A method and device for early warning and identification of abnormal discharge of a power battery

Abstract

The application discloses a power battery abnormal discharge early warning identification method and device. The method comprises the following steps: extracting non-charging data of each vehicle; dividing the non-charging data into discharge sections based on a first set condition; calculating abnormal discharge characteristic parameters of each discharge section based on the non-charging data of each discharge section; constructing a single-batch vehicle discharge statistical summary table based on the abnormal discharge characteristic parameters of each discharge section, discharge data at the start and end time; performing power battery abnormal discharge early warning identification based on the single-batch vehicle discharge statistical summary table and early warning conditions to obtain discharge early warning records; generating a single-batch early warning result list based on each discharge early warning record, and determining the risk level of the discharge early warning record based on the abnormal discharge characteristic parameters. The application can accurately identify long-time small-current abnormal discharge working conditions, thereby avoiding battery over-discharge and safety accidents in advance, providing reliable protection for power battery safety monitoring, and improving the intelligent level of new energy vehicle battery management.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle power battery safety technology, and more specifically, relates to a method and device for early warning and identification of abnormal discharge of power batteries. Background Technology

[0002] As the core power source of new energy vehicles, the safety performance of power batteries is directly related to the operational safety of vehicles and the safety of users' lives and property. It is one of the key factors restricting the sustainable and healthy development of the new energy vehicle industry.

[0003] In the actual use of new energy vehicles, abnormal operating conditions of the power battery are often a major cause of safety accidents. Among them, the risk of prolonged abnormal discharge is particularly noteworthy. According to practical cases, a fire occurred in a certain model project. Before the failure, the vehicle exhibited typical prolonged abnormal discharge: the vehicle was in a stationary state (specifically, the vehicle was awakened and the high voltage was applied, and the high voltage relay of the power battery remained closed). The power battery continued to discharge with an extremely small current (usually less than 1A on average) for more than 24 hours, eventually causing the battery to be completely discharged to 0, and the entire vehicle lost power.

[0004] Under such abnormal operating conditions, power batteries face a significant risk of over-discharge. Prolonged over-discharge can severely damage the battery's chemical properties and structural stability, potentially leading to a sharp decline in battery performance, internal short circuits, thermal runaway, and other serious consequences, thereby posing a major safety hazard of vehicle fire.

[0005] However, existing monitoring and early warning methods for electric vehicle power batteries are insufficient in identifying the abnormal discharge conditions of long-term low current under the above-mentioned static state, making it difficult to accurately capture and provide early warnings. This results in risky vehicles not being monitored and investigated in a timely manner, posing a potential threat to the safe operation of new energy vehicles.

[0006] Therefore, in order to effectively identify the aforementioned long-term abnormal discharge risk conditions, detect vehicles with potential safety hazards in advance and take targeted measures to reduce the incidence of safety accidents caused by battery over-discharge, there is an urgent need for an electric vehicle power battery abnormal discharge early warning identification method that can accurately monitor abnormal changes in key parameters such as power battery current and discharge duration, thereby achieving risk identification and early warning.

[0007] The information disclosed in the background section of this invention is intended only to enhance the understanding of the general background of this invention, and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention

[0008] The purpose of this invention is to propose a method and device for early warning and identification of abnormal discharge of power batteries, which solves the problem that traditional detection methods are insufficient in identifying long-term low-current abnormal discharge conditions under static conditions, making it difficult to accurately capture and provide early warnings, resulting in the inability to monitor and focus on dangerous vehicles in a timely manner. This invention achieves accurate identification of long-term low-current abnormal discharge conditions, and can avoid battery over-discharge and safety accidents in advance.

[0009] To achieve the above objectives, in a first aspect, the present invention proposes a method for early warning and identification of abnormal discharge in power batteries, comprising:

[0010] Extract non-charging data for each vehicle;

[0011] The non-charging data is divided into discharge segments based on the first set condition;

[0012] The abnormal discharge characteristic parameters of each discharge segment are calculated based on the non-charging data of each discharge segment.

[0013] A single batch of vehicle discharge statistics table is constructed based on the abnormal discharge characteristic parameters of each discharge segment and the discharge data at the start and end times.

[0014] Based on the single batch vehicle discharge statistics table and warning conditions, abnormal discharge warnings of power batteries are identified, and discharge warning records are obtained.

[0015] A list of early warning results for each batch is generated based on each discharge early warning record, and the risk level of the discharge early warning record is determined based on the abnormal discharge characteristic parameters.

[0016] Optionally, the extraction of non-charging data for each vehicle includes:

[0017] Extract full battery data for each vehicle;

[0018] Remove invalid data from the full battery data to obtain the full valid battery data;

[0019] The charging data in the full valid battery data is removed to obtain the non-charging data of each vehicle, and the last frame of data of each vehicle is extracted and recorded from the full valid battery data.

[0020] Optionally, the step of dividing the non-charging data into discharge segments based on the first set conditions includes:

[0021] When two adjacent frames of data in the non-charging data meet the first set condition, it is determined that the two adjacent frames of data belong to the same discharge segment.

[0022] The first set conditions include:

[0023] The interval between two adjacent frames of data shall not exceed ten minutes;

[0024] The SOC difference between the next frame and the previous frame is less than or equal to zero.

[0025] Optionally, the abnormal discharge characteristic parameters of each discharge segment include:

[0026] Discharge duration, equivalent discharge current, average discharge current, change in SOC, change in discharge mileage, maximum discharge current, and minimum discharge current.

[0027] Optionally, the warning conditions include:

[0028] The discharge duration exceeds 24 hours;

[0029] Both the equivalent discharge current and the average discharge current do not exceed 1A;

[0030] The change in mileage during the discharge process is 0.

[0031] The state of charge (SOC) at the end of the discharge is 0.

[0032] The maximum current during the discharge process shall not exceed 10A.

[0033] Optionally, determining the risk level of the discharge early warning record based on abnormal discharge characteristic parameters includes:

[0034] If the discharge duration is greater than or equal to 72 hours and the equivalent discharge current is less than or equal to 0.5A, the risk level of the corresponding discharge warning record is determined to be high.

[0035] If the discharge duration is greater than or equal to 72 hours, or the equivalent discharge current is less than or equal to 0.5A, the risk level of the corresponding discharge warning record is determined to be medium.

[0036] If the discharge duration is less than 72 hours and the equivalent discharge current is greater than 0.5A, the risk level of the corresponding discharge warning record is determined to be low.

[0037] Optionally, it also includes:

[0038] The discharge warning record is compared with the last frame of data for the corresponding vehicle;

[0039] When the comparison result meets the second set condition, it is determined that the vehicle corresponding to the discharge warning record is online normally after the warning, and the time of the last frame of data is recorded in the corresponding position of the discharge warning record in the single batch warning result list.

[0040] The discharge warning record with an empty time corresponding to the last frame of data is selected from the single batch warning result list to obtain the final warning result list for this batch.

[0041] The final list of warning results for each batch forms the historical cumulative warning list.

[0042] Optionally, the second setting condition includes:

[0043] The time difference between the last frame of data and the time recorded by the discharge warning is greater than or equal to 1 hour;

[0044] The difference between the mileage in the last frame of data and the mileage recorded by the discharge warning is greater than 0;

[0045] The highest and lowest single-cell voltages in the last frame of data satisfy the following:

[0046] 2.0≤V max ≤4.2 and 2.0≤V min ≤4.2;

[0047] Among them, V max The highest single-cell voltage, V min This is the lowest single-cell voltage.

[0048] Optionally, it also includes:

[0049] At the start of each subsequent batch calculation, the last frame of data for the corresponding vehicle will be compared with the discharge warning record of the corresponding vehicle in the historical cumulative warning list;

[0050] When the comparison result meets the second set condition, the discharge warning record corresponding to the last frame of data is determined to be invalid. This discharge warning record is deleted from the historical cumulative warning list, and the final warning result list corresponding to each batch is merged into the historical cumulative warning list to complete the update of the historical cumulative warning list.

[0051] Secondly, the present invention proposes a power battery abnormal discharge early warning and identification device, comprising:

[0052] The extraction module is used to extract non-charging data from each vehicle;

[0053] A segmentation module is used to segment the non-charging data into discharge segments based on a first set condition.

[0054] The technology module is used to calculate the abnormal discharge characteristic parameters of each discharge segment based on the non-charging data of each discharge segment;

[0055] The module is used to construct a single batch of vehicle discharge statistics summary table based on the abnormal discharge characteristic parameters of each discharge segment and the discharge data at the start and end times;

[0056] The identification module is used to identify abnormal discharge warnings of power batteries based on the single batch vehicle discharge statistics summary table and warning conditions, and to obtain discharge warning records;

[0057] The generation and determination module is used to generate a list of early warning results for a single batch based on each discharge early warning record, and to determine the risk level of the discharge early warning record based on the abnormal discharge characteristic parameters.

[0058] The beneficial effects of this invention are as follows: By extracting non-charging data from each vehicle, focusing on the discharge and idle scenarios that need to be monitored, and eliminating interference data from the charging process, this ensures that subsequent analysis is only for operating conditions that may have abnormal discharge risks, laying a data foundation for accurate anomaly identification. By dividing the non-charging data into discharge segments based on a first set condition, continuous non-charging data is divided into independent discharge segments, transforming complex raw data into basic units that can be analyzed individually, making the characteristics of each discharge process easier to extract and evaluate. By calculating the abnormal discharge characteristic parameters of each discharge segment, the abstract discharge process is transformed into specific, comparable, and analyzable indicators, providing data basis for judging whether the discharge is abnormal. By constructing a single batch vehicle discharge statistical summary table, a summary table containing complete information is formed, facilitating rapid querying and batch analysis. By using early warning conditions, high-risk records that meet the characteristics of long-term idle low-current discharge to depletion are accurately screened from the statistical summary table, achieving accurate identification of long-term low-current abnormal discharge conditions. The screened early warning records are compiled into a list and the risk level is determined, improving the efficiency and targeting of risk response and providing clear guidance for subsequent vehicle monitoring and fault diagnosis. This invention can accurately identify abnormal discharge conditions with low current for extended periods, thereby preventing battery over-discharge and safety accidents in advance, providing reliable protection for power battery safety monitoring, and improving the level of intelligent management of new energy vehicle batteries.

[0059] The system of the present invention has other features and advantages that will be apparent from or will be set forth in detail in the accompanying drawings and following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description

[0060] The above and other objects, features and advantages of the present invention will become more apparent from the accompanying drawings, in which like reference numerals generally denote like parts.

[0061] Figure 1 A flowchart illustrating the steps of a power battery abnormal discharge early warning identification method according to Embodiment 1 of the present invention is shown. Detailed Implementation

[0062] The invention will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0063] like Figure 1 As shown, this embodiment provides a method for identifying abnormal discharge warnings of power batteries, including:

[0064] S1. Extract non-charging data for each vehicle;

[0065] In this step, extracting the non-charging data for each vehicle specifically includes:

[0066] Extract full battery data for each vehicle;

[0067] Remove invalid data from the full battery data to obtain the full valid battery data;

[0068] The charging data in the full valid battery data is removed to obtain the non-charging data of each vehicle, and the last frame of data of each vehicle is extracted and recorded from the full valid battery data.

[0069] Specifically, for each vehicle (uniquely identified by its Vehicle Identification Number VIN), raw data (full battery data) generated during all operational stages of its power battery's entire lifecycle is collected, covering all operating conditions including charging, discharging, and stationary states. Key extracted fields include, but are not limited to: Vehicle Identification Number (VIN), data recording time (time), current vehicle state (e.g., charging, discharging, stationary), remaining battery capacity (SOC), total mileage (ODO), real-time current (current), and highest single-cell voltage (V). max ), minimum single cell voltage (V) min ) and the battery number (V) corresponding to the lowest single cell voltage. minnum This step is fundamental to data processing, aiming to obtain complete and comprehensive raw data to provide data support for subsequent screening, analysis, and early warning, ensuring that no information that may be related to abnormal discharge is overlooked.

[0070] The extracted full data is cleaned to remove data that is meaningless or contains obvious errors, ensuring the accuracy of subsequent analysis. Specific removal rules include:

[0071] Remove zero-value data: When the vehicle state is 0, or the highest single-cell voltage (V) is... max ) and minimum single cell voltage (V minIf both values ​​are 0, the data is considered invalid (possibly due to equipment malfunction, signal loss, etc.).

[0072] Remove null data: When the vehicle state is empty, or the highest single-cell voltage (V) is not specified. max ) and minimum single cell voltage (V min If both are empty, the data is considered invalid (due to the lack of key parameters, it cannot reflect the true state of the battery).

[0073] Through the above processing, the retained full amount of valid battery data can accurately and reliably reflect the operating status of the power battery, laying the foundation for subsequent precise analysis.

[0074] Data from non-charging conditions is filtered out from the full range of valid battery data, focusing on scenarios where abnormal discharge risks may exist. For example, based on the vehicle state field, data with a state value of 110 and 170 (110 for fast charging and 170 for slow charging) are removed, while data with other state values ​​are retained, i.e., battery data in the vehicle's discharging (e.g., power consumption during driving) and stationary (e.g., parked and unused) states. This step eliminates interference from the charging process in abnormal discharge analysis, allowing subsequent analysis resources to focus on the target conditions and improving the targeting of risk identification.

[0075] For each vehicle (each VIN), the record with the longest time frame (i.e., the most recently generated valid data for that vehicle) is selected from its full battery data set. Key information from this frame is recorded, including the vehicle identification number (VIN), recording time, remaining battery capacity (SOC), total mileage (ODO), and highest single-cell voltage (V). max ), minimum single cell voltage (V) min )wait.

[0076] The core function of this data is to serve as a key basis for determining whether a vehicle has returned to normal operation after a warning. By comparing it with parameters such as time, mileage, and voltage in the warning record, it verifies whether the vehicle started and drove normally after the warning and whether the battery status returned to normal, thereby eliminating invalid warnings and ensuring the timeliness and accuracy of the warning results. At the same time, in multiple batches of data updates, it provides the latest vehicle status reference for the dynamic updating of the historical warning list.

[0077] S2. Divide the non-charging data into discharge segments based on the first set conditions;

[0078] In this step, dividing the non-charging data into discharge segments based on the first set conditions specifically includes:

[0079] When two adjacent frames of data in the non-charging data meet the first set condition, it is determined that the two adjacent frames of data belong to the same discharge segment.

[0080] The first set of conditions includes:

[0081] The interval between two adjacent frames of data shall not exceed ten minutes;

[0082] The SOC difference between the next frame and the previous frame is less than or equal to zero.

[0083] Specifically, dividing the non-charging data into discharge segments based on the first set condition is a key step in grouping continuous non-charging state data according to the correlation of the discharge process. Its core logic and function are as follows:

[0084] After acquiring the non-charging data (i.e., discharge and idle data excluding charging status) of each vehicle, the data frames need to be compared one by one in chronological order. When two adjacent data frames simultaneously meet the first set condition, they are determined to belong to the same discharge segment; if not, a discharge segment is divided with the previous data frame as the endpoint, and the next data frame is used as the starting frame of the new discharge segment.

[0085] Among them, the two elements of the first setting condition define the correlation of the discharge process from the two dimensions of time continuity and the characteristics of charge change:

[0086] The interval between two adjacent data frames should not exceed ten minutes: This condition ensures the temporal tightness of the two data frames and avoids misjudging discontinuous processes caused by long interruptions as the same discharge segment. For example, if two data frames are 8 minutes apart when the vehicle is stationary, it means that there is no obvious interruption in the discharge process and they can be classified as the same stage; if the interval is 15 minutes, it is considered that there may be a state change in between (such as the vehicle being briefly awakened and then stationary again), and it needs to be divided into different discharge segments.

[0087] The SOC difference between the next frame and the previous frame is less than or equal to zero: a SOC (remaining battery power) difference ≤ 0 means that the battery power has not increased, which is consistent with the battery power change pattern during discharge or rest (the battery power decreases during discharge, and may remain unchanged or decrease slightly during rest). If the SOC difference is positive (i.e., the battery power increases), there may be abnormal charging or data errors. A new discharge segment needs to be defined based on this, ensuring that the battery power change within each discharge segment conforms to the basic characteristics of the non-charging state.

[0088] This division method breaks down the originally continuous non-charging data into independent and intrinsically related discharge segments. Each discharge segment represents a continuous and stable non-charging process, providing a structured analysis unit for subsequent calculation of the characteristic parameters of a single discharge segment (such as discharge duration and current changes), making it possible to accurately identify abnormal discharge processes.

[0089] S3. Calculate the abnormal discharge characteristic parameters of each discharge segment based on the non-charging data of each discharge segment;

[0090] In this step, the abnormal discharge characteristic parameters of each discharge segment include:

[0091] Discharge duration, equivalent discharge current, average discharge current, change in SOC, change in discharge mileage, maximum discharge current, and minimum discharge current.

[0092] Specifically, calculating abnormal discharge characteristic parameters based on non-charging data for each discharge segment is a crucial step in transforming structured discharge segment data into quantifiable analytical indicators. By extracting features from dimensions such as time, current, charge, and mileage during the discharge process, specific data support is provided for subsequent anomaly identification. The abnormal discharge characteristic parameters for each discharge segment include discharge duration, equivalent discharge current, average discharge current, SOC change, mileage change during the discharge process, maximum discharge current, and minimum discharge current. The calculation methods for each parameter are as follows:

[0093] Discharge duration: Based on the divided discharge segment information, the time (start time) of the first frame of data and the time (end time) of the last frame of data within each discharge segment are extracted, and the duration is calculated using the time difference. The calculation formula is:

[0094] t_diff i =t_end i -t_sta i , i = 1 ~ n (1)

[0095] In the formula, t_diff i t_end represents the discharge duration of the i-th discharge segment, in hours (h). i t_sta is the end time of the i-th discharge segment; i : The start time of the i-th discharge segment.

[0096] Equivalent discharge current: The equivalent discharge current I is defined as the number of ampere-hours discharged per unit time. equ The discharge rate is used to characterize the discharge rate of a battery during a certain discharge period, serving as a key indicator for evaluating whether the discharge process is abnormal. It is obtained by calculating the integral of the current over time within the discharge period and then dividing by the discharge duration. Specifically, the current value of each frame of data in the discharge period is accumulated and summed with the corresponding time interval (i.e., ∫Idt, equivalent to the total discharge ampere-hours), and then the sum is divided by the discharge duration of that period. The result, expressed in amperes (A), characterizes the actual discharge rate per unit time, comprehensively reflecting the overall intensity of the discharge process. The calculation formula is:

[0097]

[0098] In the formula, I equ I represents the equivalent discharge current for each discharge segment, in A, and I represents the magnitude of the discharge current at each moment in each discharge segment, in A.

[0099] Average discharge current: Calculated by arithmetic averaging of the current values ​​across all frames within a discharge segment. This involves summing all current data within the segment and dividing by the number of data frames (i.e., the number of discharge moments). The result is expressed in amperes (A) and reflects the average current magnitude during the discharge segment. The calculation formula is:

[0100] I avg =average(I j ),j=j~n(3)

[0101] In the formula, I avg The average discharge current for each discharge segment, in amperes (A); I j The value of the discharge current at time j in each discharge segment is expressed in A; n is the number of discharge times in the discharge segment.

[0102] Change in soc: This is obtained by subtracting the remaining charge at the beginning of the discharge segment (soc_sta) from the remaining charge at the end of the discharge segment (soc_end). The formula is: Change in soc = soc_end - soc_sta. The result is usually negative (because the charge decreases during the discharge process) and is used to reflect the magnitude of the change in battery charge during that discharge segment.

[0103] Mileage change during discharge: Calculated by subtracting the cumulative mileage at the start of the discharge segment from the cumulative mileage at the end of the discharge segment (odo_end). The formula is: Mileage change = odo_end - odo_sta, in kilometers (km). This value is used to determine whether the vehicle is in motion during the discharge process (0 indicates stationary discharge).

[0104] Maximum current during discharge: The maximum value is selected from the current values ​​of all frames of data within the discharge segment, and the unit is amperes (A). It is used to reflect the peak current that occurs during the discharge process of this segment.

[0105] Minimum current during discharge: The minimum value is selected from the current values ​​of all frames of data within the discharge segment. The unit is amperes (A). It is used to reflect the current valley value that occurs during the discharge segment, especially in the static discharge scenario, it can reflect the degree of weak power consumption.

[0106] Through the above calculations, the core characteristics of each discharge segment are quantified by the system. These parameters together constitute a multi-dimensional index system describing the discharge process, providing accurate data for subsequent identification of abnormal discharge conditions (such as long-term, low-current, and static discharge) based on early warning conditions.

[0107] S4. Construct a single batch vehicle discharge statistics table based on the abnormal discharge characteristic parameters of each discharge segment and the discharge data at the start and end times.

[0108] Specifically, constructing a single batch of vehicle discharge statistics table based on the abnormal discharge characteristic parameters of each discharge segment and the discharge data at the start and end times is a process of systematically integrating the discharge segment data. The aim is to form a summary table covering all dimensions of the discharge process, providing a standardized data carrier for subsequent early warning identification.

[0109] The specific construction method is as follows:

[0110] Each independent discharge segment is recorded as a separate row. The core information of that discharge segment is included in the table in a uniform format, and the content is divided into three main categories:

[0111] The raw data for the start and end times of the discharge segment includes: the start time (time_sta), the end time (time_end); the vehicle status at the start time (state_sta), the vehicle status at the end time (state_end); the remaining battery power at the start time (soc_sta), the remaining battery power at the end time (soc_end); the cumulative mileage at the start time (odo_sta), the cumulative mileage at the end time (odo_end); the current at the start time (current_sta), the current at the end time (current_end); the highest single-cell voltage at the start time (vmax_sta), the highest single-cell voltage at the end time (vmax_end); the lowest single-cell voltage at the start time (vmin_sta), the lowest single-cell voltage at the end time (vmin_end); and the battery number corresponding to the lowest single-cell voltage at the start time (vminnum_sta), the battery number corresponding to the lowest single-cell voltage at the end time (vminnum_end). This data is directly derived from the raw records of the first and last frames of the discharge segment and is used to visually present the initial and final states of the discharge process.

[0112] The calculated abnormal discharge characteristic parameters include discharge duration, equivalent discharge current, average discharge current, change in SOC, change in mileage during discharge, maximum current during discharge, and minimum current during discharge. These parameters are quantitative results of the characteristics of the discharge process, reflecting key information such as the intensity, duration, and changes in charge and mileage of the discharge process from different dimensions.

[0113] By integrating the two types of data according to the discharge segments one by one, the resulting single-batch vehicle discharge statistics table has a clear structured feature. Each row of records completely corresponds to a discharge segment, including not only the specific state data of the start and end of the discharge process, but also the core characteristic parameters in the process, realizing a centralized presentation of the entire chain of information of a single discharge segment from "start-process-end".

[0114] The construction of this master table provides a unified comparison benchmark for all discharge segments within a single batch, facilitating quick querying and batch analysis of differences between different discharge segments. On the other hand, it also lays a data foundation for subsequent screening of abnormal discharge records based on early warning conditions, enabling early warning identification to be carried out directly based on the integrated structured data, thereby improving analysis efficiency and accuracy.

[0115] S5. Based on the single batch vehicle discharge statistics table and warning conditions, identify abnormal discharge warnings of the power battery and obtain discharge warning records.

[0116] In this step, the warning conditions include:

[0117] The discharge duration exceeds 24 hours;

[0118] Both the equivalent discharge current and the average discharge current do not exceed 1A;

[0119] The change in mileage during the discharge process is 0.

[0120] The state of charge (SOC) at the end of the discharge is 0.

[0121] The maximum current during the discharge process shall not exceed 10A.

[0122] Specifically, identifying abnormal discharge warnings for power batteries based on a single batch of vehicle discharge statistics and preset warning conditions is a key step in accurately locating high-risk operating conditions from the integrated discharge segment data. Its core logic involves cross-validating multiple conditions to filter out abnormal discharge records that meet the characteristics of prolonged static low-current discharge leading to battery depletion. The specific identification process is as follows:

[0123] Using each row (i.e., each discharge segment) in the single batch vehicle discharge statistics table as an analysis unit, we verify one by one whether the discharge segment simultaneously meets all warning conditions:

[0124] Discharge duration exceeds 24 hours: that is, the value of the "Discharge Duration" field in the summary table is ≥24 hours, ensuring that the discharge process is continuous and excluding normal scenarios of short-term discharge.

[0125] The equivalent discharge current and the average discharge current both do not exceed 1A: The "equivalent discharge current" ≤ 1A and the "average discharge current" ≤ 1A in the summary table. This condition locks in the small current discharge characteristics, which is consistent with the weak power consumption mode in the static state and excludes the large current discharge situation during normal driving.

[0126] Discharge process mileage change is 0: The value of the "Discharge process mileage change" field in the summary table is 0, which means that the vehicle did not move during the discharge, further confirming that the vehicle was in a stationary state, rather than a normal discharge while driving.

[0127] A soc value of 0 at the end of discharge: The value of the "soc_end" field (remaining charge at the end of discharge) in the summary table is 0, which indicates that the battery has been completely discharged and there is a serious risk of over-discharge. This is one of the core characteristics of abnormal discharge.

[0128] The maximum current during the discharge process does not exceed 10A: The value of the "maximum current during discharge process" field in the summary table is ≤10A, which verifies again that there is no obvious large current fluctuation during the discharge process, ensuring that the entire discharge process is always in a low current state, and eliminating normal power consumption interference in a short period of time (such as temporary vehicle wake-up).

[0129] When a discharge segment simultaneously meets all the above conditions, it is determined to be an abnormal discharge condition requiring attention, and is extracted from the master table to form a discharge warning record. Each warning record contains complete information about the discharge segment in the master table, including the vehicle identification number (VIN), discharge start and end time, and various characteristic parameters, providing a precise target for subsequent risk level classification and vehicle investigation.

[0130] This multi-condition superposition identification method can effectively filter out normal discharge processes and accurately capture high-risk operating conditions characterized by "long duration, static placement, low current, and complete discharge," providing a direct and reliable basis for power battery safety early warning.

[0131] S6. Generate a list of early warning results for each batch based on each discharge early warning record, and determine the risk level of the discharge early warning record based on the abnormal discharge characteristic parameters.

[0132] In this step, determining the risk level of the discharge warning record based on abnormal discharge characteristic parameters specifically includes:

[0133] If the discharge duration is greater than or equal to 72 hours and the equivalent discharge current is less than or equal to 0.5A, the risk level of the corresponding discharge warning record is determined to be high.

[0134] If the discharge duration is greater than or equal to 72 hours, or the equivalent discharge current is less than or equal to 0.5A, the risk level of the corresponding discharge warning record is determined to be medium.

[0135] If the discharge duration is less than 72 hours and the equivalent discharge current is greater than 0.5A, the risk level of the corresponding discharge warning record is determined to be low.

[0136] Specifically, generating a batch of early warning results based on each discharge early warning record and determining the risk level according to abnormal discharge characteristic parameters is a process of structuring and risk-stratifying the identified abnormal discharge conditions, providing clear guidance for subsequent targeted handling. The specific operation is as follows:

[0137] Generate a single batch of early warning results list: All discharge early warning records filtered by the early warning conditions are summarized to form a single batch of early warning results list. The list contains the core information of each early warning record, such as the vehicle identification number (vin), early warning time (i.e., discharge end time time_end), mileage at the end time (odo_end), the lowest single cell voltage number at the end time (vminnum_end), as well as key abnormal discharge characteristic parameters such as discharge duration, equivalent discharge current, and average discharge current, comprehensively reflecting the basic situation of abnormal discharge for each vehicle.

[0138] Risk levels are determined based on abnormal discharge characteristic parameters: using "discharge duration" and "equivalent discharge current" from the discharge warning record as core judgment indicators, risk levels are classified according to the following rules:

[0139] High-risk level: When the discharge duration recorded by the discharge warning is ≥72 hours and the equivalent discharge current is ≤0.5A, it is judged as high-risk. This situation means that the vehicle has been continuously discharging with an extremely weak current for an extremely long time (such as more than 3 days), the risk of battery over-discharge is extremely high, and because the discharge process is highly concealed, it may have already caused serious damage to the battery, requiring priority emergency investigation.

[0140] Medium Risk Level: A vehicle is classified as medium risk if its discharge warning record meets either the criteria of "discharge duration ≥ 72 hours" or "equivalent discharge current ≤ 0.5A". For example, a vehicle with a discharge duration of 80 hours but an equivalent discharge current of 0.6A, or a discharge duration of 50 hours but an equivalent discharge current as low as 0.3A, both fall under the medium risk category. While these situations do not reach the severity of high risk, they still present a significant risk of over-discharge and require inspection at the next lower priority level.

[0141] Low-risk level: When the discharge duration recorded in the discharge warning is less than 72 hours and the equivalent discharge current is greater than 0.5A, it is judged as low-risk. This type of abnormal discharge has a relatively short duration and the discharge current does not reach an extremely low level. The urgency and severity of battery over-discharge are low, and it can be included in the routine monitoring and investigation plan.

[0142] Through the above methods, the single-batch early warning result list not only integrates all abnormal discharge records that need attention, but also, through the clear division of risk levels, allows for differentiated treatment of potential hazards of different severity, significantly improving the efficiency and accuracy of risk handling and providing an actionable guideline for the safety management of power batteries.

[0143] The method in this embodiment further includes: comparing the discharge warning record with the last frame of data of the corresponding vehicle;

[0144] When the comparison result meets the second set condition, it is determined that the vehicle corresponding to the discharge warning record is online normally after the warning, and the time of the last frame of data is recorded in the corresponding position of the discharge warning record in the single batch warning result list.

[0145] The discharge warning record with an empty time corresponding to the last frame of data is selected from the single batch warning result list to obtain the final warning result list for this batch.

[0146] The final list of warning results for each batch forms the historical cumulative warning list.

[0147] In this step, the second set of conditions includes:

[0148] The time difference between the last frame of data and the time recorded by the discharge warning is greater than or equal to 1 hour;

[0149] The difference between the mileage in the last frame of data and the mileage recorded by the discharge warning is greater than 0;

[0150] The highest and lowest single-cell voltages in the last frame of data satisfy the following:

[0151] 2.0≤V max ≤4.2 and 2.0≤V min ≤4.2;

[0152] Among them, V max The highest single-cell voltage, V min This is the lowest single-cell voltage.

[0153] Specifically, comparing the discharge warning record with the last frame of data of the corresponding vehicle (obtained in S1) is a further verification of the effectiveness of the warning record, thereby filtering out vehicles that truly pose a continuous risk. The specific process is as follows:

[0154] First, for each discharge warning record in the single batch warning result list, the last frame of data for the corresponding vehicle (same VIN) is retrieved and compared. If the comparison result simultaneously meets the second set condition, the vehicle is determined to have returned to normal operation after the warning. The second set condition includes:

[0155] The time difference between the last frame of data and the time recorded by the discharge warning (warning time) is ≥1 hour to ensure that the vehicle has enough time to return to normal after the warning and to eliminate the interference of instantaneous data fluctuations;

[0156] The difference between the mileage in the last frame of data and the mileage recorded in the discharge warning (mileage at the end time) is >0, indicating that the vehicle actually drove after the warning, verifying that the vehicle has left the abnormal discharge state.

[0157] The highest single cell voltage (V) in the last frame of data max ) and minimum single cell voltage (V min ) is in the range of 2.0≤V max ≤4.2 and 2.0≤V min A reading of ≤4.2 indicates that the battery voltage has returned to normal and there is no risk of abnormal high or low voltage.

[0158] If the above conditions are met, the time of the last frame of data will be recorded at the corresponding position of the discharge warning record in the single batch warning result list, and marked as having been normally online; if not met, the position will be empty.

[0159] Subsequently, discharge warning records with empty timestamps corresponding to the last frame of data are selected from the single batch warning result list. These records indicate that the vehicle did not go online normally after the warning, and the abnormal risk may continue, thus forming the final warning result list for this batch.

[0160] By summarizing the final warning results of each batch, a historical cumulative warning list is formed. This list dynamically reflects the situation of vehicles with persistent abnormal discharge risks in each batch, providing a complete basis for long-term tracking and management of vehicle risks.

[0161] The method in this embodiment further includes: at the beginning of each subsequent batch calculation, comparing the last frame data of the corresponding vehicle with the discharge warning record of the corresponding vehicle in the historical cumulative warning list;

[0162] When the comparison result meets the second set condition, the discharge warning record corresponding to the last frame of data is determined to be invalid. The discharge warning record is deleted from the historical cumulative warning list, and the final warning result list corresponding to each batch is merged into the historical cumulative warning list to complete the update of the historical cumulative warning list.

[0163] Specifically, updating the historical cumulative warning list at the start of each subsequent batch calculation is a crucial step to ensure that the list always reflects the latest risk status of vehicles. The specific process is as follows:

[0164] First, for each discharge warning record in the historical cumulative warning list, the last frame of data extracted from the corresponding vehicle (same VIN) during the current batch calculation is retrieved and compared. When the comparison result simultaneously meets the second set condition, the discharge warning record is determined to be invalid. After completing the invalidation of historical warning records, the final warning result list obtained from the current batch calculation is merged with the updated historical cumulative warning list to form a new historical cumulative warning list.

[0165] In this way, the historical cumulative warning list can dynamically remove vehicle records that have returned to normal, while incorporating new batches of valid warning records, thus maintaining accurate tracking of the risk of abnormal vehicle discharge and providing a reliable and timely basis for long-term vehicle safety management.

[0166] Example 2

[0167] This embodiment provides a power battery abnormal discharge early warning and identification device, including:

[0168] The extraction module is used to extract non-charging data from each vehicle;

[0169] A segmentation module is used to segment the non-charging data into discharge segments based on a first set condition.

[0170] The technology module is used to calculate the abnormal discharge characteristic parameters of each discharge segment based on the non-charging data of each discharge segment;

[0171] The module is used to construct a single batch of vehicle discharge statistics summary table based on the abnormal discharge characteristic parameters of each discharge segment and the discharge data at the start and end times;

[0172] The identification module is used to identify abnormal discharge warnings of power batteries based on the single batch vehicle discharge statistics summary table and warning conditions, and to obtain discharge warning records;

[0173] The generation and determination module is used to generate a list of early warning results for a single batch based on each discharge early warning record, and to determine the risk level of the discharge early warning record based on the abnormal discharge characteristic parameters.

[0174] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A method for early warning and identification of abnormal discharge in power batteries, characterized in that, include: Extract non-charging data for each vehicle; The non-charging data is divided into discharge segments based on a first set condition, wherein the first set condition includes: the interval between two adjacent frames of data does not exceed ten minutes; and the SOC difference between the next frame of data and the previous frame of data is less than or equal to zero. Abnormal discharge characteristic parameters for each discharge segment are calculated based on the non-charging data of each discharge segment. These abnormal discharge characteristic parameters include: discharge duration, equivalent discharge current, average discharge current, SOC change, discharge process mileage change, maximum discharge current, and minimum discharge current. A single batch of vehicle discharge statistics table is constructed based on the abnormal discharge characteristic parameters of each discharge segment and the discharge data at the start and end times. Based on the single batch vehicle discharge statistics table and warning conditions, abnormal discharge warnings of power batteries are identified, and discharge warning records are obtained. The warning conditions include: discharge duration exceeding 24 hours; equivalent discharge current and average discharge current not exceeding 1A; mileage change during discharge is 0; SOC at the end of discharge is 0; and maximum current during discharge does not exceed 10A. A list of early warning results for each batch is generated based on each discharge early warning record, and the risk level of the discharge early warning record is determined based on the abnormal discharge characteristic parameters.

2. The method for early warning and identification of abnormal discharge of power battery according to claim 1, characterized in that, The extraction of non-charging data for each vehicle includes: Extract full battery data for each vehicle; Remove invalid data from the full battery data to obtain the full valid battery data; The charging data in the full valid battery data is removed to obtain the non-charging data of each vehicle, and the last frame of data of each vehicle is extracted and recorded from the full valid battery data.

3. The method for early warning and identification of abnormal discharge of power battery according to claim 1, characterized in that, The risk level determination of discharge early warning records based on abnormal discharge characteristic parameters includes: If the discharge duration is greater than or equal to 72 hours and the equivalent discharge current is less than or equal to 0.5A, the risk level of the corresponding discharge warning record is determined to be high. If the discharge duration is greater than or equal to 72 hours, or the equivalent discharge current is less than or equal to 0.5A, the risk level of the corresponding discharge warning record is determined to be medium. If the discharge duration is less than 72 hours and the equivalent discharge current is greater than 0.5A, the risk level of the corresponding discharge warning record is determined to be low.

4. The method for early warning and identification of abnormal discharge of power battery according to claim 2, characterized in that, Also includes: The discharge warning record is compared with the last frame of data for the corresponding vehicle; When the comparison result meets the second set condition, it is determined that the vehicle corresponding to the discharge warning record is online normally after the warning, and the time of the last frame of data is recorded in the corresponding position of the discharge warning record in the single batch warning result list. The discharge warning record with an empty time corresponding to the last frame of data is selected from the single batch warning result list to obtain the final warning result list for this batch. The final list of warning results for each batch forms the historical cumulative warning list.

5. The method for early warning and identification of abnormal discharge of power battery according to claim 4, characterized in that, The second setting conditions include: The time difference between the last frame of data and the time recorded by the discharge warning is greater than or equal to 1 hour; The difference between the mileage in the last frame of data and the mileage recorded by the discharge warning is greater than 0; The highest and lowest single-cell voltages in the last frame of data satisfy the following: 2.0≤V max ≤4.2 and 2.0≤V min ≤4.2; Among them, V max The highest single-cell voltage, V min This is the lowest single-cell voltage.

6. The method for early warning and identification of abnormal discharge of power battery according to claim 5, characterized in that, Also includes: At the start of each subsequent batch calculation, the last frame of data for the corresponding vehicle will be compared with the discharge warning record of the corresponding vehicle in the historical cumulative warning list; When the comparison result meets the second set condition, the discharge warning record corresponding to the last frame of data is determined to be invalid. This discharge warning record is deleted from the historical cumulative warning list, and the final warning result list corresponding to each batch is merged into the historical cumulative warning list to complete the update of the historical cumulative warning list.

7. A power battery abnormal discharge early warning and identification device, characterized in that, include: The extraction module is used to extract non-charging data from each vehicle; The segmentation module is used to segment the non-charging data into discharge segments based on a first set condition, wherein the first set condition includes: the interval between two adjacent frames of data does not exceed ten minutes; and the SOC difference between the next frame of data and the previous frame of data is less than or equal to zero. The technical module is used to calculate the abnormal discharge characteristic parameters of each discharge segment based on the non-charging data of each discharge segment. The abnormal discharge characteristic parameters include: discharge duration, equivalent discharge current, average discharge current, SOC change, discharge process mileage change, discharge process maximum current and discharge process minimum current. The module is used to construct a single batch of vehicle discharge statistics summary table based on the abnormal discharge characteristic parameters of each discharge segment and the discharge data at the start and end times; The identification module is used to identify abnormal discharge warnings of power batteries based on the single batch vehicle discharge statistics table and warning conditions, and obtain discharge warning records. The warning conditions include: discharge duration exceeding 24 hours; equivalent discharge current and average discharge current not exceeding 1A; mileage change during discharge process is 0; SOC at the end of discharge is 0; maximum current during discharge process does not exceed 10A. The generation and determination module is used to generate a list of early warning results for a single batch based on each discharge early warning record, and to determine the risk level of the discharge early warning record based on the abnormal discharge characteristic parameters.

Images