Management method for a battery pack, electronic device, and vehicle

Abstract

Embodiments of this disclosure provide a management method, electronic device, and vehicle for a battery pack. The management method includes: acquiring a first data set associated with a plurality of interconnected battery cells, the first data set including electrical data of the plurality of battery cells over a predetermined time period; generating a second data set by selecting electrical data from at least a portion of the predetermined time period from the first data set based on the state of charge of the plurality of battery cells over the predetermined time period; determining at least one set of characteristic values ​​for the plurality of battery cells based on the second data set, the at least one set of characteristic values ​​including a set of characteristic values ​​associated with the internal resistance of the plurality of battery cells; and determining whether any abnormality exists in the plurality of battery cells based on the at least one set of characteristic values. The solution of this disclosure can effectively improve the accuracy of battery fault warning at a lower cost.

Description

Technical Field

[0001] Embodiments of this disclosure relate to the field of battery management technology, and more specifically to a management method for a battery pack, an electronic device, a vehicle including the electronic device, and a computer-readable medium. Background Technology

[0002] To mitigate the impact of fossil fuels on the climate and environment, the new energy industry has developed rapidly, particularly with the widespread and extensive application of energy storage batteries such as lithium-ion batteries in various fields. For example, new energy vehicles and the energy storage power supplies for wind and solar power plants all require a large number of energy storage batteries. However, with the widespread use of energy storage batteries such as lithium-ion batteries, fires caused by spontaneous combustion of batteries have become increasingly frequent, raising growing concerns about the safety of energy storage batteries.

[0003] Currently, to ensure the safety of energy storage batteries, fault diagnosis functions can be installed in the battery system. However, most fault diagnosis solutions can only detect faults after they have occurred, thus failing to prevent potential losses. Therefore, some fault diagnosis solutions propose providing early warnings before battery failures occur, enabling effective measures to eliminate safety hazards before they happen. However, existing fault warning solutions suffer from low accuracy and poor practicality. Frequent false alarms not only worsen the user experience and increase battery maintenance costs, but may also lead to the omission of genuine anomalies or faults. Summary of the Invention

[0004] In view of the above problems, according to the exemplary embodiments of the present disclosure, a method for managing battery packs, an electronic device, a vehicle, and a computer-readable medium are provided.

[0005] In a first aspect of this disclosure, a management method for a battery pack is provided, the method comprising: acquiring a first data set associated with a plurality of battery cells connected to each other, the first data set including electrical data of the plurality of battery cells over a predetermined time period; generating a second data set by selecting electrical data from at least a portion of the predetermined time period from the first data set based on the state of charge of the plurality of battery cells over the predetermined time period; determining at least a set of characteristic values ​​of the plurality of battery cells based on the second data set, the at least a set of characteristic values ​​including a set of characteristic values ​​associated with the internal resistance of the plurality of battery cells; and determining whether any abnormality exists in the plurality of battery cells based on the at least a set of characteristic values.

[0006] In a second aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory coupled to the processor, the memory having instructions stored therein, the instructions causing the device to perform actions when executed by the processor, the actions including: acquiring a first data set associated with a plurality of battery cells connected to each other, the first data set including electrical data of the plurality of battery cells over a predetermined time period; generating a second data set by selecting electrical data from at least a portion of the predetermined time period from the first data set based on the state of charge of the plurality of battery cells over the predetermined time period; determining at least a set of characteristic values ​​of the plurality of battery cells based on the second data set, the at least a set of characteristic values ​​including a set of characteristic values ​​associated with the internal resistance of the plurality of battery cells; and determining whether the plurality of battery cells are abnormal based on the at least a set of characteristic values.

[0007] In a third aspect of this disclosure, a vehicle is provided, comprising: a battery pack including a plurality of battery cells connected to each other; and electronic equipment according to the second aspect.

[0008] In a fourth aspect of this disclosure, a computer-readable medium is provided having computer-readable instructions stored thereon, which, when executed by a processing unit, cause the processing unit to perform the management method according to the first aspect.

[0009] It should be understood that the description in the Summary of the Invention is not intended to limit the key or essential features of the embodiments of this disclosure, nor is it intended to restrict the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the various embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0011] Figure 1 A schematic diagram of an example scenario in which embodiments of the present disclosure may be implemented is shown.

[0012] Figure 2 A schematic flowchart of a battery pack management method according to an embodiment of the present disclosure is shown.

[0013] Figure 3 A schematic flowchart illustrating an example process for obtaining a first data set according to an embodiment of the present disclosure is shown.

[0014] Figure 4 A schematic flowchart illustrating an example process for generating a second data set according to an embodiment of the present disclosure is shown.

[0015] Figure 5A The graph showing the relationship between the open-circuit voltage and the state of charge of a ternary lithium battery is presented.

[0016] Figure 5B The graph shows the relationship between the internal resistance and the state of charge of a ternary lithium battery.

[0017] Figure 6 A schematic flowchart illustrating an example process for determining a set of feature values ​​based on a second data set according to an embodiment of the present disclosure is shown.

[0018] Figure 7 A schematic flowchart illustrating an example process for determining a set of feature values ​​based on a second data set according to an embodiment of the present disclosure is shown.

[0019] Figure 8 A schematic flowchart illustrating an example process for determining whether a plurality of battery cells are abnormal based on at least one set of feature values, according to an embodiment of the present disclosure.

[0020] Figure 9 A schematic block diagram of an example device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation

[0021] The embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0022] In the description of embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment / implementation" or "this embodiment / implementation" should be understood as "at least one embodiment / implementation". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0023] Figure 1 A schematic diagram of an example scenario 1000 in which embodiments of the present disclosure may be implemented is shown. It should be understood that the devices or apparatus shown in this example scenario 1000 are merely examples, and the devices or apparatus that may appear in different application scenarios will vary depending on the actual situation. The scope of the present disclosure is not limited in this respect.

[0024] like Figure 1As shown, scenario 1000 includes a vehicle 100. As an example, vehicle 100 can be an electric vehicle, a hybrid vehicle, or other type of vehicle equipped with an energy storage battery. Vehicle 100 includes an energy storage device 110, and the energy storage device 110 includes a battery pack 111. The battery pack 111 may include multiple battery cells 1111 connected to each other. For example, the multiple battery cells 1111 may be connected together in series and / or parallel. It should be noted that... Figure 1 The number of battery cells shown is merely exemplary, and the battery pack 111 may include any number of battery cells 111 as needed. For example, a battery pack 111 for a vehicle may include hundreds or thousands of battery cells.

[0025] Vehicle 100 may also include a control platform 120 for monitoring and managing vehicle 100, as well as implementing other vehicle-related control functions. In some embodiments, the energy storage device 110 of vehicle 100 may include a Battery Management System (BMS) 112 for monitoring, managing, and maintaining battery pack 111, such as measuring and collecting various real-time physical and electrical parameter data of the battery, battery state estimation, equalization management, thermal management, charge and discharge control, etc. It is understood that although BMS 112 is set separately from control platform 120 here, BMS 112 can also be integrated into control platform 120, which can achieve the same purpose of this disclosure. Thus, control platform 120 or BMS 112 can acquire measurement or sampling data associated with battery pack 111 and its battery cells of vehicle 100. The control platform 120 or BMS 112 can store this data and analyze and diagnose the battery status based on this data in real time or at an appropriate time after data collection, in order to determine whether there are any abnormal conditions in each battery cell 1111 in the battery pack 111 that may lead to battery failure and safety accidents, thereby providing early warning before battery failure occurs.

[0026] In some embodiments, scenario 1000 may include a cloud 200. For example, the cloud 200 may communicate with the vehicle 100 or its energy storage device 110 via wired or wireless communication to exchange data with the vehicle 100 or its energy storage device 110. Furthermore, the cloud 200 may also communicate with other vehicles or related devices via wireless or wired communication networks to achieve data exchange. With the help of the cloud 200, large amounts of measurement or sampling data (e.g., long-term data over several days or weeks) from the vehicle 100 or other vehicles can be stored in the cloud 200. This avoids the need for the vehicle itself to have large-capacity storage devices to store the data, reducing vehicle costs and space occupation, and also avoids data loss due to insufficient vehicle data storage capacity.

[0027] In some embodiments, scenario 1000 may include computing device 300, which may be a remote computing platform external to vehicle 100. Computing device 300 may acquire measurement or sampling data related to vehicle 100 or other vehicles from cloud 200, and process this data to analyze and diagnose the battery pack status in vehicle 100 or other vehicles, thereby providing battery fault warnings for these vehicles. However, it is understood that vehicle 100's control platform 120 or BMS 112 may also acquire data from cloud 200 when needed, thereby directly utilizing vehicle 100's control platform 120 or BMS 112 to process the data, allowing battery fault warnings to be implemented by vehicle 100's own computing and control devices.

[0028] As can be seen from the above, in scenario 1000, the measurement or sampling data of battery pack 111 can be stored in the storage device of vehicle 100 and / or stored in the cloud 200, and can be obtained by the control platform 120 of vehicle 100 and / or BMS 112 and / or computing device 300 outside vehicle 100 for processing to realize fault early warning analysis of battery pack 111.

[0029] Some conventional battery fault prediction schemes utilize abnormal voltage fluctuations in battery cells within a battery pack to detect battery anomalies. However, this approach is susceptible to factors such as data quality, environmental conditions, and the nonlinear characteristics of the battery system. For example, battery cell voltages fluctuate drastically with changes in State of Charge (SOC), meaning that a cell exhibiting significant voltage fluctuations does not necessarily indicate an anomaly. Furthermore, as SOC changes over time (e.g., gradually decreasing during discharge or gradually increasing during charging), the cell voltage itself changes accordingly. Therefore, cell voltages inevitably vary over a given period. Considering the inconsistency in SOC among individual cells within the battery pack, it is difficult to distinguish whether a cell's voltage fluctuation relative to others is due to SOC inconsistency or an inherent anomaly within the cell itself. These issues contribute to a high false alarm rate in battery fault prediction systems.

[0030] Embodiments of this disclosure propose an improved management method for battery packs. By acquiring electrical data of the battery over a predetermined time period and filtering the data based on the state of charge (SOC), the impact of SOC on electrical data, particularly battery internal resistance, can be eliminated or reduced, thereby improving the accuracy of fault warnings.

[0031] Figure 2A schematic flowchart of a battery pack management method 2000 according to an embodiment of the present disclosure is shown. Method 2000 can be implemented in scenario 1000 and executed by computing device 300, or by control platform 120 or BMS 112 of vehicle 100. It is understood that the implementation of method 2000 is not limited to this, but can be implemented in other scenarios requiring battery pack management, such as energy storage power systems in photovoltaic power plants. For the purposes of discussion, [further details will be provided]. Figure 1 Let's describe method 2000, and assume that the method is executed by computing device 300.

[0032] At box 2001, computing device 300 acquires a first data set associated with a plurality of battery cells connected to each other. The first data set includes electrical data of the plurality of battery cells 1111 over a predetermined time period. As an example, the electrical data included in the first data set may be electrical data such as voltage and current of each battery cell in the battery pack 111 during the predetermined time period, which can be acquired by measurement or sampling. Furthermore, the electrical data may also include data acquired after certain processing, such as battery charge data during the predetermined time period, which can be obtained by calculation from measured or sampled current data.

[0033] At box 2002, computing device 300 generates a second data set by selecting electrical data from at least a portion of the predetermined time period from a first data set based on the state of charge (SOC) of multiple battery cells 1111 over a predetermined time period. As an example, the SOC of each battery cell in the resistive cell 111 reflects the remaining charge of each cell; for example, an SOC of 100% or 1 indicates that the cell is fully charged, while an SOC of 0% or 0 indicates that the cell is fully discharged. The SOC of each battery cell can be obtained, for example, by integrating the cell's discharge current or through other suitable means. Changes in SOC can affect some characteristics of the battery cell, such as internal resistance and voltage fluctuations. For example, when the SOC is high or low, the internal resistance and voltage fluctuations of the battery cell become very sensitive and exhibit non-linear characteristics. Therefore, the first data set can be filtered based on the SOC of the battery cells during the predetermined time period, and, for example, data exhibiting non-linear characteristics can be removed.

[0034] At box 2003, computing device 300 determines at least one set of characteristic values ​​for a plurality of battery cells based on a second data set. This at least one set of characteristic values ​​includes a set of characteristic values ​​associated with the internal resistance of the plurality of battery cells 1111. As an example, each characteristic value in the set corresponds to one battery cell, and each characteristic value can reflect whether a certain characteristic of the battery cell corresponding to that characteristic value is abnormal, such as whether it deviates from the overall average characteristic within the battery pack. The set of characteristic values ​​associated with the internal resistance of the plurality of battery cells 1111 can reflect whether the internal resistance of the battery cells is abnormal. For battery packs in applications such as vehicles, battery manufacturers typically screen battery cells, resulting in similar and consistent internal resistance values ​​among battery cells within the same battery pack. Because the initial internal resistance of the battery cells is similar, and their environments are also similar, the resistance consistency of the battery cells within the battery pack is relatively good before the batteries reach the level of aging required for retirement. As batteries age, the internal resistance of battery cells is affected when abnormal conditions such as micro-short circuits or leakage occur. In other words, the internal resistance of a battery cell can reflect the internal mechanism or state changes of the battery to a certain extent. Therefore, by determining a set of characteristic values ​​associated with the internal resistance of multiple battery cells in a resistor array, it is possible to determine with relatively high accuracy whether abnormal battery cells have occurred in the battery array.

[0035] At box 2004, computing device 300 determines whether any of the multiple battery cells 1111 are abnormal based on at least one set of feature values. As an example, computing device 300 can process at least one set of features and determine whether any of the multiple battery cells 1111 are abnormal, thereby enabling timely and effective measures to be taken before the battery malfunctions or an accident occurs (e.g., battery spontaneous combustion), such as replacing or maintaining the battery pack 111, to eliminate potential safety hazards in advance.

[0036] Figure 3 A schematic flowchart of an example process 3000 for obtaining the first data set is shown. Figure 3 The process shown can be performed in 3000. Figure 2 Implemented at box 2001.

[0037] At box 3001, computing device 300 receives a sequence of sampled data associated with multiple battery cells 1111. As an example, computing device 300 may receive the sequence of sampled data related to the battery pack 111 of vehicle 100 from cloud 200 or via a vehicle network. This sequence of sampled data may be acquired by vehicle 100's BMS 112 or other measurement devices and transmitted to cloud 200 or vehicle network, and may be measurement or sampling data of the vehicle battery over a relatively long period of time.

[0038] At box 3002, computing device 300 calculates the average current I based on current values ​​in the sampled data sequence whose absolute values ​​are greater than the current threshold. avg The current threshold is greater than zero. For example, the current threshold could be 0.5A, meaning the computing device 300 can select current data greater than 0.5A or less than -0.5A from the sampled data sequence and calculate the average current value for each battery cell. Data with current close to zero might be data from when the battery pack is not charging or discharging, such as when the vehicle is in standby mode. Such data cannot be used to assess the battery's state and can therefore be excluded. Thus, the calculated average current I... avg This can represent the average charging and discharging current of a battery cell. It should be noted that, for the sake of simplicity, it is assumed here that multiple battery cells in the battery pack are connected in series, so each battery cell 1111 has the same average current as the battery pack 111 as a whole.

[0039] At box 3003, computing device 300 calculates the average current I based on the calculated value. avg The predetermined time period is calculated by considering the specified rate of change of the state of charge (SOC). Specifically, by determining the length of the predetermined time period, the length of the sliding time window (i.e., the predetermined time period) corresponding to the first data set can be determined. This length will affect the selection of electrical data used for analysis and diagnosis, and thus the results of battery fault early warning analysis and diagnosis. For example, if the predetermined time period is too short, the amount of data in the first data set corresponding to the predetermined time period will be small, and characteristics such as internal resistance and voltage fluctuations will be unstable. Conversely, if the predetermined time period is too long, the SOC changes will be large, and characteristics such as voltage fluctuations will be more due to normal fluctuations caused by SOC changes. In this case, characteristics such as voltage fluctuations cannot be used to analyze the state of the battery cell. Therefore, factors such as battery capacity and driving habits need to be considered to select an appropriate length for the predetermined time period or sliding time window. As an example, the change in the SOC of the battery cell, ΔSOC, can be determined or specified, for example, 1-5%, to ensure that the SOC change within the predetermined time period or sliding time window is small and does not cause excessive voltage changes due to SOC changes. Based on the specified change ΔSOC and the average current calculated at box 3002, the duration of the predetermined time period or sliding time window can be calculated using the following equation:

[0040]

[0041] Where t1 is the duration of the predetermined time period or sliding time window, and Ca is the rated capacity of battery pack 111 or battery cell 1111.

[0042] At block 3004, computing device 300 preprocesses sampled data within a predetermined time period in the sampled data sequence to generate a first data set. In some embodiments, computing device 300 acquires a set of voltage values ​​and a set of current values ​​for a plurality of battery cells 1111 within the predetermined time period, calculates a set of charge values ​​for the plurality of battery cells 1111 based on the set of current values, and generates the first data set based on the set of voltage values, the set of current values, and the set of charge values.

[0043] As an example, voltage and current data for each battery cell can be obtained from the sampled data sequence according to a predetermined time period or the duration t1 of a sliding time window. The voltage of multiple battery cells 1111 can be expressed as... In the matrix, x represents the number of samples within the predetermined time period, and y represents the number of battery cells 1111 within the battery pack 110. For simplicity and clarity, it is assumed here that multiple battery cells 1111 are connected in series, or that battery cells connected in parallel within the battery pack 111 are considered as a single battery cell, thus treating the battery pack as being formed by multiple battery cells connected in series. Therefore, multiple battery cells in the battery pack can have the same current magnitude at each sampling time, and the current of multiple battery cells can be expressed as [I1 I2 … I x The subscript x represents the number of samples. Then, based on the sampled data within a predetermined time period, the cumulative electricity consumption within that time period can be calculated using the following equation (i.e., the ampere-hour integral formula):

[0044]

[0045] The calculated capacity of multiple battery cells can be expressed as [Q1 Q2 … Q x ].

[0046] Figure 4 A schematic flowchart of an example process 4000 for generating a second data set is shown. Figure 5A The graph showing the relationship between the open-circuit voltage and the state of charge (SOC) of a ternary lithium battery is presented. Figure 5B The graph shows the relationship between the internal resistance and the state of charge (SOC) of a ternary lithium battery. Figure 4 The process shown in 4000 can be performed... Figure 2 Implemented at box 2002.

[0047] At box 4001, computing device 300 acquires the state of charge of each of the multiple battery cells over a predetermined time period.

[0048] At box 4002, computing device 300 selects electrical data from a first data set for a time period in which the state of charge (SOC) of multiple battery cells is within a predetermined range to generate a second data set.

[0049] Specifically, such as Figure 5A and 5B As shown, when the State of Charge (SOC) is within a predetermined range (e.g., the middle range of SOC, approximately 30-80%), the battery's internal resistance tends to stabilize, and the open-circuit voltage has a nearly linear relationship with the SOC, resulting in relatively small voltage fluctuations caused by changes in SOC. However, when the SOC is high or low, the battery's internal resistance and open-circuit voltage will exhibit non-linear and drastic changes with SOC. In other words, when the SOC of a battery pack, such as a lithium battery, is within the predetermined range or interval described above, battery characteristics such as internal resistance and voltage fluctuations can remain essentially stable. Therefore, in the process of determining battery anomalies by analyzing battery characteristics (e.g., internal resistance and voltage fluctuations), selecting electrical data corresponding to the aforementioned predetermined range or interval of SOC in the first dataset will help improve the accuracy of battery anomaly detection.

[0050] In some embodiments of this disclosure, the predetermined range of SOC is from 35% to 75% of the state of charge. Specifically, for lithium batteries, battery characteristics such as internal resistance and voltage fluctuation are relatively stable within an SOC range of 30% to 80%. Meanwhile, considering that there is a 5% error in the estimation or calculation of SOC, selecting an SOC range of 35% to 75% is more preferred.

[0051] In some embodiments, the computing device 300 selects electrical data from at least a portion of a predetermined time period from a first data set based on additional conditions. These additional conditions include at least one of the following: the voltage and current values ​​of multiple battery cells 1111 are not empty during the selected time period; and the absolute value of the current value of multiple battery cells 1111 during the selected time period is greater than zero. Specifically, in addition to selecting electrical data for time periods where the State of Charge (SOC) is within a predetermined range, further data filtering can be performed. For example, if data for some battery cells is not collected at a certain sampling moment, this may result in some battery cells having empty data at that sampling moment. In this case, all electrical data for that sampling moment can be removed from the first data set. Furthermore, electric vehicles may experience long-term standby scenarios where the vehicle's battery current is low and stable, making the data unsuitable for analyzing battery anomalies. Therefore, if the current at some sampling moments is zero or close to zero, it indicates that the vehicle and battery may be in a standby scenario, and the electrical data corresponding to these sampling moments can be discarded instead of being included in the second data set.

[0052] The voltage in the second data set obtained after screening can be expressed, for example, as where p in the matrix is the number of sampling times after screening, so p < x. The current in the second data set can be expressed as [I1 I2 … I p , and the electric quantity can be expressed as [Q1 Q2 … Q p .

[0053] Before further processing the second data set, the computing device 300 can determine the validity of the second data set. In some embodiments of the present disclosure, the computing device 300 can further determine the data volume of the second data set and / or the maximum change amount of the current values in the second data set, and if the data volume is lower than the data volume threshold and / or the maximum change amount is lower than the change amount threshold, the computing device 300 determines that the second data set is invalid and discards the second data set, and if the data volume exceeds the data volume threshold and / or the maximum change amount exceeds the change amount threshold, the computing device 300 determines that the second data set is valid.

[0054] For example, the maximum value I p and the minimum value I max can be found in the current matrix [I1 I2 … I min of the second data set, and the data volume num in the second data set can be counted, where the theoretical data volume num_total in the predetermined time period t1 = t1 / Δt, and Δt is the sampling time interval. Further, a current change amount threshold can be set, for example, 10A. Thus, when the maximum change amount I max -I min of the current values in the second data set is lower than the current change amount threshold of 10A, it can be determined that there is not enough change in the battery current during the time period corresponding to the current second data set. In addition, a data volume threshold can also be set, for example, the data volume threshold is set to 100, or set to a certain proportion of the theoretical data volume num_total, for example, 70% of num_total. When the data volume in the second data set is lower than the data volume threshold, it can be determined that the second data set obtained after screening the first data set lacks sufficient valid data. Lack of sufficient valid data and insufficient current change amount may both lead to an increase in accidental errors, thus affecting the fault warning analysis and even resulting in incorrect analysis results. Therefore, when it is determined that the second data set has at least one of the situations of insufficient data volume and insufficient current change amount, the second data set can be discarded, and return to box 2001 or box 3001 to re-acquire data.

[0055] Figure 6 FIG. shows a schematic flowchart of an example process 6000 for determining a set of eigenvalue associated with the battery internal resistance based on the second data set. Figure 6The process shown in 6000 can be performed... Figure 2 Implemented at box 2003.

[0056] At box 6001, for each battery cell, computing device 300 performs a linear fit based on the voltage, current and charge values ​​of the corresponding battery cell in the second dataset to obtain the internal resistance value of the corresponding battery cell.

[0057] As an example, the voltage matrix from the second data set can be used... Select the voltage data [V] of the nth battery cell. n,1 V n,1 … V n,p ], where 1≤n≤y. Furthermore, in the case of multiple battery cells connected in series, the current and charge data of the nth battery cell are [I1 I2 … I p ] and [Q1 Q2 … Q p Then, with voltage as the dependent variable and current and charge as independent variables, a multidimensional linear fit was performed, and the fitted equation is shown below:

[0058] V n = a*I+b*Q+c (3)

[0059] In equation (3), a, b, and c are the fitting parameters, where parameter a is the internal resistance r of the nth battery cell during the sliding time window or a predetermined time period. n It should be noted that in equation (3), in addition to current as the independent variable, the effect of electrical quantity on voltage is also considered. Figure 5A (An example is shown showing the relationship between charge or SOC and voltage), so the charge is further incorporated into the linear fit, which improves the fitting accuracy and thus improves the accuracy of the obtained internal resistance value.

[0060] By repeating the above steps, the internal resistance values ​​of other battery cells during a predetermined time period or sliding time window can be obtained. A set of internal resistance values ​​for the multiple battery cells 1111 (i.e., y battery cells) of the battery pack 111 can be expressed as [r1 r2 …r y ].

[0061] Conventional methods for obtaining the internal resistance of lithium batteries require hybrid power pulse characteristic (HPPC) testing in a laboratory environment. The internal resistance is calculated by using the ratio of voltage change to current change at different times. These conventional methods are only feasible in a laboratory setting and require high-precision charging and discharging peripherals, resulting in high costs. The improved online calculation scheme for battery internal resistance disclosed in this paper accurately calculates the battery internal resistance by selecting appropriate data segments and employing multivariate regression analysis, reducing the computational burden and eliminating the need for high-precision voltage signals and additional costly acquisition equipment.

[0062] At box 6002, for each battery cell, the computing device 300 calculates based on a set of internal resistance values ​​[r1 r2 … r] for multiple battery cells. y Determine the internal resistance Z fraction of the corresponding battery cell.

[0063] As an example, based on the internal resistance [r1 r2 … r] of multiple battery cells 1111... y The standard deviation r_std and the average value r_avg of the internal resistance sequence can be calculated. Therefore, the internal resistance Z-score Z of the nth battery cell can be determined. r，n It can be calculated using the following equation:

[0064]

[0065] The internal resistance Z-score calculated by equation (4) is a dimensionless value. Furthermore, the SOC, temperature, and aging degree within the same battery pack 111 are not significantly different. Therefore, the internal resistance Z-score obtained by comparing the battery cells within the battery pack 111 will be basically unrelated to SOC, temperature, and aging degree, and can thus serve as a characteristic of internal resistance consistency.

[0066] At box 6003, a set of internal resistance Z-scores for multiple battery cells is determined as a set of eigenvalues ​​from at least one set of eigenvalues. Specifically, by repeating the calculation steps of equation (4) for other battery cells, the internal resistance Z-score of each battery cell in battery pack 111 can be calculated, thereby obtaining a set of internal resistance Z-scores (or internal resistance Z-score sequence or matrix) [Z] for multiple battery cells 1111. r,1 Z r,2 … Z r,y ], and use it as one of the at least one set of features mentioned in box 2003.

[0067] In some embodiments of this disclosure, at least one set of feature values ​​in block 2003 may further include a set of feature values ​​associated with the voltage fluctuation levels of the multiple battery cells. Specifically, in addition to feature values ​​associated with the battery internal resistance, feature values ​​associated with the voltage fluctuation levels of the multiple battery cells may be further introduced, which can increase the accuracy of battery fault early warning analysis.

[0068] Figure 7 A schematic flowchart of an example process 7000 is shown, which determines a set of characteristic values ​​associated with the degree of voltage fluctuation of multiple battery cells based on a second dataset.

[0069] At box 7001, for each battery cell, computing device 300 obtains the voltage entropy of the corresponding battery cell based on the voltage value of the corresponding battery cell in the second data set.

[0070] As an example, in the voltage matrix of the second data set In the process, determine the maximum voltage V of the nth battery cell. max Minimum voltage V min And the total data volume is N. Based on the maximum voltage V max and minimum voltage V min The voltage data is divided into m equal-width voltage intervals, with each interval having a width of [value missing]. Therefore, m voltage ranges (V) can be obtained. min V min +ΔV]、(V min +ΔV,V min +2ΔV]、……、(V min +(m-1)ΔV,V max Then, we can consider the number N of the nth battery cell in different voltage ranges. i Statistical analysis was performed, and the probability density distribution P of the nth battery cell in the m voltage intervals was calculated according to the following equation. i :

[0071]

[0072] In this way, the distribution frequency or probability of the voltage of the nth battery cell in different voltage ranges can be calculated. When dividing the voltage ranges, the number of ranges should not be too large; for example, m can be selected as 5 to 10.

[0073] Based on the probability distribution of the nth battery cell across m voltage intervals, the voltage entropy E of the nth battery cell can be calculated using the following equation, which is similar to the formula for calculating information entropy. n :

[0074]

[0075] Repeating the above steps determines the voltage entropy of each battery cell in battery pack 111 over a predetermined time period, and yields a set of voltage entropies (or voltage entropy sequences or matrices) [E1E2 … E] for multiple battery cells in battery pack 111. y ].

[0076] At box 7002, for each battery cell, computing device 300 calculates based on a set of voltage entropies [E1 E2 … E] for multiple battery cells. y [The voltage entropy Z fraction of the corresponding battery cell is determined.]

[0077] As an example, based on the voltage entropy [E1 E2 … E] of multiple battery cells 1111 y The standard deviation E_std and the average value E_avg of the voltage entropy sequence can be calculated. Therefore, the voltage entropy Z-score of the nth battery cell is... E,n It can be calculated using the following equation:

[0078]

[0079] At box 7003, a set of voltage entropy Z fractions for multiple battery cells is determined as a set of eigenvalues ​​from at least one set of eigenvalues.

[0080] Specifically, for the other battery cells in battery pack 111, by repeating the calculation steps of equation (7), the voltage entropy Z-scores of all battery cells in battery pack 111 can be calculated, thereby obtaining a set of voltage entropy Z-scores (or voltage entropy Z-score sequences or matrices) of multiple battery cells 1111 [Z E,1 Z E,2 … Z E,y ], and use it as one of the at least one set of features mentioned in box 2003.

[0081] Therefore, the computing device 300 can at least determine a set of characteristic values ​​associated with the battery's internal resistance based on the second data set. Alternatively, the computing device 300 can determine a first set of characteristic values ​​associated with the battery's internal resistance and a second set of characteristic values ​​associated with the degree of battery voltage fluctuation (e.g., voltage entropy) based on the second data set. It is understood that the embodiments of this disclosure are not limited thereto, and characteristic values ​​of other properties can be introduced according to the actual needs of battery fault early warning diagnosis, thereby further improving the analysis and diagnosis process.

[0082] Figure 8 A schematic flowchart of an example process 8000 for determining whether multiple battery cells are abnormal based on at least one set of feature values ​​is shown. Figure 8 The process shown can be performed in 8000. Figure 2Implemented at box 2004.

[0083] At box 8001, computing device 300 uses an outlier algorithm to determine at least one eigenvalue threshold corresponding to at least one set of eigenvalues. As an example, computing device 300 may use an outlier algorithm such as the 3σ criterion or quartic difference to determine the threshold TH1 of the battery internal resistance Z-fraction. Furthermore, if the voltage entropy of the battery cell is provided, a similar outlier algorithm may be used to determine the threshold TH2 of the voltage entropy Z-fraction. Alternatively, TH1 and / or TH2 may be determined using empirical values.

[0084] At box 8002, computing device 300 determines whether at least one set of feature values ​​contains feature values ​​that exceed the corresponding feature value threshold. For example, it sequentially determines the internal resistance Z fraction of multiple battery cells [Z]. r,1 Z r,2 … Z r,y The internal resistance Z-score of each battery cell in the battery pack is checked to see if it exceeds the threshold TH1. If the internal resistance Z-score of one or more battery cells exceeds the threshold TH1, it indicates that there are outlier battery cells in the battery pack 111; otherwise, it indicates that there are no outlier battery cells in the battery pack 111. Furthermore, given the voltage entropy of the battery cells, the voltage entropy Z-scores of multiple battery cells can be determined sequentially. E,1 Z E,2 … Z E,y The system checks whether the voltage entropy Z-score of each battery cell in the battery pack exceeds the threshold TH2. If the internal resistance Z-score of one or more battery cells exceeds the threshold TH1 and the voltage entropy Z-score exceeds TH2, it indicates that there are outlier battery cells in battery pack 111; otherwise, it indicates that there are no outlier battery cells in battery pack 111. Using both the internal resistance Z-score and the voltage entropy Z-score to identify outlier battery cells can reduce the interference and impact of data quality differences on outlier detection, further improving the accuracy of the detection.

[0085] At box 8003, if at least one set of feature values ​​contains feature values ​​exceeding the corresponding feature value threshold, the computing device 300 increments the counter. As an example, the computing device 300 can provide an indicator `battery_fault` and a counter `cumNum`. When an outlier battery cell is determined to exist, `battery_fault` can be set to 1; otherwise, it can be set to 0. Further, when `battery_fault` = 1, the counter `cumNum` = `cumNum + battery_fault`, and when `battery_fault` = 0, the counter `cumNum` = 0. That is, when it is determined based on the current data that an outlier battery cell exists in battery pack 111, the counter `cumNum` can be incremented; if no outlier battery cell exists, the counter `cumNum` is reset to zero.

[0086] At box 8004, the computing device 300 determines whether the counter count exceeds a counting threshold. At box 8005, if the counter count exceeds the counting threshold, the computing device 300 determines that the battery pack or multiple battery cells are abnormal. Specifically, the previous steps can be repeated to repeatedly determine whether outlier battery cells exist based on newly acquired data. Thus, when the number of times outlier battery cells are determined to exist exceeds a certain counting threshold TH3 (e.g., 2 to 5), it is determined that the battery pack 111 or multiple battery cells are indeed abnormal. In this way, the impact of random errors in the data can be reduced, further improving the accuracy of battery anomaly detection.

[0087] The consistency characteristics of a battery cell can include properties such as State of Charge (SOC), voltage, and internal resistance. SOC consistency only indicates inconsistency in the battery cell's charge level and is unrelated to safety. Battery internal resistance and voltage fluctuations, characterized by voltage entropy, can reflect the internal mechanism or state changes of the battery to some extent. However, both battery internal resistance and voltage entropy are affected by SOC, exhibiting non-linear changes, for example, at high or low SOC. Therefore, simply analyzing and diagnosing battery anomalies based on battery internal resistance (and voltage entropy) may result in a high false alarm rate. In the embodiments of this disclosure, data is filtered based on SOC, and the state of battery internal resistance (and voltage entropy) is determined based on the filtered data, enabling more accurate battery anomaly diagnosis. Furthermore, this disclosure employs an improved algorithm to calculate battery internal resistance, which improves the accuracy of internal resistance calculation while reducing computational load and overall cost.

[0088] Figure 9 A schematic block diagram of an example device 9000 that can be used to implement embodiments of the present disclosure is shown. Device 9000 can be implemented as... Figure 1The computing device 300, or the control platform 120 or BMS 112 of the vehicle 100. Device 9000 can be used to implement... Figure 2-4 and Figure 6-8 The method.

[0089] As shown in the figure, device 9000 includes a central processing unit (CPU) 9001, which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) 9002 or loaded from storage unit 9008 into random access memory (RAM) 9003. RAM 9003 can also store various programs and data required for the operation of device 9000, such as the measurement data mentioned above. CPU 9001, ROM 9002, and RAM 9003 are interconnected via bus 9004. Input / output (I / O) interface 9005 is also connected to bus 9004.

[0090] Multiple components in device 9000 are connected to I / O interface 9005, including: input unit 9006, such as keyboard, mouse, etc.; output unit 9007, such as various types of monitors, speakers, etc.; storage unit 9008, such as disk, optical disk, etc.; and communication unit 9009, such as network card, modem, wireless transceiver, etc. Communication unit 9009 allows device 9000 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0091] Processing unit 9001 performs the methods or processes described above, such as method 2000. For example, in some embodiments, method 2000 may be implemented as a computer software program or computer program product tangibly contained in a machine-readable medium, such as a non-transient computer-readable medium, such as storage unit 9008. In some embodiments, part or all of the computer program may be loaded and / or installed on device 9000 via ROM 9002 and / or communication unit 9009. When the computer program is loaded into RAM 9003 and executed by CPU 9001, one or more steps of method 2000 described above may be performed. Alternatively, in other embodiments, CPU 9001 may be configured to perform method 2000 by any other suitable means (e.g., by means of firmware).

[0092] Those skilled in the art will understand that the various steps of the methods disclosed above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using device-executable program code, which can then be stored in a storage device for execution by the computing device. Alternatively, they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this disclosure is not limited to any particular combination of hardware and software.

[0093] It should be understood that although several devices or sub-devices of the device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more devices described above can be embodied in one device. Conversely, the features and functions of one device described above can be further divided and embodied by multiple devices.

[0094] The above description is merely an optional embodiment of this disclosure and is not intended to limit this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for managing a battery pack, comprising: Acquire a first data set associated with multiple battery cells connected to each other, the first data set including electrical data of the multiple battery cells over a predetermined time period; Based on the state of charge of the plurality of battery cells during the predetermined time period, electrical data from at least a portion of the predetermined time period are selected from the first data set to generate a second data set; Based on the second data set, at least one set of characteristic values ​​of the plurality of battery cells are determined, the at least one set of characteristic values ​​including a set of characteristic values ​​associated with the internal resistance of the plurality of battery cells; as well as Based on the at least one set of feature values, determine whether any of the plurality of battery cells are abnormal. The determination of at least one set of feature values ​​for the plurality of battery cells based on the second data set includes: For each battery cell, Linear fitting is performed based on the voltage, current, and charge values ​​of the corresponding battery cell in the second dataset to obtain the internal resistance value of the corresponding battery cell, wherein the linear fitting uses the voltage value as the dependent variable and the current and charge values ​​as independent variables. Based on a set of internal resistance values ​​for the plurality of battery cells, the internal resistance Z fraction of the corresponding battery cell is determined. as well as A set of internal resistance Z fractions for the plurality of battery cells is determined as one set of characteristic values ​​in the at least one set of characteristic values.

2. The management method according to claim 1, wherein the at least one set of characteristic values ​​further includes a set of characteristic values ​​associated with the voltage fluctuation level of the plurality of battery cells, and The determination of at least one set of feature values ​​for the plurality of battery cells based on the second data set further includes: For each battery cell, Based on the voltage value of the corresponding battery cell in the second data set, the voltage entropy of the corresponding battery cell is obtained, and Based on a set of voltage entropies for the plurality of battery cells, the voltage entropy Z fraction of the corresponding battery cell is determined; as well as A set of voltage entropy Z fractions for the plurality of battery cells is determined as a set of characteristic values ​​in the at least one set of characteristic values.

3. The management method according to claim 1, wherein generating a second data set by selecting electrical data from at least a portion of the predetermined time period based on the state of charge of the plurality of battery cells during the predetermined time period comprises: Obtain the state of charge of each of the plurality of battery cells during the predetermined time period; as well as A second data set is generated by selecting electrical data from the first data set during a time period in which the state of charge of the plurality of battery cells is within a predetermined range.

4. The management method according to claim 3, wherein the predetermined range is from 35% to 75% of the state of charge.

5. The management method according to claim 3, wherein generating the second data set by selecting electrical data from at least a portion of the predetermined time period based on the state of charge of the plurality of battery cells during the predetermined time period further comprises: Electrical data for at least a portion of the predetermined time period is selected from the first data set based on additional conditions, the additional conditions including at least one of the following: The voltage and current values ​​of the plurality of battery cells are not empty during the selected time period; and The absolute value of the current value of the plurality of battery cells is greater than zero during the selected time period.

6. The management method according to claim 1, further comprising, before determining at least one set of characteristic values ​​of the plurality of battery cells based on the second data set: Determine the amount of data in the second data set and / or the maximum change in current values ​​in the second data set; If the data volume is lower than the data volume threshold and / or the maximum change is lower than the change threshold, the second data set is determined to be invalid and the second data set is discarded. as well as If the amount of data exceeds the data amount threshold and / or the maximum change exceeds the change threshold, the second data set is determined to be valid.

7. The management method of claim 1, wherein obtaining a first data set associated with a plurality of battery cells connected to each other comprises: Receive the sampling data sequence associated with the plurality of battery cells; The average current is calculated based on the current values ​​in the sampled data sequence whose absolute values ​​are greater than a current threshold, wherein the current threshold is greater than zero. The duration of the predetermined time period is calculated based on the calculated average current and the specified rate of change of state of charge. as well as The sampled data in the sampled data sequence that falls within the predetermined time period is preprocessed to generate the first data set.

8. The management method according to claim 7, wherein preprocessing the sampled data in the sampled data sequence that falls within the predetermined time period to generate the first data set comprises: Obtain a set of voltage values ​​of the plurality of battery cells during the predetermined time period; Obtain a set of current values ​​for the plurality of battery cells during the predetermined time period; A set of charge values ​​for the plurality of battery cells is calculated based on the set of current values; as well as The first data set is generated based on the set of voltage values, the set of current values, and the set of electrical charge values.

9. The management method according to claim 1, wherein determining whether there is an abnormality in the plurality of battery cells based on the at least one set of feature values ​​includes: An outlier algorithm is used to determine at least one feature value threshold corresponding to each of the at least one set of feature values; If at least one set of feature values ​​contains a feature value that exceeds the corresponding feature value threshold, increment the counter by one. as well as If the counter count exceeds the counting threshold, it is determined that there is an anomaly in the plurality of battery cells.

10. An electronic device, comprising: processor; as well as A memory coupled to the processor, the memory having instructions stored therein, the instructions causing the device to perform actions when executed by the processor, the actions including: Acquire a first data set associated with multiple battery cells connected to each other, the first data set including electrical data of the multiple battery cells over a predetermined time period; Based on the state of charge of the plurality of battery cells during the predetermined time period, electrical data from at least a portion of the predetermined time period are selected from the first data set to generate a second data set; Based on the second data set, at least one set of characteristic values ​​for the plurality of battery cells is determined, the at least one set of characteristic values ​​including a set of characteristic values ​​associated with the internal resistance of the plurality of battery cells; and Based on the at least one set of feature values, it is determined whether the plurality of battery cells are abnormal, wherein determining the at least one set of feature values ​​of the plurality of battery cells based on the second data set includes: For each battery cell, Linear fitting is performed based on the voltage, current, and charge values ​​of the corresponding battery cell in the second dataset to obtain the internal resistance value of the corresponding battery cell, wherein the linear fitting uses the voltage value as the dependent variable and the current and charge values ​​as independent variables. Based on a set of internal resistance values ​​for the plurality of battery cells, determine the internal resistance Z fraction of the corresponding battery cell; and A set of internal resistance Z fractions for the plurality of battery cells is determined as one set of characteristic values ​​in the at least one set of characteristic values.

11. The electronic device of claim 10, wherein the at least one set of characteristic values ​​further includes a set of characteristic values ​​associated with the degree of voltage fluctuation of the plurality of battery cells, and The determination of at least one set of feature values ​​for the plurality of battery cells based on the second data set further includes: For each battery cell, Based on the voltage value of the corresponding battery cell in the second data set, the voltage entropy of the corresponding battery cell is obtained, and Based on a set of voltage entropies for the plurality of battery cells, the voltage entropy Z fraction of the corresponding battery cell is determined; as well as A set of voltage entropy Z fractions for the plurality of battery cells is determined as a set of characteristic values ​​in the at least one set of characteristic values.

12. The electronic device of claim 10, wherein generating the second data set by selecting electrical data from at least a portion of the predetermined time period based on the state of charge of the plurality of battery cells during the predetermined time period comprises: Obtain the state of charge of each of the plurality of battery cells during the predetermined time period; as well as A second data set is generated by selecting electrical data from the first data set during a time period in which the state of charge of the plurality of battery cells is within a predetermined range.

13. The electronic device of claim 12, wherein the predetermined range is from 35% to 75% of the state of charge.

14. The electronic device of claim 12, wherein generating the second data set by selecting electrical data from at least a portion of the predetermined time period based on the state of charge of the plurality of battery cells during the predetermined time period further comprises: Electrical data for at least a portion of the predetermined time period is selected from the first data set based on additional conditions, the additional conditions including at least one of the following: The voltage and current values ​​of the plurality of battery cells are not empty during the selected time period; and The absolute value of the current value of the plurality of battery cells is greater than zero during the selected time period.

15. The electronic device of claim 10, further comprising, before determining at least one set of characteristic values ​​of the plurality of battery cells based on the second data set: Determine the amount of data in the second data set and / or the maximum change in current values ​​in the second data set; If the data volume is lower than the data volume threshold and / or the maximum change is lower than the change threshold, the second data set is determined to be invalid and the second data set is discarded. as well as If the amount of data exceeds the data amount threshold and / or the maximum change exceeds the change threshold, the second data set is determined to be valid.

16. The electronic device of claim 10, wherein acquiring a first data set associated with a plurality of battery cells connected to each other comprises: Receive the sampling data sequence associated with the plurality of battery cells; The average current is calculated based on the current values ​​in the sampled data sequence whose absolute values ​​are greater than a current threshold, wherein the current threshold is greater than zero. The duration of the predetermined time period is calculated based on the calculated average current and the specified rate of change of state of charge. as well as The sampled data in the sampled data sequence that falls within the predetermined time period is preprocessed to generate the first data set.

17. The electronic device of claim 16, wherein preprocessing the sampled data in the sampled data sequence that falls within the predetermined time period to generate the first data set comprises: Obtain a set of voltage values ​​of the plurality of battery cells during the predetermined time period; Obtain a set of current values ​​for the plurality of battery cells during the predetermined time period; A set of charge values ​​for the plurality of battery cells is determined based on the set of current values; as well as The first data set is generated based on the set of voltage values, the set of current values, and the set of electrical charge values.

18. The electronic device of claim 10, wherein determining whether the plurality of battery cells are abnormal based on the at least one set of feature values ​​comprises: An outlier algorithm is used to determine at least one feature value threshold corresponding to each of the at least one set of feature values; If at least one set of feature values ​​contains a feature value that exceeds the corresponding feature value threshold, increment the counter by one. as well as If the counter count exceeds the counting threshold, it is determined that there is an anomaly in the plurality of battery cells.

19. A vehicle comprising: A battery pack, comprising multiple battery cells connected to each other; as well as The electronic device according to any one of claims 10-18.

20. A computer-readable medium having computer-readable instructions stored thereon, which, when executed by a processing unit, cause the processing unit to perform the management method according to any one of claims 1-9.

Images