Battery management device and battery management method

Abstract

According to some embodiments, a battery management device comprises: an interface configured to obtain battery data of a battery; and a controller configured to calculate a first SOH value of the battery on the basis of the battery data, calculate a first SOH correction value for the first SOH value by using a regression equation for calculating a SOH correction value on the basis of error factors causing errors in the first SOH value, and calculate a second SOH value by applying the first SOH correction value to the first SOH value.

Description

Battery management device and battery management method

[0001] Cross-citation with related applications

[0002] This application claims the benefit of priority to Korean Patent Application No. 10-2024-0142672, filed October 18, 2024, and Korean Patent Application No. 10-2024-0066780, filed May 22, 2024, the entire contents of which are incorporated herein by reference.

[0003] Technology field

[0004] The embodiments disclosed in this document relate to a battery management device and a battery management method.

[0005] Recently, active research and development has been conducted on secondary batteries. The term "secondary battery" refers to a rechargeable battery, encompassing both conventional Ni / Cd and Ni / MH batteries, as well as recent lithium-ion batteries. Among secondary batteries, lithium-ion batteries can boast higher energy densities than conventional Ni / Cd and Ni / MH batteries. They can be manufactured in small and lightweight designs, making them highly versatile power sources for mobile devices. Recently, their use has expanded to include power sources for electric vehicles, attracting attention as a next-generation energy storage medium.

[0006] Using the charging current during battery charging and the state of charge (SOC) values ​​before and after charging, the corresponding state of health (SOH) value for the battery can be calculated. The SOC values ​​used to calculate SOH can be derived through a mapping relationship with the open circuit voltage (OCV). If variables such as charging temperature are not considered during the SOH calculation process, SOH errors due to temperature differences can occur. In addition, errors in the OCV-SOC mapping relationship itself can also result in errors in the SOH value.

[0007] One of the purposes of the embodiments disclosed in this document is to provide a battery management device and a battery management method capable of estimating a more accurate SOH value by correcting an error that occurs when calculating an SOH value using charging current and charging SOC values.

[0008] The technical objectives of the embodiments disclosed in this document are not limited to the technical tasks mentioned above, and other technical tasks not mentioned will be clearly understood by those skilled in the art from the descriptions below.

[0009] According to some embodiments, a battery management device includes an interface configured to obtain battery data of a battery; and a controller configured to calculate a first SOH value of the battery based on the battery data, calculate a first SOH correction value for the first SOH value using a regression equation that calculates an SOH correction value based on error factors that cause an error in the first SOH value, and calculate a second SOH value by applying the first SOH correction value to the first SOH value.

[0010] According to some embodiments, the battery data includes charge data that is accumulated and collected each time the charge of the battery satisfies a valid condition, and the controller is configured to calculate the second SOH value based on the charge data.

[0011] According to some embodiments, the charge data includes charge open circuit voltage (OCV) data and charge current data for each of the valid charges satisfying the validity condition within the period of interest, and the controller is configured to calculate the first SOH value for each of the valid charges based on SOC data corresponding to the charge OCV data and the charge current data.

[0012] According to some embodiments, the regression coefficients of the regression equation for the error factors are learned to calculate the first SOH correction value based on the first SOH values ​​for the effective charges.

[0013] According to some embodiments, a first SOH deviation value of each of the first SOH values ​​relative to a representative value of the first SOH values ​​is set as a residual of the regression equation, and the regression coefficients are learned to minimize the residuals of the regression equation.

[0014] According to some embodiments, the error factors include at least one of a charge start SOC according to the SOC data, a charge end SOC according to the SOC data, a charge start temperature, and a charge end temperature.

[0015] According to some embodiments, the regression equation is constructed based on a multivariate polynomial that includes terms composed of combinations of the error factors.

[0016] According to some embodiments, the controller is configured to calculate first SOH correction values ​​for the effective charges by applying the error factors for each of the effective charges to the learned regression coefficients of the regression equation.

[0017] According to some embodiments, the validity condition includes a first condition regarding whether the charge amount exceeds a threshold amount and a second condition regarding whether the charge pause time exceeds a threshold time.

[0018] According to some embodiments, the learning of the regression coefficients is performed based on a lightweight gradient boosting machine (light GBM) or linear regression.

[0019] According to some embodiments, a battery management method includes the steps of: obtaining battery data of a battery; calculating a first SOH value of the battery based on the battery data; calculating a first SOH correction value for the first SOH value using a regression equation that calculates an SOH correction value based on error factors that cause an error in the first SOH value; and calculating a second SOH value by applying the first SOH correction value to the first SOH value.

[0020] According to some embodiments, the battery data includes charge data that is accumulated each time the charge of the battery satisfies a valid condition, and the second SOH value is calculated based on the charge data.

[0021] According to some embodiments, the charge data includes charge open circuit voltage (OCV) data and charge current data for each of the valid charges satisfying the validity condition within the period of interest, and the step of calculating the first SOH value includes the step of calculating the first SOH value for each of the valid charges based on SOC data corresponding to the charge OCV data and the charge current data.

[0022] According to some embodiments, the regression coefficients of the regression equation for the error factors are learned to calculate the first SOH correction value based on the first SOH values ​​for the effective charges.

[0023] According to some embodiments, a first SOH deviation value of each of the first SOH values ​​relative to a representative value of the first SOH values ​​is set as a residual of the regression equation, and the regression coefficients are learned to minimize the residuals of the regression equation.

[0024] According to some embodiments, the error factors include at least one of a charge start SOC according to the SOC data, a charge end SOC according to the SOC data, a charge start temperature, and a charge end temperature.

[0025] According to some embodiments, the regression equation is constructed based on a multivariate polynomial that includes terms composed of combinations of the error factors.

[0026] According to some embodiments, the step of calculating the first SOH correction value includes the step of calculating the first SOH correction values ​​for the valid charges by applying the error factors for each of the valid charges to the learned regression coefficients of the regression equation.

[0027] According to some embodiments, the validity condition includes a first condition regarding whether the charge amount exceeds a threshold amount and a second condition regarding whether the charge pause time exceeds a threshold time.

[0028] According to some embodiments, the learning of the regression coefficients is performed based on a lightweight gradient boosting machine (light GBM) or linear regression.

[0029] According to the embodiments disclosed in this document, a battery management device and a battery management method can be provided that can estimate a more accurate SOH value by correcting an error that occurs when calculating an SOH value using charging current and charging SOC values.

[0030] The technical effects according to the embodiments disclosed in this document are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art according to the disclosure of this document.

[0031] FIG. 1 illustrates elements constituting a battery management system according to some embodiments.

[0032] FIG. 2 illustrates elements constituting a battery management device according to some embodiments.

[0033] FIG. 3 illustrates a process for calculating SOH values ​​using charging current and charging SOC values ​​according to some embodiments.

[0034] FIG. 4 illustrates a method of accumulating charge data whenever the charging of a battery satisfies a valid condition according to some embodiments.

[0035] FIG. 5 illustrates charging data that is accumulated each time a battery's charge satisfies a valid condition according to some embodiments.

[0036] Figure 6 illustrates SOH values ​​calculated based on charging data according to some embodiments.

[0037] FIG. 7 illustrates a machine learning model utilized to correct SOH values ​​based on charging data according to some embodiments.

[0038] FIG. 8 illustrates a method for setting a regression equation learned by a machine learning model to correct SOH values ​​based on charging data according to some embodiments.

[0039] FIG. 9 illustrates a process for calculating a first SOH correction value and a second SOH value using a regression equation according to some embodiments.

[0040] Figure 10 illustrates the difference between the variation of SOH according to the error factor before correction and the variation of SOH according to the error factor after correction according to some embodiments.

[0041] Figure 11 illustrates the difference between the SOH distribution before correction and the SOH distribution after correction according to some embodiments.

[0042] FIG. 12 illustrates steps of a battery management method according to some embodiments.

[0043] Hereinafter, embodiments described in this document are described with reference to the attached drawings. However, this is not intended to limit the disclosure of this document to specific embodiments, and it should be understood that various modifications, equivalents, and / or alternatives of the embodiments described in this document are included.

[0044] The embodiments and terminology used in this document are not intended to limit the technical features described in this document to a specific embodiment, but should be understood to encompass various modifications, equivalents, or alternatives of the embodiment. In connection with the description of the drawings, similar reference numerals may be used to refer to similar or related components. The singular form of a noun corresponding to an item may include one or more of the item, unless the context clearly indicates otherwise.

[0045] In this document, the phrases "A or B", "at least one of A and B", "at least one of A or B", "A, B, or C", "at least one of A, B, and C", and "at least one of A, B, or C" can each include any one of the items listed together in that phrase, or all possible combinations thereof. Terms such as "first", "second", "first", "second", "A", "B", "(a)", or "(b)" may be used merely to distinguish the corresponding component from other corresponding components, and do not limit the corresponding components in any other respect (e.g., importance or order) unless specifically stated otherwise.

[0046] In this document, when a component (e.g., a first component) is referred to as being “connected,” “coupled,” or “connected,” with or without the terms “functionally” or “communicatively,” or “coupled” or “connected,” it means that the component can be connected to the other component directly (e.g., wired or wirelessly), or indirectly (e.g., via a third component).

[0047] The methods according to various embodiments disclosed in this document may be provided as a computer program product. The computer program product may be traded as a commodity between a seller and a buyer. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., compact disc read-only memory, CD-ROM), or may be distributed online (e.g., downloaded or uploaded) through an application store or directly between two driver devices. In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily generated in a machine-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0048] According to the embodiments disclosed in this document, each component (e.g., a module or a program) of the above-described components may include one or more entities, and some of the entities may be separated and placed in other components. According to the embodiments disclosed in this document, one or more components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to the integration. According to the embodiments disclosed in this document, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0049] FIG. 1 may illustrate elements constituting a battery management system according to some embodiments.

[0050] Referring to FIG. 1, a battery management system (100) may include a power usage device (110), a battery (120), and a battery management device (130). However, the present invention is not limited thereto, and some components may be omitted from the battery management system (100) or other general-purpose components may be further included in the battery management system (100).

[0051] The power usage device (110) may be configured to charge or discharge the battery (120). The power usage device (110) may discharge the battery (120) while consuming power, and may charge the battery (120) while generating power. In an embodiment, the power usage device (110) may include a mobility device such as an electric vehicle (EV), a hybrid electric vehicle (HEV), or an electric bike. The mobility device may drive a motor based on the power of the battery (120) or charge the battery (120) with power generated through regenerative braking. In an embodiment, the power usage device (110) may include a charger / discharger that provides charge / discharge cycles to the battery (120).

[0052] The battery (120) may include a battery pack or the like that is a diagnostic target of the system (100). The battery pack of the battery (120) may include a plurality of battery modules, and each battery module may include a plurality of battery cells. According to an embodiment, the battery (120) may be mounted on various types of mobility devices.

[0053] The battery management device (130) can perform operations for diagnosing or managing the battery (120). The battery management device (130) can acquire battery data of the battery (120) and diagnose or manage the status of the battery (120) to be diagnosed based on the battery data. According to an embodiment, the battery management device (130) can include a battery management system (BMS) configured together with the battery (120) in an on-board manner, and / or an external device remotely placed from the battery (120) in an off-board manner. The external device can include a charger of a battery charging station, a battery diagnostic device, a cloud computing server, etc.

[0054] The battery management system (100) may further include a management server. The management server may manage the management results of the battery management device (130). The management server may exchange data with the battery management device (130) via wired / wireless communication. When a defect in the battery (120) is diagnosed or its lifespan is predicted, the results may be transmitted to the management server and recorded in a database. According to an embodiment, the battery management device (130) may perform diagnostic operations by executing battery management software, and the management server may provide update information of the battery management software to the battery diagnostic device (130).

[0055] FIG. 2 may illustrate elements constituting a battery management device according to some embodiments.

[0056] Referring to FIG. 2, the battery management device (130) may include an interface (131) and a controller (132). However, the present invention is not limited thereto, and some components may be omitted from the battery management device (130), or other general-purpose components may be further included in the battery management device (130).

[0057] The interface (131) can obtain battery data of the battery (120). According to an embodiment, the interface (131) may include a communication unit configured to receive battery data and / or a sensor unit configured to measure battery data. According to an embodiment, when the battery management device (130) is implemented in an off-board form, the communication unit may receive battery data in a wired data communication, wireless data communication, or the like. Alternatively, when the battery management device (130) is implemented in an on-board form, the sensor unit may be configured to measure values ​​such as voltage, current, temperature, and resistance from the battery (120).

[0058] The controller (132) may have a structure for executing commands that implement the operations of the battery management device (130). The controller (132) may be implemented as an array of multiple logic gates for processing various operations or as a general-purpose microprocessor, and may be composed of a single processor or multiple processors. For example, the controller (132) may be implemented in the form of at least one of a microprocessor, a CPU, a GPU, and an AP.

[0059] The controller (132) can operate with a memory configured to store various data, commands, mobile applications, computer programs, etc. The memory can be configured separately from or integrally with the controller (132). The controller (132) can process various operations by executing commands stored in the memory. For example, the memory can be implemented as a non-volatile device such as a ROM, a PROM, an EPROM, an EEPROM, a flash memory, a PRAM, an MRAM, an RRAM, an FRAM, etc., or a volatile device such as a DRAM, an SRAM, an SDRAM, a PRAM, etc., and can be implemented in the form of an HDD, an SSD, an SD, a Micro-SD, etc., or a combination thereof.

[0060] The interface (131) may be configured to obtain battery data of the battery (120). According to an embodiment, the interface (131) may include a communication unit configured to receive the battery data and / or a sensor unit configured to measure the battery data. According to an embodiment, when the battery management device (130) is implemented in an off-board form, the communication unit may receive the battery data in a manner such as wired data communication, wireless data communication, etc. Alternatively, when the battery management device (130) is implemented in an on-board form, the sensor unit may be configured to measure values ​​such as voltage, current, temperature, and resistance from the battery (120). According to an embodiment, the battery data may include charge data related to charging of the battery (120) and / or discharge data related to discharging.

[0061] The controller (132) may be configured to calculate a first SOH value of the battery (120) based on battery data. In an embodiment, the battery data may include data regarding charging and / or discharging of the battery (120), and the first SOH value may be an SOH value calculated using charging current and charging SOC values. The first SOH value may be raw data prior to correction for errors in the SOH value caused by error factors.

[0062] The controller (132) may be configured to calculate a first SOH correction value for the first SOH value using a regression equation that calculates an SOH correction value based on error factors that cause an error in the first SOH value. According to an embodiment, the regression equation may have error factors as independent variables or input variables, and may have the first SOH correction value as a dependent variable or output variable. The regression equation may be configured in the form of a multivariate polynomial, and the regression coefficients for each term of the multivariate polynomial may be learned through various machine learning techniques. According to an embodiment, the error factors may include variables such as voltage, current, temperature, and resistance related to charging and / or discharging of the battery (120). According to an embodiment, the first SOH correction value may be set as a deviation value with respect to an average value or a median value of the first SOH value for a certain period of time, and may be adjusted to a specific value according to learning of the regression equation.

[0063] The controller (132) may be configured to calculate a second SOH value by applying a first SOH correction value to the first SOH value. In an embodiment, the second SOH value may be calculated by subtracting the first SOH correction value from the first SOH value. The second SOH value may be a result of correcting for errors due to error factors present in the first SOH value.

[0064] According to an embodiment, the battery data may include charging data that is accumulated and collected whenever the charging of the battery (120) satisfies a validity condition, and the controller (132) may be configured to calculate a second SOH value based on the charging data. The validity condition is for ensuring data validity, and may include conditions regarding the amount of charging and / or discharging of the battery (120), conditions regarding the time point of data collection, etc. For example, among data regarding charging, discharging, SOC fluctuation, etc. of the battery (120) for 3 months, 6 months, 1 year, or other periods, data that satisfies the validity condition may be collected as charging data. Calculation of the first SOH value and learning of a regression equation may be performed based on the charging data.

[0065] According to an embodiment, the charge data may include charge open circuit voltage (OCV) data and charge current data of each valid charge that satisfies a valid condition within a period of interest, and the controller (132) may be configured to calculate a first SOH value for each valid charge based on SOC data and charge current data corresponding to the charge OCV data. The valid charges may include charges that satisfy conditions regarding a charge amount, conditions regarding a collection point in time of OCV data, etc. among charges of the battery (120) within the period of interest. The SOC data corresponding to the charge OCV data may be derived based on a pre-prepared OCV-SOC mapping profile, etc. According to an embodiment, the first SOH value may be calculated based on a charge start SOC, a charge end SOC, and a charge current of each valid charge.

[0066] According to an embodiment, regression coefficients of a regression equation for error factors can be learned to calculate a first SOH correction value based on first SOH values ​​for valid charges. The regression equation can have error factors as independent / input variables and the first SOH correction value as a dependent / output variable. Terms of the regression equation can be set based on error factors, and regression coefficients for the terms of the regression equation can be adjusted through learning of the regression equation. The first SOH correction value can be calculated based on the learned regression coefficients.

[0067] According to an embodiment, the first SOH deviation value of each of the first SOH values ​​with respect to the representative value of the first SOH values ​​may be set as the residual of the regression equation, and the regression coefficients may be learned to minimize the residuals of the regression equation. For example, if there are n valid charges within the period of interest, there may be n first SOH values ​​corresponding to each of them. The mean, median, mode, etc. for the n first SOH values ​​may be set as the representative value, and the first SOH deviation value may be the difference between each first SOH value and the representative value. According to an embodiment, before learning the regression equation, the first SOH deviation value may be set as the output and dependent variable of the regression equation, and when learning the regression equation is completed, the first SOH deviation value may function as the first SOH correction value. For example, a regression equation can be set to output a first SOH deviation value by applying regression coefficients to terms formed through a combination of error factors, wherein the first SOH deviation value can function as a residual of the regression equation. The regression coefficients can be trained in a direction that minimizes the residual sum of squares (RSS) of the n first SOH deviation values, or other regression learning techniques other than the residual sum of squares (RSS) can be utilized.

[0068] According to an embodiment, the error factors may include at least one of a charge start SOC based on SOC data, a charge end SOC based on SOC data, a charge start temperature, and a charge end temperature. For example, the charge start temperature is not used in the calculation process of the first SOH value, but the first SOH value may vary depending on the charge start temperature when other conditions are the same. Accordingly, the charge start temperature may correspond to an error factor of the first SOH value, and the remaining charge start SOC, charge end SOC, and charge end temperature may also correspond to error factors in a similar manner.

[0069] In an embodiment, the regression equation can be constructed based on a multivariate polynomial that includes terms that are combinations of error factors. For example, if the error factors are the start SOC of charge (x1) and the end SOC of charge (x2), and the degree of the multivariate polynomial is 3, the possible combinations of error factors are x1 3 , x2 3 , x1 2 *x2, x1*x2 2 , x1 2 , x1*x2, x2 2 , x1, x2, constant term, etc. These 10 terms can form a multivariate polynomial of the regression equation, and 10 regression coefficients for this can be learned. Unlike this example, the number of error factors and the degree of the multivariate polynomial can be set differently. Since various combinations of error factors are utilized in the form of a multivariate polynomial, the error factors interact with each other and influence each other, so that factors that cause errors in the first SOH value can be reflected in the SOH correction process.

[0070] According to an embodiment, the controller (132) may be configured to calculate first SOH correction values ​​for the valid charges by applying error factors for each of the valid charges to the learned regression coefficients of the regression equation. For example, the regression equation may be a two-variable third-order polynomial with terms x1 3 , x2 3 , x1 2 *x2, x1*x2 2 , x1 2 , x1*x2, x2 2 , x1, x2, and a constant term, the 10 regression coefficients can be adjusted through learning, and when learning for the 10 regression coefficients is completed, the first SOH deviation value corresponding to the difference between the first SOH value and the representative value can be calculated by substituting x1 and x2 into a two-variable third-order polynomial. The first SOH deviation value calculated by the learned regression equation can be the first SOH correction value.

[0071] According to an embodiment, the validity condition may include a first condition regarding whether the amount of charge exceeds a threshold amount and a second condition regarding whether the charging pause time exceeds the threshold time. For example, the first condition may be a condition where the amount of charge is greater than or equal to 25% of the dod. A dod of greater than or equal to 25% may mean that the difference between the charge end SOC and the charge start SOC is greater than or equal to 25%. The second condition may be a condition where the charge pause time is greater than or equal to 2 hours. The charge pause time may mean a pause time from the completion of charging until the voltage is measured. If the pause time is not sufficient, it may be difficult to regard the measured voltage as OCV. If the pause time is sufficient, the measured voltage may be regarded as OCV. The exemplary values ​​such as 25% or more in the first condition and 2 hours or more in the second condition may be changed to different values.

[0072] In an embodiment, learning of regression coefficients may be performed based on a light gradient boosting machine (light GBM) or linear regression. Regression coefficients for terms of a multivariate polynomial generated by a combination of error factors may be learned based on a light gradient boosting machine (light GBM) or linear regression. The regression coefficients may be learned in a direction that minimizes the residual sum of squares (RSS) or a similar metric.

[0073] FIG. 3 illustrates a process for calculating SOH values ​​using charging current and charging SOC values ​​according to some embodiments.

[0074] Referring to FIG. 3, a formula (310) illustrating a process of calculating an SOH value using charging current and charging SOC values ​​and a table (320) for explaining variables calculated in the formula (310) can be illustrated.

[0075] In formula (310), the first SOH value is the charge start SOC (SOC start ), charging end SOC (SOC end ), charging current (I) and the capacity of the battery (120). The first SOH value can be calculated based on the formula (310) for effective charging that satisfies the effective condition. The charging current (I) can be integrated while effective charging is in progress.

[0076] Table (320) may illustrate an example of calculating the first SOH value based on the formula (310) in different states. The start-of-charge OCV, start-of-charge SOC, end-of-charge OCV, and end-of-charge SOC in the beginning-of-life (BOL) state and the end-of-life (EOL) state may be the same, but the integrated value of the charge current (I) may differ due to battery aging, which may lead to a difference in the first SOH value.

[0077] Although equation (310) does not consider temperature factors, the charging temperature may be a factor causing an error in the first SOH value. In addition, equation (310) only considers the difference between the charge start SOC and the charge end SOC, but if the charge start SOC or the charge end SOC is different for the same difference value, the first SOH value may fluctuate. Therefore, the charge start SOC or the charge end SOC may also be an error factor. In addition to these, when calculating the first SOH value based on equation (310), various factors causing an error in the first SOH value may be utilized as error factors in the regression equation.

[0078] FIG. 4 illustrates a method of accumulating charge data whenever the charging of a battery satisfies a valid condition according to some embodiments.

[0079] Referring to FIG. 4, a graph (400) may be illustrated illustrating a method of accumulating charging data whenever the charging of a battery (120) satisfies a valid condition. The horizontal axis of the graph (400) may represent time.

[0080] In the graph (400), the voltage measured after a rest time of more than 2 hours has elapsed can be treated as an OCV value, and the OCV value can be converted into an SOC value through a mapping table, etc. The SOC value converted from the OCV value can be recorded as one point in the graph (400). According to an embodiment, the measurement of the OCV value and the conversion of the SOC value can be performed whenever the use of the power usage device (110) is terminated or whenever the charging of the power usage device (110) is completed. When the power usage device (110) is a mobility device, as in the graph (400), the SOC can gradually decrease as the mobility device drives and then increase significantly due to the charging of the mobility device.

[0081] If the SOC difference between adjacent SOC values ​​exceeds, for example, 25% or another appropriate value, the SOC difference may be recorded as one cycle. Multiple cycles recorded over a period of 3 months, 6 months, 1 year, or other period of interest may form charging data. A first SOH value may be calculated for each of the multiple cycles of charging data. For example, if there are n cycles over a period of interest, n first SOH values ​​may be calculated, and through an SOH correction process, one or more but no more than n second SOH values ​​may be calculated.

[0082] FIG. 5 illustrates charging data that is accumulated each time a battery's charge satisfies a valid condition according to some embodiments.

[0083] Referring to FIG. 5, a table (500) may be illustrated illustrating charging data that is accumulated whenever the charging of a battery (120) satisfies a valid condition. The table (500) may represent n cycles collected during a period of interest.

[0084] Based on the start voltage (OCV_s) and end voltage (OCV_e) recorded in the cycle of each row of the table (500), the start SOC (SOC_s) and end SOC (SOC_e) can be calculated, and the current integration (CI) based on the charging current for the cycle of each row can be calculated. For the cycle of each row, the first SOH value (SOH_raw) can be calculated based on the start SOC (SOC_s), end SOC (SOC_e) and the current integration (CI). When n cycles are collected during the period of interest, n first SOH values ​​(SOH_raw) for these can be calculated.

[0085] The charging data of the table (500) may include the starting temperature (T_s) and the ending temperature (T_e) of each charging cycle. The starting temperature (T_s) and the ending temperature (T_e) may not be reflected in the calculation process of the first SOH value (SOH_raw). However, the starting temperature (T_s) and the ending temperature (T_e) may be error factors that affect the calculation result of the first SOH value (SOH_raw). The error due to the starting temperature (T_s) and the ending temperature (T_e) may be reflected in the learning process of the regression equation, and the learned regression equation may output the first SOH correction value to correct the error due to the starting temperature (T_s) and the ending temperature (T_e).

[0086] Figure 6 illustrates SOH values ​​calculated based on charging data according to some embodiments.

[0087] Referring to FIG. 6, a first graph (610) and a second graph (620) illustrating SOH values ​​calculated based on charging data may be illustrated. The horizontal axis of the first graph (610) and the second graph (620) may represent the driving distance of a battery-equipped vehicle, and the vertical axis may represent the battery SOH.

[0088] The first graph (610) and the second graph (620) may be graphs for different first and second vehicles. In both the first graph (610) and the second graph (620), a tendency for the SOH to decrease as the driving distance increases can be confirmed. The vertical axis SOH of the first graph (610) and the second graph (620) may correspond to the first SOH value in a state in which SOH correction by the battery management device (130) is not performed. In both the first graph (610) and the second graph (620), a high dispersion of the SOH values ​​may be shown. The SOH dispersion may appear higher as the depth of discharge (dod) value is lower. For example, dod25 may represent SOH values ​​calculated for data in which the SOC difference due to charging is 25% or more, and dod45 and dod65 may also represent similar meanings.

[0089] FIG. 7 illustrates a machine learning model utilized to correct SOH values ​​based on charging data according to some embodiments.

[0090] Referring to FIG. 7, the ML model (700) can train the regression coefficients of the regression equation (X_polynomial) based on the learning data (y_train), and output the predicted data (y_predict) using the trained regression equation (X_polynomial).

[0091] The regression equation (X_polynomial) can take the form of a multivariate polynomial. The terms of the multivariate polynomial can be set based on the combination of error factors that cause errors in the first SOH value. For example, if the error factors are the start SOC of charging (x1) and the end SOC of charging (x2), and the degree of the multivariate polynomial is 2, the possible combinations of error factors are x1 2 , x2 2, x1*x2, x1, x2, constant term, etc. In this case, six regression coefficients β1~β6 for six terms can form a multivariate polynomial, and the dependent / output variable of the regression equation (X_polynomial) can be training data (y_train) or prediction data (y_predict).

[0092] [Mathematical Formula 1]

[0093] β1*x1 2 + β2*x2 2 + β3*x1*x2 + β4*x1 + β5*x2 + β6*c = y

[0094] Mathematical expression 1 may represent the form of an exemplary regression equation. Before learning of regression coefficients is completed, y may be training data (y_train), and after learning of regression coefficients is completed, y may be prediction data (y_predict). According to an embodiment, the training data (y_train) or the prediction data (y_predict) may be a first SOH deviation value corresponding to the difference between the representative value of n first SOH values ​​and each first SOH value, and the first SOH deviation value may function as a residual of the regression equation. For example, the regression coefficients may be learned in a direction that minimizes the residual sum of squares (RSS), and a lightweight gradient boosting machine (light GBM), linear regression, or any other appropriate technique may be utilized to learn the regression equation. When the adjustment of the regression coefficients β1 to β6 is completed through learning, when the input variables (x1, x2) are input into the regression equation, the predicted data (y_predict) can be output, and at this time, the predicted data (y_predict) can be the first SOH correction value.

[0095] FIG. 8 illustrates a method for setting a regression equation learned by a machine learning model to correct SOH values ​​based on charging data according to some embodiments.

[0096] Referring to FIG. 8, a first table (810) and a second table (820) may be illustrated to illustrate a method of setting a regression equation learned by a machine learning model to correct SOH values ​​based on charging data.

[0097] The first table (810) may indicate the types of error factors that cause errors in the first SOH value. For example, the error factors of the first table (810) may include a charge start SOC, a charge end SOC, a charge start temperature, a charge end temperature, etc. The terms of the regression equation may be set based on a combination of the error factors of the first table (810). The second table (820) may indicate an example of setting the terms of the regression equation based on a combination of four error factors (x1 to x4). In the second table (820), the terms of the regression equation may include a linear term, a quadratic term, a cubic term, a quartic term, etc. When the number of error factors is 4, the regression equation may be a polynomial of degree 4 or higher. The number of error factors of the first table (810) and the terms of the second table (820) are all exemplary, and a regression equation of a different form may be utilized for SOH correction.

[0098] As in the second table (820), the terms of the regression equation may include terms generated by a combination of two or more error factors, such as x1*x2, x3*x4, etc., which may be a difference from the single-variable polynomial regression or multivariable linear regression of the regression equation. Therefore, through learning the regression equation, the error that two or more error factors interact with each other and induce in the first SOH value can be considered. Accordingly, when the SOH correction procedure using the single-variable polynomial regression equation is repeated four times for each of the four error factors, the error caused by the interaction of two or more error factors is not considered, whereas the regression equation in the form of a polynomial equation based on the combination of the four error factors can perform a more accurate SOH correction by taking this into account.

[0099] FIG. 9 illustrates a process for calculating a first SOH correction value and a second SOH value using a regression equation according to some embodiments.

[0100] Referring to FIG. 9, a table (900) illustrating a process of calculating a first SOH correction value and a second SOH value using a regression equation may be illustrated. The table (900) may represent n valid charges.

[0101] A first SOH representative value (SOH_raw_rep) can be calculated for n first SOH values ​​(SOH_raw_1 to SOH_raw_n). The first SOH representative value (SOH_raw_rep) can be a mean, a median, a mode, etc. Based on the difference with the first SOH representative value (SOH_raw_rep), n first SOH deviation values ​​(y_raw_1 to y_raw_n) can be calculated. According to an embodiment, the n first SOH deviation values ​​(y_raw_1 to y_raw_n) can function as residuals of a regression equation.

[0102] Regression learning can be performed in the direction of minimizing n residual sums of squares (RSS) or similar indices for n residuals, and the regression coefficients of the regression equation can be updated through such learning. After learning is completed, error factors can be input into the regression equation to output a y_trained value, and at this time, the y_trained value can correspond to the first SOH deviation value. When the first SOH deviation value is applied to each first SOH value, a second SOH value with SOH correction completed can be calculated.

[0103] Figure 10 illustrates the difference between the variation of SOH according to the error factor before correction and the variation of SOH according to the error factor after correction according to some embodiments.

[0104] Referring to FIG. 10, a first graph (1010) and a second graph (1020) can be illustrated illustrating the difference between the variation of SOH according to the error factor before correction and the variation of SOH according to the error factor after correction.

[0105] The horizontal axis of the first graph (1010) may be the charging start SOC, and the vertical axis may be the first SOH value before correction. The horizontal axis of the second graph (1020) may be the charging start SOC, and the vertical axis may be the second SOH value after correction. As in the first graph (1010), if the value of the charging start SOC fluctuates, the first SOH value also fluctuates, so the charging start SOC may cause an error in the first SOH value. On the other hand, as in the second graph (1020), even if the value of the charging start SOC fluctuates, the second SOH value hardly fluctuates and is constant, so the error caused by the charging start SOC can be mostly eliminated in the second SOH value.

[0106] Figure 11 illustrates the difference between the SOH distribution before correction and the SOH distribution after correction according to some embodiments.

[0107] Referring to FIG. 11, a first graph (1110) and a second graph (1120) can be illustrated illustrating the difference between the SOH distribution before correction and the SOH distribution after correction.

[0108] The horizontal axis of the first graph (1110) may be the driving distance of a battery-equipped vehicle, and the vertical axis may be the first SOH value before correction. The horizontal axis of the second graph (1120) may be the driving distance of a battery-equipped vehicle, and the vertical axis may be the second SOH value after correction. As illustrated, a tendency for SOH to decrease as the driving distance increases may be shown in the second graph (1120). In addition, it may be confirmed that in all cases of dod25, dod45, and dod65, the SOH distribution of the second graph (1120) is lower than the SOH distribution of the first graph (1120), which may mean that the SOH error factor has been resolved, enabling more accurate SOH estimation.

[0109] FIG. 12 illustrates steps of a battery management method according to some embodiments.

[0110] Referring to FIG. 12, the battery management method (1200) may include steps (1210) to (1240). However, the present invention is not limited thereto, and some steps may be omitted or other general steps may be added, and the steps of the battery management method (1200) may be executed in a different order than the illustrated order.

[0111] The battery management method (1200) may be composed of steps that are processed in a time-series manner in the battery management device (130). Therefore, even if the content is omitted below, the content described above for the battery management device (130) may be equally applied to the battery management method (1200).

[0112] Steps (1210) to (1240) of the battery management method (1200) can be performed by the interface (131) and controller (132) of the battery management device (130).

[0113] In step (1210), the battery management device (130) may perform a step of acquiring battery data of the battery.

[0114] In step (1220), the battery management device (130) may perform a step of calculating a first SOH value of the battery based on battery data.

[0115] In step (1230), the battery management device (130) may perform a step of calculating a first SOH correction value for the first SOH value using a regression equation that calculates an SOH correction value based on error factors that cause an error in the first SOH value.

[0116] In step (1240), the battery management device (130) may perform a step of calculating a second SOH value by applying a first SOH correction value to the first SOH value.

[0117] According to an embodiment, the battery management method (1200) may be implemented in the form of a computer program stored on a computer-readable storage medium. That is, the computer program may include instructions for implementing the battery management method (1200), and the program instructions may be stored on the computer-readable storage medium. The computer program may include a mobile application.

[0118] According to an embodiment, the computer-readable storage medium may include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute computer program instructions such as ROMs, RAMs, flash memories, and the like. The computer program instructions may include machine language codes generated by a compiler and high-level language codes that can be executed by a computer using an interpreter, etc.

[0119] The terms "include," "comprise," or "have" used herein, unless otherwise specifically stated, imply that the corresponding component may be included, and therefore should be interpreted to include other components rather than to exclude other components. All terms, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiments disclosed in this document pertain, unless otherwise defined. Commonly used terms, such as terms defined in dictionaries, should be interpreted to be consistent with their contextual meaning in the relevant art, and shall not be interpreted in an idealized or overly formal sense, unless explicitly defined in this document.

[0120] The above description is merely an illustrative description of the technical idea disclosed in this document, and those skilled in the art to which the embodiments disclosed in this document pertain may make various modifications and variations without departing from the essential characteristics of the embodiments disclosed in this document. Therefore, the embodiments disclosed in this document are not intended to limit the technical idea of ​​the embodiments disclosed in this document, but to explain it, and the scope of the technical idea disclosed in this document is not limited by these embodiments. The protection scope of the technical idea disclosed in this document should be interpreted by the claims below, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of rights of this document.

[0121] [Explanation of symbols]

[0122] 100: Battery management system 110: Power usage device

[0123] 120: Battery 130: Battery management unit

[0124] 131: Interface 132: Controller

Claims

1. An interface configured to obtain battery data of a battery; and Calculating a first SOH value of the battery based on the above battery data, Calculate a first SOH correction value for the first SOH value using a regression equation that calculates an SOH correction value based on error factors that cause errors in the first SOH value, A battery management device comprising a controller configured to calculate a second SOH value by applying the first SOH correction value to the first SOH value.

2. In paragraph 1, The above battery data includes charging data that is accumulated whenever the charging of the battery satisfies the validity condition, A battery management device, wherein the controller is configured to calculate the second SOH value based on the charging data.

3. In paragraph 2, The above charging data includes charging open circuit voltage (OCV) data and charging current data for each valid charge that satisfies the above validity condition within the period of interest, A battery management device, wherein the controller is configured to calculate the first SOH value for each of the valid charges based on SOC data corresponding to the charge OCV data and the charge current data.

4. In paragraph 3, A battery management device, wherein the regression coefficients of the regression equation for the above error factors are learned to calculate the first SOH correction value based on the first SOH values ​​for the effective charges.

5. In paragraph 4, The first SOH deviation value of each of the first SOH values ​​compared to the representative value of the first SOH values ​​is set as the residual of the regression equation, A battery management device, wherein the above regression coefficients are learned to minimize the residuals of the above regression equation.

6. In paragraph 5, A battery management device, wherein the above error factors include at least one of a charge start SOC according to the SOC data, a charge end SOC according to the SOC data, a charge start temperature, and a charge end temperature.

7. In paragraph 6, A battery management device, wherein the above regression equation is constructed based on a multivariate polynomial including terms composed of combinations of the above error factors.

8. In paragraph 7, A battery management device, wherein the controller is configured to calculate first SOH correction values ​​for the effective charges by applying the error factors for each of the effective charges to the learned regression coefficients of the regression equation.

9. In paragraph 2, A battery management device, wherein the above validity conditions include a first condition regarding whether the charge amount exceeds a threshold amount and a second condition regarding whether the charge pause time exceeds a threshold time.

10. In paragraph 5, A battery management device in which learning of the above regression coefficients is performed based on a lightweight gradient boosting machine (light GBM) or linear regression.

11. Step of acquiring battery data of the battery; A step of calculating a first SOH value of the battery based on the battery data; A step of calculating a first SOH correction value for the first SOH value using a regression equation that calculates an SOH correction value based on error factors that cause an error in the first SOH value; and A battery management method comprising a step of calculating a second SOH value by applying the first SOH correction value to the first SOH value.

12. In paragraph 11, The above battery data includes charging data that is accumulated whenever the charging of the battery satisfies the validity condition, A battery management method, wherein the second SOH value is calculated based on the charging data.

13. In paragraph 12, The above charging data includes charging open circuit voltage (OCV) data and charging current data for each valid charge that satisfies the above validity condition within the period of interest, The step of calculating the above first SOH value is: A battery management method, comprising the step of calculating the first SOH value for each of the effective charges based on the SOC data corresponding to the charge OCV data and the charge current data.

14. In paragraph 13, A battery management method, wherein the regression coefficients of the regression equation for the above error factors are learned to calculate the first SOH correction value based on the first SOH values ​​for the effective charges.

15. In paragraph 14, The first SOH deviation value of each of the first SOH values ​​compared to the representative value of the first SOH values ​​is set as the residual of the regression equation, A battery management method, wherein the above regression coefficients are learned to minimize the residuals of the above regression equation.

16. In paragraph 15, A battery management method, wherein the above error factors include at least one of a charge start SOC according to the SOC data, a charge end SOC according to the SOC data, a charge start temperature, and a charge end temperature.

17. In paragraph 16, A battery management method, wherein the above regression equation is constructed based on a multivariate polynomial including terms composed of combinations of the above error factors.

18. In paragraph 17, The step of calculating the above first SOH correction value is: A battery management method comprising a step of calculating first SOH correction values ​​for the effective charges by applying the error factors for each of the effective charges to the learned regression coefficients of the regression equation.

19. In paragraph 12, A battery management method, wherein the above validity conditions include a first condition regarding whether the charge amount exceeds a threshold amount and a second condition regarding whether the charge pause time exceeds a threshold time.

20. In paragraph 15, A battery management method, wherein learning of the above regression coefficients is performed based on a lightweight gradient boosting machine (light GBM) or linear regression.

Images