained. If the event type of the current frame is a preset charging event, then obtain the reference current value of the battery data of the previous frame and the current current value of the battery data of the current frame. If the difference between the first state of charge and the current state of charge is greater than a first preset state of charge threshold, or the current state of charge and the second state of charge meet preset conditions, or the time difference between the acquisition time of the current frame and the acquisition time of the first frame of the block is greater than a first preset time interval, or the event type of the current frame is a preset charging event, and the current difference between the current current value and the reference current value is greater than a jump threshold, then the battery data of the current frame is used as the first battery data of a new block group; otherwise, the battery data of the current frame is assigned to the block group to which the previous frame belongs. The jump threshold is determined based on the preset current range to which the current value of the battery data of the current frame belongs.

5. The method according to claim 4, characterized in that, The step of determining the block group to which the battery data of the current frame belongs based on the first frame of the block, the previous frame, and the current frame further includes: If the event type of the current frame is a preset discharge event, and the difference between the current state of charge and the second state of charge is greater than the second preset state of charge threshold, then it is determined that the current state of charge and the second state of charge meet the preset conditions. If the event type of the current frame is a preset charging event, and the current state of charge is less than the second state of charge, and the difference between the two is less than a third preset state of charge threshold, then it is determined that the current state of charge and the second state of charge meet the preset conditions.

6. The method according to claim 3, characterized in that, The step of determining the block group to which the battery data of the current frame belongs based on the acquisition time of the previous frame and the acquisition time of the current frame includes: If the time difference between the acquisition time of the current frame and the acquisition time of the previous frame is greater than the second preset interval threshold, then the battery data of the current frame is used as the first battery data of a new block group; otherwise, the battery data of the current frame is assigned to the block group to which the battery data of the previous frame belongs.

7. The method according to claim 1, characterized in that, The charging status includes charging, not charging, and charging completed; the vehicle status includes starting, turning off, and other statuses. The step of determining the event type of each sampling frame based on the charging state and vehicle state of the battery data in each sampling frame includes: If the charging state is the charging state, then the event type is determined to be a preset charging event; otherwise, if the vehicle state is the starting state or the other state, then the event type is determined to be a preset discharging event. If the charging state is not the charging state and the vehicle state is the off state, then the event type is determined to be a preset static event.

8. A power battery data grouping device, characterized in that, The device includes: The acquisition module is used to acquire multiple battery data collected from the target power battery in multiple sampling frames. Each battery data includes the acquisition time, charging status, and vehicle status of the corresponding sampling frame. The determination module is used to determine the event type of each sampling frame based on the charging state and vehicle state of the battery data in each sampling frame. The event type represents the operating condition of the target power battery in the corresponding sampling frame. The partitioning module is used to divide the multiple battery data into at least one block group according to the acquisition time and event type of each sampling frame, wherein the battery data in the same block group belong to the same event type of the sampling frame.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory being used to store a program, and the processor being used to implement the power battery data grouping method as described in any one of claims 1-7 when executing the program.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the power battery data grouping method as described in any one of claims 1-7.

Images