Power battery safety early warning method, device, equipment and medium

Abstract

The application belongs to the technical field of battery safety control, and discloses a power battery safety early warning method, device, equipment and medium, wherein the method comprises the following steps: acquiring voltage data and SOC data of the power battery in a power battery early warning period, wherein the voltage data comprises voltage data of each single body; extracting voltage data of each single body according to different SOC intervals to obtain voltage data sets of each single body in different SOC intervals; calculating the difference of each single body in each SOC interval; and calculating the consistency of each single body in each SOC interval; if the difference and the consistency both satisfy preset conditions, it is determined that the power battery has a safety risk. The method can identify the battery safety risk in advance, does not need complex modeling and accurate parameter extraction, is simple and easy to deploy, can comprehensively evaluate the battery state, and improves the early warning accuracy and reliability.

Description

Technical Field

[0001] This invention relates to the field of battery safety control technology, and in particular to a method, device, equipment and medium for early warning of power battery safety. Background Technology

[0002] In recent years, with the widespread application of high-power, high-density battery systems in electric vehicles and energy storage power stations, the safety of power batteries has received high attention. During long-term use, lithium-ion batteries are prone to imbalances in parameters (such as voltage and resistance) between cells due to factors such as differences in manufacturing consistency, different local aging rates, and uneven heat dissipation. This can ultimately lead to serious accidents such as thermal runaway, fire, or even explosion.

[0003] Traditional power battery safety testing methods mainly rely on static threshold comparison (e.g., alarms are triggered when the voltage of a single cell exceeds or falls below a set value). However, this method cannot identify chronic deviations and early evolution trends, and alarms are easily triggered only at the critical point of thermal runaway, resulting in delayed warnings.

[0004] Furthermore, while some existing methods employ modeling to analyze cell status, most require complex data-driven modeling or precise extraction of cell parameters, making large-scale application in practical deployments difficult. Therefore, a simple, effective, and highly deployable method for early warning of power battery faults is urgently needed. Summary of the Invention

[0005] The main objective of this invention is to provide a method, device, equipment, and medium for safety early warning of power batteries, aiming to solve the technical problems of delayed early warning and difficult deployment in existing power battery early warning methods.

[0006] To achieve the above-mentioned objectives, the first aspect of this invention provides a power battery safety early warning method, the method comprising:

[0007] The voltage data and SOC (State of Charge) data of the power battery are acquired within a power battery warning cycle, wherein the voltage data includes the voltage data of each individual cell;

[0008] The voltage data of each individual unit is extracted according to different SOC intervals to obtain voltage datasets of each individual unit in different SOC intervals;

[0009] Calculate the difference between each of the said cells within each of the said SOC ranges, where the difference refers to the degree of deviation of the voltage of one cell from that of other cells within the same SOC range; and,

[0010] Calculate the consistency of each of the aforementioned cells within each of the aforementioned SOC intervals. The consistency refers to the degree to which the voltage variation pattern of a cell within different SOC intervals matches that of other cells.

[0011] If both the difference and the consistency meet the preset conditions, then the power battery is determined to have a safety risk.

[0012] Further, after extracting the voltage data of each individual unit according to different SOC intervals to obtain the voltage dataset of each individual unit in different SOC intervals, the process includes:

[0013] Calculate the average voltage of each of the aforementioned cells in different SOC ranges;

[0014] Within each SOC interval, the average voltage values ​​of each individual cell are concatenated to obtain a multidimensional average voltage vector of the individual cell corresponding to each SOC interval.

[0015] Further, the differences between each of the said monomers within each of the said SOC intervals are calculated, including:

[0016] Normalize each element in the average voltage vector of the individual cells corresponding to each SOC interval to obtain the normalized value of each individual cell in each SOC interval.

[0017] Based on the normalized values ​​of each monomer in each SOC interval, the differences of each monomer in different SOC intervals are determined.

[0018] Further, the elements in the average voltage vector of the individual cells corresponding to each SOC interval are normalized to obtain the normalized value of each individual cell in each SOC interval, including:

[0019] Normalize each element in the voltage mean vector of the individual cells corresponding to each SOC interval to obtain the normalized value of each individual cell in each SOC interval, including:

[0020] The normalized value is calculated using an improved Z-score normalization algorithm, and the formula is as follows:

[0021]

[0022] in, This represents the improved normalized value of monomer i within the SOC interval j. Let be the average voltage of cell i within the SOC range j. This represents the average voltage of all cells within the SOC range j. Let j be the standard deviation of the average voltage of all cells within the SOC range. This is the working condition correction factor. is the coefficient of variation of the average voltage of all cells within the SOC interval j.

[0023] Further, calculating the consistency of each of the said monomers within each of the said SOC intervals includes:

[0024] For each of the said entities, its normalized value in different SOC intervals is obtained and formed into a normalized multidimensional vector;

[0025] Multiply the elements of the normalized multidimensional vector pairwise to obtain the first vector product matrix corresponding to each individual entity. This vector product matrix is ​​an upper triangular matrix, and is calculated only when x ≤ y. , Let x be the normalized value of monomer i in the SOC interval. Let y be the normalized value of monomer i in the SOC interval y, where i is the monomer number and x and y are the SOC interval numbers.

[0026] The consistency of each individual entity within each SOC interval is determined by the first vector product matrix corresponding to each individual entity.

[0027] Further, calculating the consistency of each of the said monomers within each of the said SOC intervals includes:

[0028] Multiply the normalized values ​​of each individual in different SOC intervals pairwise to obtain a second vector product matrix;

[0029] The consistency of the individual entity within each SOC interval is determined by the second vector product matrix.

[0030] Furthermore, acquiring the voltage data and SOC data of the power battery within a power battery warning cycle includes:

[0031] The battery management system of the power battery acquires the total current of the battery pack, the initial SOC data, and the initial voltage data of each individual cell within a power battery warning cycle.

[0032] The initial SOC data corresponding to a total current greater than or equal to a current threshold and the initial voltage data of each individual cell are filtered to obtain the voltage data and SOC data.

[0033] A second aspect of the present invention provides a power battery safety warning device, comprising:

[0034] The acquisition unit is used to acquire the voltage data and SOC data of the power battery within a power battery warning cycle, wherein the voltage data includes the voltage data of each individual cell;

[0035] The extraction unit is used to extract the voltage data of each of the individual units according to different SOC intervals to obtain the voltage dataset of each individual unit in different SOC intervals.

[0036] The first calculation unit is used to calculate the differences of each of the said monomers within each of the said SOC intervals;

[0037] The second calculation unit is used to calculate the consistency of each of the said individual units within each of the said SOC intervals;

[0038] The determination unit is used to determine that the power battery has a safety risk if both the difference and the consistency meet preset conditions.

[0039] A third aspect of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the power battery safety warning method described in any of the above claims.

[0040] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the power battery safety warning method described in any of the preceding claims.

[0041] Beneficial effects:

[0042] The power battery safety early warning method, device, equipment, and medium of this invention differ from traditional methods that rely on static threshold comparisons. This method, by calculating the differences and consistency of individual cells within different SOC ranges, can capture the chronic deviations and early evolution trends of parameters between cells, avoiding triggering alarms only at the thermal runaway critical point and achieving early warning, thus providing more time to address battery safety issues. Compared to existing methods that employ complex data-driven modeling or require precise cell parameter extraction, this method only needs to acquire battery voltage and SOC data. Safety early warning can be achieved through simple steps such as data extraction and difference and consistency calculations, eliminating the need for complex modeling processes and precise parameter extraction. It is highly deployable and easy to apply on a large scale in practice. By considering the differences and consistency of individual cells simultaneously across different SOC ranges, the state of the power battery can be assessed more comprehensively and meticulously, covering various potential situations of parameter imbalance between cells, thus improving the accuracy and reliability of safety early warning. Attached Figure Description

[0043] Figure 1 A flowchart illustrating a power battery safety warning method according to an embodiment of the invention;

[0044] Figure 2 A schematic diagram of the structure of a power battery safety warning device according to an embodiment of the invention;

[0045] Figure 3 This is a schematic diagram of the structure of a computer device according to an embodiment of the invention.

[0046] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0048] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of features, integers, steps, operations, elements, modules, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, modules, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any modules and all combinations of one or more associated listed items.

[0049] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0050] Reference Figure 1 This invention provides a power battery safety early warning method, the method comprising:

[0051] S1: Obtain the voltage data and SOC data of the power battery within a power battery warning cycle, wherein the voltage data includes the voltage data of each individual cell.

[0052] A power battery warning cycle refers to a pre-set time period for collecting and analyzing battery data to monitor battery safety, such as 3 days or 15 days. Voltage data is the collection of voltage values ​​for each individual cell in the battery. SOC (State of Charge) data is the battery's state of charge data, reflecting the remaining battery capacity. The battery management system (BMS) of an electric vehicle collects voltage data and corresponding SOC data for each individual cell within a warning cycle (e.g., 3 consecutive days). For example, taking a ternary lithium battery pack of a certain brand of electric vehicle as an example, the BMS collects the voltage (accuracy ±0.01V) and SOC (accuracy ±1%) of each cell at a frequency of 1Hz for 3 consecutive days to obtain the voltage and SOC data for that cycle. Obtaining continuous and comprehensive battery operation data provides fundamental data support for subsequent analysis of the characteristics of each individual cell in different SOC ranges.

[0053] S2: Extract the voltage data of each individual unit according to different SOC intervals to obtain voltage datasets of each individual unit in different SOC intervals.

[0054] SOC intervals are continuous sub-intervals into which the SOC range (0-100%) is divided. According to a preset SOC interval division rule (e.g., each 10% interval is divided into 10 intervals such as 0-10%, 10-20%, etc.), voltage data for each cell within each SOC interval is extracted from the data obtained in step S1, forming voltage datasets for each cell in different SOC intervals. For example, if the SOC is divided into 10 intervals (0-10%, 10-20%, ..., 90-100%), the voltage and SOC data for a specific cell collected in step S1 are filtered to select all voltage data for that cell when the SOC is between 0-10%, forming the voltage dataset for that cell in the 0-10% SOC interval. Similarly, voltage datasets for other SOC intervals are obtained. Classifying battery operation data by SOC intervals facilitates targeted analysis of the voltage characteristics of battery cells at different charge levels, providing categorized data for subsequent calculations of differences and consistency.

[0055] S3: Calculate the differences of each of the said monomers within each of the said SOC intervals.

[0056] Difference refers to the degree of deviation of a single cell's voltage from that of other single cells within the same State of Charge (SOC) range. Based on the voltage datasets of each single cell in different SOC ranges obtained in step S2, an appropriate algorithm (such as normalization) is used to calculate the degree of voltage difference of each single cell relative to other single cells within that SOC range. For example, for a certain SOC range (e.g., 30-40%), voltage datasets of all single cells within this range are collected, and the deviation of each single cell's voltage from the average voltage of all single cells in this range is calculated to measure the difference of that single cell within that range. By calculating the difference, single cells with abnormally deviating voltages can be identified, and these single cells may pose safety hazards.

[0057] S4: Calculate the consistency of each of the said monomers within each of the said SOC intervals.

[0058] Consistency refers to the degree to which the voltage variation pattern of a single cell matches that of other cells within different SOC ranges. Using the voltage dataset from step S2, the voltage variation relationship of each cell across different SOC ranges is analyzed and compared with other cells to determine its degree of consistency. For example, for a given cell, its voltage variation in SOC ranges such as 0-10% and 10-20% is analyzed and compared with the voltage variation patterns of other cells in the battery pack to calculate the degree of fit, thus obtaining the consistency of that cell. Consistency reflects whether the overall pattern of single-cell voltage variation is normal; cells with poor consistency are more likely to cause battery safety problems.

[0059] S5: If both the difference and the consistency meet the preset conditions, then it is determined that the power battery has a safety risk.

[0060] The preset conditions are threshold ranges for variability and consistency pre-defined based on battery safety standards and historical data. When the variability and consistency of a single cell exceed these ranges, a safety risk is considered to exist. The variability and consistency of each cell calculated in steps S3 and S4 are compared with the preset thresholds. If both preset risk conditions are met (e.g., excessive variability and low consistency), the power battery is determined to have a safety risk. For example, the variability threshold is set to a certain value (e.g., deviation from the average by more than 5%), and the consistency threshold is 0.7 (range 0-1, closer to 1 for better consistency). If the variability of a single cell exceeds 5% across multiple SOC ranges and the consistency is below 0.7, it indicates that its performance fluctuates greatly under different SOC states, deviates significantly from the average, and there is a risk of aging, poor contact, or thermal runaway. This single cell is determined to have a safety risk, and consequently, the entire power battery is determined to have a safety risk. A corresponding warning is then issued, which may include: cell number, anomaly type, anomaly value magnitude, and possible causes. The cell number can be recorded in the system log or sent to a remote monitoring platform as a warning to assist in subsequent maintenance or replacement decisions. Accurate assessment of the safety risks of power batteries facilitates timely implementation of safety measures and ensures safe battery use.

[0061] This embodiment focuses on electric vehicle power batteries. By acquiring voltage and SOC data within a warning period, and classifying them by SOC intervals, it calculates the differences and consistency of each individual cell within each SOC interval. Finally, it determines the battery safety risk based on preset conditions. By analyzing the voltage differences and consistency of individual battery cells by SOC interval, it can accurately capture abnormal characteristics of the battery at different charge stages. Compared with existing non-partitioned or single-dimensional analysis methods, this method can more comprehensively and accurately identify battery safety risks, providing a more reliable basis for battery safety management.

[0062] In one embodiment, after extracting the voltage data of each individual unit according to different SOC intervals to obtain voltage datasets of each individual unit in different SOC intervals, the process includes:

[0063] S21: Calculate the average voltage of each of the aforementioned cells in different SOC ranges.

[0064] The average voltage value refers to the arithmetic mean of all voltage data for a single battery cell within a specific SOC range. For each cell's voltage dataset obtained in step S2 across different SOC ranges, the arithmetic mean of each dataset is calculated to obtain the average voltage value for each cell across different SOC ranges. For example, for a single cell's voltage dataset in the 0-10% SOC range, containing 100 voltage data points with values ​​of 3.2V, 3.21V, ..., 3.19V, its arithmetic mean is (3.2 + 3.21 + ... + 3.19) / 100 = 3.2V. Similarly, the average voltage value for this cell in other SOC ranges is obtained. By calculating the average voltage value, the discrete voltage data is transformed into a more representative statistical quantity, facilitating subsequent vector construction and analysis.

[0065] S22: Within each SOC interval, the average voltage values ​​of each individual cell are concatenated to obtain a multidimensional average voltage vector of the individual cell corresponding to each SOC interval.

[0066] A multidimensional voltage mean vector is a vector composed of the average voltage values ​​of all cells within a given state of charge (SOC) range, arranged sequentially. For each SOC range, the average voltage values ​​of all cells within that range are collected and concatenated according to their cell numbers to form the multidimensional voltage mean vector corresponding to that SOC range. For example, a battery pack has 12 cells. In the 0-10% SOC range, the average voltage values ​​of each cell are 3.2V, 3.22V, 3.18V, ..., 3.21V. Concatenating these values ​​in the order of cell 1 to cell 12 yields the vector [3.2, 3.22, 3.18, ..., 3.21], which is the multidimensional voltage mean vector for that SOC range. Constructing a multidimensional voltage mean vector integrates the voltage characteristics of multiple cells into a single vector, facilitating subsequent overall analysis of the battery pack's voltage distribution characteristics within that SOC range and providing structured data for calculating differences and consistency.

[0067] In this embodiment, the average value of the individual voltage datasets for each SOC interval is calculated and concatenated to form a multidimensional voltage mean vector, further providing a simpler and more structured data format for subsequent difference and consistency analysis. Specifically, by calculating the voltage average value and constructing a multidimensional voltage mean vector, the complex voltage data is statistically and structurally processed, making subsequent difference and consistency analysis more efficient and accurate. Compared with directly using the raw voltage data, this reduces the complexity of data processing and improves analysis efficiency.

[0068] In one implementation, the above calculation of the differences of each of the said monomers within each of the said SOC intervals includes:

[0069] S31: Normalize each element in the average voltage vector of the individual cells corresponding to each SOC interval to obtain the normalized value of each individual cell in each SOC interval.

[0070] The normalization process described above scales the data proportionally to fit within a specific range. Here, it's used to eliminate differences in the dimensions and numerical ranges of the average voltage across different SOC ranges, facilitating comparison. For each SOC range's corresponding multidimensional voltage mean vector, each element (i.e., the average voltage of each individual cell within that SOC range) is normalized to obtain the normalized value for each individual cell within that SOC range. For example, taking the multidimensional voltage mean vector [3.2, 3.22, 3.18, ..., 3.21] for a certain SOC range as an example, using the min-max normalization method, each element is subtracted from the minimum value of the vector (e.g., 3.18), and then divided by the difference between the maximum and minimum values ​​of the vector (e.g., 3.22 - 3.18 = 0.04). This yields the normalized vector, and each element represents the normalized value of the corresponding individual cell within that SOC range. Normalization eliminates the differences in dimensions and numerical ranges of the average voltage values ​​of different cells, making the voltage characteristics of different SOC ranges and different cells comparable, thus laying the foundation for accurate calculation of differences. The above normalization process can also utilize Z-score, standardized residuals, and IQR (interquartile range).

[0071] S32: Based on the normalized values ​​of each monomer in each SOC interval, determine the differences of each monomer in different SOC intervals.

[0072] Differential analysis, based on normalized values, measures the degree of deviation of an individual cell from other cells in terms of voltage characteristics. By utilizing the normalized values ​​of each cell within each State of Charge (SOC) range, and calculating the deviation of that cell's normalized value from those of other cells, the differential characteristic within that SOC range is determined. For example, if a cell has a normalized value of 0.8 in a certain SOC range, while the normalized values ​​of other cells in that range are mostly between 0.5 and 0.6, the average deviation of that cell's normalized value from those of other cells is calculated, thus determining that the cell has a significant differential characteristic within that SOC range. Calculating differential characteristic based on normalized values ​​more accurately reflects the degree of anomaly in a cell's voltage relative to other cells, improving the accuracy of differential characteristic identification.

[0073] In this embodiment, the voltage mean vector elements of each SOC interval are normalized to determine the differences between individual cells in each SOC interval, making the difference calculation more scientific and comparable. Using normalization to calculate the differences solves the incomparability problem caused by differences in units and numerical ranges of voltage data from different SOC intervals and individual cells. Compared to directly using the original voltage average value to calculate the differences, the results are more accurate and can more precisely identify individual cells with abnormal voltage.

[0074] In one embodiment, the normalization of each element in the voltage mean vector of the individual cells corresponding to each SOC interval to obtain the normalized value of each individual cell in each SOC interval includes:

[0075] The normalized value is calculated using an improved Z-score normalization algorithm, and the formula is as follows:

[0076]

[0077] in, This represents the improved normalized value of monomer i within the SOC interval j. Let be the average voltage of cell i within the SOC range j. This represents the average voltage of all cells within the SOC range j. Let j be the standard deviation of the average voltage of all cells within the SOC range. This is the operating condition correction factor (with a value range of 0.1-0.3, adaptively adjusted according to the historical fluctuations in the operating conditions of the power battery). The coefficient of variation of the average voltage of all cells within the SOC range j. The normalized value calculated by this formula can be combined with the dispersion of voltage data within the SOC range and the influence of operating condition fluctuations to more accurately reflect the difference of individual cell voltage relative to the whole group of cells. Compared with traditional Z-score normalization, it reduces the misjudgment rate of differences under extreme operating conditions.

[0078] The improved Z-score normalization algorithm described above introduces a working condition correction coefficient and a coefficient of variation based on the traditional Z-score algorithm, to better adapt to normalization methods with different working conditions and data dispersion. The traditional Z-score formula is: (X represents data, The mean, (Standard deviation). In this embodiment, the average voltage of each cell i within the SOC interval j is... Combined with the average voltage of all individual cells in this SOC range Standard deviation Coefficient of variation and operating condition correction factor Substitute into the improved Z-score formula to calculate the normalized value. For example, the average voltage of a single cell i within the SOC range j. =3.3V, the average voltage of all cells in this SOC range =3.2V, standard deviation =0.1V, coefficient of variation =0.1 / 3.2≈0.03125, working condition correction factor =0.2, then substituting into the formula, we get: =0.9938. Compared to traditional Z-score normalization, the improved algorithm introduces a working condition correction coefficient. and coefficient of variation The improved Z-score normalization algorithm can adjust according to the degree of fluctuation in operating conditions and the degree of data dispersion. Under extreme operating conditions (such as large fluctuations in operating conditions and high data dispersion), it can avoid misjudgments caused by the traditional Z-score not considering these factors, making the normalized value more accurately reflect the true difference in cell voltage. Compared with the traditional Z-score, the improved Z-score normalization algorithm can better adapt to different operating conditions and voltage data dispersion of power batteries, making the normalized value more accurate. This improves the accuracy of subsequent difference calculations, reduces the misjudgment rate of differences under extreme operating conditions, and provides a more reliable basis for battery safety early warning.

[0079] In one embodiment, calculating the consistency of each of the said monomers within each of the said SOC intervals includes:

[0080] S401: For each of the said monomers, obtain its normalized value in different SOC intervals and form a normalized multidimensional vector.

[0081] A normalized multidimensional vector refers to a vector composed of the normalized values ​​of a single battery cell across multiple SOC intervals, arranged sequentially. In this application, the normalized values ​​are obtained using the Z-score method. For each cell, combining its average voltage across different SOC intervals, and the average voltage and standard deviation of all cells in the group across the corresponding SOC intervals, the normalized value within each SOC interval is calculated using the Z-score formula. These normalized values ​​are then arranged sequentially according to the SOC intervals to form a normalized multidimensional vector. For example, a battery pack has 12 cells, and the SOC is divided into 9 intervals: 0-20%, 20-30%, 30-40%, ..., 90-100%. For single cell 1, the average voltage X = 3.2V in the 0-20% SOC range, and the average voltage of all cells in the group in this range is 3.15V with a standard deviation of 0.1V. Therefore, the normalized value for this range is 0.5. Similarly, the normalized values ​​for single cell 1 in the other 8 SOC ranges are calculated, forming a normalized multidimensional vector [0.5, 0.3, -0.2, ..., 0.4]. Through Z-score normalization, the average voltage of single cell in different SOC ranges is transformed into a value of deviation from the mean of the entire group, eliminating the differences in dimensions and numerical ranges, making the voltage characteristics of different SOC ranges comparable, and providing a foundation for subsequent vector product matrix construction and consistency analysis.

[0082] S402: Multiply the elements of the normalized multidimensional vector pairwise to obtain the first vector product matrix corresponding to each individual entity, wherein the first vector product matrix is ​​an upper triangular matrix, and only the case where x ≤ y is calculated. , Let x be the normalized value of monomer i in the SOC interval. Let y be the normalized value of monomer i in the SOC interval y, where i is the monomer number and x and y are the SOC interval numbers.

[0083] An upper triangular matrix is ​​a matrix that retains only elements whose row indices are less than or equal to their column indices (i.e., x ≤ y), as follows:

[0084]

[0085] The first vector product matrix is ​​an upper triangular matrix, which makes the vector product matrix more concise while retaining the correlation information of the normalized values ​​of the individual in different SOC intervals.

[0086] S403: Determine the consistency of each of the said entities within each of the said SOC intervals by using the first vector product matrix corresponding to each of the said entities.

[0087] Analyzing the distribution concentration and numerical fluctuations of elements in the upper triangular vector product matrix helps determine the consistency of individual cells. For example, observing the upper triangular matrix of cell 1 above, the relatively concentrated element values ​​indicate a high degree of consistency in the normalized value variation patterns across different SOC ranges, suggesting good consistency. Conversely, if the matrix element values ​​are scattered and fluctuate significantly, the consistency is poor. Utilizing the upper triangular vector product matrix allows for a comprehensive and accurate assessment of individual cell consistency, resulting in a more accurate alignment with the logic of the battery disclosure document and providing a reliable basis for battery safety warnings.

[0088] In this embodiment, an upper triangular matrix construction method is adopted. Z-score normalization is used to obtain the normalized values ​​of individual cells in each SOC range, which are then used to generate an upper triangular vector product matrix for final analysis of individual cell consistency. Compared with existing technologies, Z-score normalization and upper triangular matrix construction more accurately capture the voltage deviation correlation characteristics of individual cells in different SOC ranges, making the consistency assessment results more reliable and effectively improving the accuracy of battery safety warnings.

[0089] In one embodiment, the calculation of the consistency of each of the monomers within each of the SOC intervals includes:

[0090] S411: Multiply the normalized values ​​of each individual unit in different SOC intervals pairwise to obtain a second vector product matrix.

[0091] The second vector product matrix is ​​the matrix formed by multiplying the normalized values ​​of a single entity across multiple SOC intervals in pairs. For each entity, its normalized values ​​across all SOC intervals (e.g., 0-10%, 10-20%) are collected. These normalized values ​​are then combined pairwise and multiplied. The results are arranged in matrix form to obtain the second vector product matrix for that entity. For example, if the normalized values ​​of an entity in three SOC intervals (0-10%, 10-20%, 20-30%) are z1=0.8, z2=0.7, and z3=0.9 respectively, multiplying them pairwise yields the following matrix: By constructing a second vector product matrix, the correlation between the normalized values ​​of individual cells in different SOC intervals is presented in matrix form, which facilitates the overall analysis of the consistency characteristics of individual cell voltage changes.

[0092] S412: Determine the consistency of the individual entity within each of the SOC intervals using the second vector product matrix.

[0093] Consistency refers to the degree to which the voltage variation pattern of a single unit in different SOC intervals matches the patterns of other units or the overall system. This degree of consistency can be reflected from multiple dimensions using the second vector product matrix. Analyzing the distribution and dispersion of elements in the second vector product matrix helps determine the consistency of a single unit across different SOC intervals. For example, observing the second vector product matrix above, if the elements are relatively concentrated and have low dispersion, it indicates that the normalized value variation of the single unit is relatively consistent across different SOC intervals; if the elements are scattered and have high dispersion, it indicates poor consistency. Using the second vector product matrix allows for a comprehensive and multi-dimensional assessment of the consistency of single-unit voltage variations, providing more comprehensive and accurate results compared to assessments using a single indicator.

[0094] In this embodiment, a matrix is ​​constructed by multiplying the normalized values ​​of individual cells in different SOC ranges, and then the consistency of individual cells is analyzed, making the consistency assessment more comprehensive and scientific. Specifically, constructing a second vector product matrix and evaluating consistency based on it can reflect the correlation of voltage changes of individual cells in different SOC ranges from multiple perspectives. Compared with the traditional single-dimensional consistency assessment method, it can more comprehensively and accurately identify cells with poor consistency, providing a more reliable basis for battery safety early warning.

[0095] Furthermore, the determination of the consistency of the individual entity within each SOC interval using the second vector product matrix includes:

[0096] The consistency evaluation index for the vector product matrix is ​​calculated using the following formula:

[0097]

[0098] in, Let be the consistency evaluation value of individual i (ranging from 0 to 1, with values ​​closer to 1 indicating better consistency), and m be the total number of SOC intervals. It is the element in the x-th row and y-th column of the second vector product matrix of the single entity i (i.e., the product of the normalized values ​​of x and y of single entity i in the SOC interval). The average of all elements in the second vector product matrix ( );when When the value is less than the consistency threshold (ranging from 0.6 to 0.8, preset according to the type of power battery and the service life), the consistency of cell i is determined to be abnormal in each SOC range. This indicator achieves a quantitative assessment of the deviation of cell voltage from consistency by quantifying the dispersion of the second vector product matrix, avoiding the one-sidedness of the traditional "judgment based solely on the threshold of a single matrix element".

[0099] Consistency assessment indicators This metric quantifies the consistency of individuals by assessing the dispersion of the second vector product matrix. The value ranges from 0 to 1, with values ​​closer to 1 indicating better consistency. First, the average value of all elements in the second vector product matrix is ​​calculated. Then, calculate the sum of the absolute differences between each element and the mean, and then, combining this with the total number of SOC intervals m, substitute these values ​​into the formula to obtain the result. For example, taking the 3×3 second vector product matrix of the aforementioned single entity as an example, the matrix elements are... First, calculate the sum of all elements: 0.64 + 0.56 + 0.72 + 0.56 + 0.49 + 0.63 + 0.72 + 0.63 + 0.81 = 5.7. The average value is... =5.76 / 9=0.64. Then calculate the sum of the absolute differences between each element and 0.64: ​​|0.64-0.64|+|0.56-0.64|+|0.72-0.64|+|0.56-0.64|+|0.49-0.64|+|0.63-0.64|+|0.72-0.64|+|0.63-0.64|+|0.81-0.64|=0+0.08+0.08+0.15+0.01+0.08+0.01+0.17=0.66. The total number of SOC intervals is m=3. Substituting into the formula, we get: This consistency assessment index quantifies the dispersion of the second vector product matrix, achieving a quantitative and comprehensive evaluation of individual consistency. Compared with traditional methods that judge consistency based on a single matrix element or simple statistics, the results are more objective and accurate, avoiding bias.

[0100] In one embodiment, obtaining the voltage data and SOC data of the power battery within a power battery warning cycle includes:

[0101] S11: The battery management system of the power battery acquires the total current of the battery pack, the initial SOC data, and the initial voltage data of each cell within a power battery warning cycle.

[0102] A Battery Management System (BMS) is a system used to manage power batteries, collecting data such as current, voltage, and State of Charge (SOC). The total current of the entire battery pack is the total current of the power battery pack. Initial SOC and initial voltage data are unfiltered, raw SOC and voltage data. From the BMS of an electric vehicle, the total current of the power battery pack, the initial voltage data of each individual cell, and the corresponding initial SOC data are collected for a warning period (e.g., three consecutive days). For example, the BMS of an electric vehicle collects the total current of the battery pack (accuracy ±0.1A), the initial voltage of each individual cell (accuracy ±0.01V), and the initial SOC (accuracy ±1%) at a frequency of 1Hz for three consecutive days, obtaining the total current, initial SOC, and initial voltage data for that period. Obtaining raw data from the BMS ensures the directness and authenticity of the data, providing a foundation for subsequent data filtering.

[0103] S12: Filter the initial SOC data corresponding to the current of the whole package being greater than or equal to the current threshold and the initial voltage data of each individual cell to obtain the voltage data and SOC data.

[0104] The current threshold is a pre-set current value used to screen the battery's operating state. When the total current of the battery pack is greater than or equal to this threshold, the battery is in a high-current operating state, and the data may be affected by the high current and cannot accurately reflect the battery's normal characteristics. The setting logic in this application is the maximum current value corresponding to the power battery during low-power charging / discharging, such as... Level A covers low-power charging / discharging and idle conditions of the battery. Data with the entire pack current below the current threshold is selected to obtain the final voltage and SOC data used for safety warnings. By filtering data under high-current operating conditions, the interference of high current on battery voltage and SOC data is eliminated, making the voltage and SOC data analyzed later more reflective of the battery's characteristics under normal conditions and improving the accuracy of safety warnings.

[0105] Reference Figure 2 This invention also provides a power battery safety warning device for executing the power battery safety warning method in any of the above embodiments, including:

[0106] The acquisition unit 10 is used to acquire the voltage data and SOC data of the power battery within a power battery warning cycle, wherein the voltage data includes the voltage data of each cell;

[0107] Extraction unit 20 is used to extract the voltage data of each of the units according to different SOC intervals to obtain voltage datasets of each unit in different SOC intervals.

[0108] The first calculation unit 30 is used to calculate the difference between each of the said cells in each of the said SOC intervals, wherein the difference refers to the degree of deviation of the voltage of a cell from the voltage of other cells in the same SOC interval;

[0109] The second calculation unit 40 is used to calculate the consistency of each of the said cells in each of the said SOC intervals. The consistency refers to the degree to which the voltage variation pattern of a cell in different SOC intervals matches that of other cells.

[0110] The determination unit 50 is used to determine that the power battery has a safety risk if both the difference and the consistency meet preset conditions.

[0111] Reference Figure 3 The present invention also provides a computer device, the internal structure of which can be as follows: Figure 3 As shown. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor is designed to provide computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores operating devices, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores voltage signals, etc. The network interface is used to communicate with external terminals via a network connection. Furthermore, the computer device may also include input devices and a display screen. When the computer program is executed by the processor, it implements the power battery safety warning method in any of the above embodiments. Those skilled in the art will understand that... Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer equipment on which the present application is applied.

[0112] One embodiment of this application also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the power battery safety warning method in any of the above embodiments. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0113] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0114] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0115] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for early warning of power battery safety, characterized in that, The method includes: The voltage data and SOC data of the power battery are acquired within a power battery warning cycle, wherein the voltage data includes the voltage data of each individual cell; The voltage data of each individual unit is extracted according to different SOC intervals to obtain voltage datasets of each individual unit in different SOC intervals; Calculate the difference between each of the said cells within each of the said SOC ranges, where the difference refers to the degree of deviation of the voltage of one cell from that of other cells within the same SOC range; and, Calculate the consistency of each of the aforementioned cells within each of the aforementioned SOC intervals. The consistency refers to the degree to which the voltage variation pattern of a cell within different SOC intervals matches that of other cells. If both the difference and the consistency meet the preset conditions, then the power battery is determined to have a safety risk. After extracting the voltage data of each individual unit according to different SOC intervals to obtain the voltage dataset of each individual unit in different SOC intervals, the process includes: Calculate the average voltage of each of the aforementioned cells in different SOC ranges; Within each SOC interval, the average voltage values ​​of each individual cell are concatenated to obtain a multidimensional average voltage vector of the individual cell corresponding to each SOC interval. The calculation of the differences of each of the monomers within each of the SOC intervals includes: Normalize each element in the average voltage vector of the individual cells corresponding to each SOC interval to obtain the normalized value of each individual cell in each SOC interval. Based on the normalized values ​​of each monomer in each SOC interval, the differences of each monomer in different SOC intervals are determined. The step of normalizing each element in the voltage mean vector of the individual cells corresponding to each SOC interval to obtain the normalized value of each individual cell in each SOC interval includes: The normalized value is calculated using an improved Z-score normalization algorithm, and the formula is as follows: in, This represents the improved normalized value of monomer i within the SOC interval j. Let be the average voltage of cell i within the SOC range j. This represents the average voltage of all cells within the SOC range j. Let j be the standard deviation of the average voltage of all cells within the SOC range. This is the working condition correction factor. is the coefficient of variation of the average voltage of all cells within the SOC interval j.

2. The power battery safety early warning method according to claim 1, characterized in that, The calculation of the consistency of each of the monomers within each of the SOC intervals includes: For each of the said entities, its normalized value in different SOC intervals is obtained and formed into a normalized multidimensional vector; Multiply the elements of the normalized multidimensional vector pairwise to obtain the first vector product matrix corresponding to each individual entity. This vector product matrix is ​​an upper triangular matrix, and is calculated only when x ≤ y. , Let x be the normalized value of monomer i in the SOC interval. Let y be the normalized value of monomer i in the SOC interval y, where i is the monomer number and x and y are the SOC interval numbers. The consistency of each individual entity within each SOC interval is determined by the first vector product matrix corresponding to each individual entity.

3. The power battery safety early warning method according to claim 1, characterized in that, The calculation of the consistency of each of the monomers within each of the SOC intervals includes: Multiply the normalized values ​​of each individual in different SOC intervals pairwise to obtain a second vector product matrix; The consistency of the individual entity within each SOC interval is determined by the second vector product matrix.

4. The power battery safety early warning method according to claim 1, characterized in that, The acquisition of voltage and SOC data of the power battery within a power battery warning cycle includes: The battery management system of the power battery acquires the total current of the battery pack, the initial SOC data, and the initial voltage data of each individual cell within a power battery warning cycle. The initial SOC data corresponding to a total current greater than or equal to a current threshold and the initial voltage data of each individual cell are filtered to obtain the voltage data and SOC data.

5. A power battery safety warning device, used to execute the power battery safety warning method as described in any one of claims 1-4, characterized in that, include: The acquisition unit is used to acquire the voltage data and SOC data of the power battery within a power battery warning cycle, wherein the voltage data includes the voltage data of each individual cell; The extraction unit is used to extract the voltage data of each of the individual units according to different SOC intervals to obtain the voltage dataset of each individual unit in different SOC intervals. The first calculation unit is used to calculate the difference between each of the said cells in each of the said SOC intervals, where the difference refers to the degree of deviation of the voltage of a cell from the voltage of other cells in the same SOC interval; The second calculation unit is used to calculate the consistency of each of the said cells in each of the said SOC intervals. The consistency refers to the degree to which the voltage variation pattern of a cell in different SOC intervals matches that of other cells. The determination unit is used to determine that the power battery has a safety risk if both the difference and the consistency meet preset conditions.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the power battery safety early warning method as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the power battery safety early warning method as described in any one of claims 1 to 4.

Images