Apparatus and method for generating diagnostic model and apparatus for diagnosing battery

Abstract

An apparatus for generating a diagnostic model according to an embodiment of the present invention comprises: a battery information acquisition unit configured to acquire an indicator value of each of a plurality of batteries for a plurality of preset indicators in each of a plurality of cycles, and acquire a target value of the plurality of batteries in a target cycle; and a model training unit configured to calculate a correlation coefficient between the target value of the plurality of batteries and each of the plurality of indicators in each of the plurality of cycles, determine, on the basis of the plurality of correlation coefficients respectively calculated for the plurality of indicators, a target indicator among the plurality of indicators, and train a diagnostic model by using the target value of the plurality of batteries and the indicator value of each of the plurality of batteries corresponding to the target indicator.

Description

Diagnostic model generation device and method and battery diagnostic device

[0001] This application is a priority claim application for Korean Patent Application No. 10-2024-0143309 filed on October 18, 2024, and all contents disclosed in the specification and drawings of said application are incorporated into this application by reference.

[0002] The present invention relates to a diagnostic model generation device and method for generating a diagnostic model, and a battery diagnostic device for diagnosing the condition of a battery using the same.

[0003] Recently, as the demand for portable electronic products such as laptops, video cameras, and mobile phones has increased rapidly, and the development of electric vehicles, energy storage batteries, robots, and satellites has accelerated, research on high-performance batteries capable of repeated charging and discharging is actively underway.

[0004] Currently commercialized batteries include nickel-cadmium, nickel-hydrogen, nickel-zinc, and lithium batteries. Among these, lithium batteries are gaining attention for their advantages, such as the ability to freely charge and discharge with almost no memory effect compared to nickel-based batteries, a very low self-discharge rate, and high energy density.

[0005] While extensive research is being conducted on these batteries in terms of increasing capacity and density, improving lifespan and safety is also crucial. To enhance battery safety, technology capable of accurately diagnosing the battery's condition is required.

[0006] The present invention aims to provide an apparatus and method for generating a diagnostic model used to estimate the state of a battery. Additionally, the present invention aims to provide a battery diagnostic apparatus that estimates the state of a battery using a learned diagnostic model.

[0007] Other objects and advantages of the present invention may be understood from the following description and will become more clearly apparent from the embodiments of the present invention. Furthermore, it will be readily apparent that the objects and advantages of the present invention can be realized by the means and combinations thereof set forth in the claims.

[0008] A diagnostic model generating device according to one aspect of the present invention may include: a battery information acquisition unit configured to acquire indicator values ​​for each of a plurality of batteries for a plurality of pre-set indicators in each of a plurality of cycles and to acquire target values ​​for the plurality of batteries in a target cycle; and a model learning unit configured to calculate correlation coefficients between the target values ​​of the plurality of batteries and each of the plurality of indicators for each of the plurality of cycles, determine a target indicator among the plurality of indicators based on the correlation coefficients calculated for each of the plurality of indicators, and to train a diagnostic model using the target values ​​of the plurality of batteries and the indicator values ​​of the plurality of batteries corresponding to the target indicators.

[0009] The above model learning unit may be configured to calculate the correlation coefficient for each of the plurality of indicators by considering the linear relationship between the target value of the plurality of batteries and the indicator value of each of the plurality of indicators.

[0010] The above model learning unit may be configured to determine a reference coefficient among a plurality of correlation coefficients corresponding to each of the plurality of indicators, and to determine the target indicator by comparing the determined plurality of reference coefficients with each other.

[0011] The above model learning unit may be configured to compare the magnitudes of the plurality of reference coefficients and determine at least one target indicator based on the comparison result.

[0012] The above model learning unit may be configured to determine the maximum reference coefficient among the plurality of reference coefficients and to determine the indicator corresponding to the maximum reference coefficient as the target indicator.

[0013] The above model learning unit may be configured to determine a plurality of reference coefficients that are greater than or equal to a preset reference value among the plurality of reference coefficients, and to determine an indicator corresponding to the determined plurality of reference coefficients as the target indicator.

[0014] The above model learning unit may be configured to determine the maximum correlation coefficient among the above plurality of correlation coefficients as the reference coefficient.

[0015] The above target cycle can be set as a cycle after the above plurality of cycles.

[0016] A battery diagnostic device according to another aspect of the present invention may include: a memory configured to store a diagnostic model learned by a diagnostic model generating device according to one aspect of the present invention; and a processor configured to obtain a diagnostic indicator value for the target indicator of the diagnostic battery and to diagnose the state of the target cycle of the diagnostic battery by inputting the diagnostic indicator value into the learned diagnostic model.

[0017] The above processor may be configured to estimate the target value of the target cycle of the diagnostic battery.

[0018] A battery pack according to another aspect of the present invention may include a battery diagnostic device.

[0019] An automobile according to another aspect of the present invention may include a battery diagnostic device.

[0020] A method for generating a diagnostic model according to another aspect of the present invention may include: a battery information acquisition step of acquiring indicator values ​​for each of a plurality of batteries for a plurality of pre-set indicators in each of a plurality of cycles and acquiring target values ​​for the plurality of batteries in a target cycle; a correlation coefficient calculation step of calculating correlation coefficients between the target values ​​of the plurality of batteries and each of the plurality of indicators for each of the plurality of cycles; a target indicator determination step of determining a target indicator among a plurality of indicators based on the correlation coefficients calculated for each of the plurality of indicators; and a learning step of training a diagnostic model using the target values ​​of the plurality of batteries and the indicator values ​​of the plurality of batteries corresponding to the target indicators.

[0021] A computer-readable recording medium according to another aspect of the present invention may store a computer program for executing a diagnostic model generation method comprising: a battery information acquisition step of acquiring indicator values ​​for each of a plurality of batteries for a plurality of preset indicators at each of a plurality of cycles and acquiring target values ​​for the plurality of batteries at a target cycle; a correlation coefficient calculation step of calculating a correlation coefficient between the target values ​​of the plurality of batteries and each of the plurality of indicators for each of the plurality of cycles; a target indicator determination step of determining a target indicator among a plurality of indicators based on a plurality of correlation coefficients calculated for each of the plurality of indicators; and a learning step of training a diagnostic model using the target values ​​of the plurality of batteries and the indicator values ​​of the plurality of batteries corresponding to the target indicators.

[0022] According to one aspect of the present invention, the diagnostic model generating device has the advantage of being able to generate a learned diagnostic model capable of more accurately diagnosing the state of a battery in a target cycle.

[0023] In addition, according to one aspect of the present invention, since the state of the target cycle of the diagnostic battery is diagnosed according to a learned diagnostic model, the future state of the diagnostic battery can be predicted more accurately.

[0024] The effects of the present invention are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art from the description in the claims.

[0025] The following drawings attached to this specification serve to further enhance understanding of the technical concept of the invention in conjunction with the detailed description of the invention set forth below; therefore, the invention should not be interpreted as being limited only to the matters described in such drawings.

[0026] FIG. 1 is a schematic diagram illustrating a diagnostic model generation device according to one embodiment of the present invention.

[0027] FIG. 2 is a schematic diagram illustrating the correlation coefficient according to one embodiment of the present invention.

[0028] Figures 3 to 5 schematically illustrate the cycle-by-cycle correlation coefficients for the first to third indicators.

[0029] FIG. 6 is a schematic diagram illustrating a battery diagnostic device according to another embodiment of the present invention.

[0030] FIG. 7 is a schematic diagram illustrating a battery pack according to another embodiment of the present invention.

[0031] FIG. 8 is a schematic drawing illustrating an automobile according to another embodiment of the present invention.

[0032] FIG. 9 is a schematic diagram illustrating a battery diagnostic method according to another embodiment of the present invention.

[0033] Terms and words used in this specification and claims should not be interpreted as being limited to their ordinary or dictionary meanings, but should be interpreted in a meaning and concept consistent with the technical spirit of the invention, based on the principle that the inventor can appropriately define the concept of the terms to best describe his invention.

[0034] Therefore, the embodiments described in this specification and the configurations illustrated in the drawings are merely the most preferred embodiments of the present invention and do not represent all of the technical ideas of the present invention; thus, it should be understood that various equivalents and modifications that can replace them may exist at the time of filing this application.

[0035] In addition, in describing the present invention, if it is determined that a detailed description of related known components or functions may obscure the essence of the invention, such detailed description is omitted.

[0036] Terms including ordinal numbers, such as first, second, etc., are used for the purpose of distinguishing one of the various components from the rest, and are not used to limit the components by such terms.

[0037] Throughout the specification, when a part is described as "including" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0038] Additionally, throughout the specification, when it is said that a part is "connected" to another part, this includes not only cases where they are "directly connected," but also cases where they are "indirectly connected" with other components in between.

[0039]

[0040] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings.

[0041] FIG. 1 is a schematic diagram illustrating a diagnostic model generating device (100) according to one embodiment of the present invention.

[0042] Referring to FIG. 1, the diagnostic model generation device (100) may include a battery information acquisition unit (110) and a model learning unit (120).

[0043] The battery information acquisition unit (110) may be configured to acquire indicator values ​​for each of a plurality of batteries for a plurality of preset indicators in each of a plurality of cycles.

[0044] Here, a battery refers to a single, independent cell that is physically separable and equipped with a negative terminal and a positive terminal. For example, a lithium-ion battery or a lithium-polymer battery may be considered a battery. Additionally, the battery may be of the cylindrical, prismatic, or pouch type. Furthermore, a battery may refer to a battery bank, battery module, or battery pack in which multiple cells are connected in series and / or parallel. For the sake of convenience of explanation, the term "battery" below is described as referring to a single, independent cell.

[0045] The battery information acquisition unit (110) can be connected via wired and / or wireless means to enable communication with the outside. Additionally, the battery information acquisition unit (110) can acquire battery information for a plurality of batteries from the outside.

[0046] The battery information acquisition unit (110) can acquire various indicator values ​​of multiple batteries in each cycle. For example, if the number of batteries is n and m indicators are pre-set, the battery information acquisition unit (110) can acquire n × m indicator values ​​in each cycle. Here, if the number of cycles is k, the battery information acquisition unit (110) can acquire n × m × k indicator values.

[0047] In addition, multiple indicators can be pre-set as items obtainable from the battery's charge / discharge data. For example, multiple indicators can be pre-set as follows.

[0048] (1) Charging start voltage: Voltage when charging starts

[0049] (2) Charging termination voltage: Voltage when charging is terminated

[0050] (3) Discharge start voltage: Voltage when the discharge starts

[0051] (4) Discharge termination voltage: Voltage when the discharge ends

[0052] (5) Charging termination resistance: Resistance when charging ends

[0053] (6) Discharge termination resistance: Resistance when the discharge ends

[0054] (7) Charge-discharge capacity ratio: ratio of the amount charged to the amount discharged

[0055] (8) Capacity: Amount of charge or discharge of the battery

[0056] (9) SOH (State of health): Battery degradation

[0057] (10) CC (Constant current) charging capacity: Amount of CC charging

[0058] (11) CC charging capacity ratio: The ratio of CC charging capacity to the total charging amount

[0059] (12) CC charging time: Time taken to charge CC

[0060] (13) CV (Constant voltage) charging capacity: Amount of CV charging

[0061] (14) CV charging capacity ratio: The ratio of CV charging capacity to the total charging amount

[0062] (15) CV charging time: Time taken to charge the CV

[0063] (16) Peak voltage or derivative capacitance: Peak voltage or derivative capacitance included in the derivative capacitance profile representing the correspondence between voltage (V) and derivative capacitance (dQdV).

[0064] Here, the differential capacitance profile may include one or more peaks. If the differential capacitance profile includes multiple peaks, the voltage or differential capacitance for each of the multiple peaks may be set as an indicator.

[0065] (17) Peak capacitance or derivative voltage: Peak capacitance or derivative voltage included in the derivative voltage profile representing the correspondence between capacitance (Q) and derivative voltage (dVdQ).

[0066] Here, the differential voltage profile may include one or more peaks. If the differential voltage profile includes multiple peaks, the capacitance or differential voltage for each of the multiple peaks may be set as an indicator.

[0067] (18) Change amount or rate of change for each of indicators (1) to (17): Change amount or rate of change of the value of indicators (1) to (17) relative to the reference value of indicators (1) to (17) (e.g., value of the initial cycle).

[0068] The battery information acquisition unit (110) may be configured to acquire target values ​​of multiple batteries of a target cycle.

[0069] Specifically, the target cycle can be set as a cycle after a plurality of cycles. For example, the battery information acquisition unit (110) can acquire battery information of a plurality of batteries for the first to the 100th cycles. And, the battery information can acquire target values ​​of a plurality of batteries for the 300th cycle corresponding to the target cycle.

[0070] Here, the target value can be set to a value that indicates the state of the battery in the target cycle. Specifically, the target value can be set to a value that directly or indirectly indicates the battery's lifespan or degradation level in the target cycle. For example, the target value can be set to the battery's SOH, SOHQ (State of health quality), SOHC (State of health capacity), SOHR (State of health resistance), or SOHP (State of health power) in the target cycle. Although limited embodiments regarding the target value have been described above, any value capable of indicating the battery's performance indicators in the target cycle can be applied as the target value.

[0071] As previously explained, the battery information acquisition unit (110) can be connected to communicate with the outside. Additionally, the battery information acquisition unit (110) can acquire target values ​​of multiple batteries of a target cycle from the outside.

[0072] The battery information acquisition unit (110) can be connected via wired and / or wireless means to communicate with the model learning unit (120). Additionally, the battery information acquisition unit (110) can transmit various acquired information to the model learning unit (120).

[0073] The model learning unit (120) can be configured to calculate the correlation coefficient between the target value of a plurality of batteries and each of a plurality of indicators for each of a plurality of cycles.

[0074] Specifically, the model learning unit (120) can calculate correlation coefficients for multiple batteries in each cycle. Here, the correlation coefficient is a value corresponding to each indicator and can be calculated as a value representing the correlation between the indicators of multiple batteries for one cycle.

[0075] FIG. 2 is a schematic diagram illustrating the correlation coefficient according to one embodiment of the present invention.

[0076] For example, in the embodiment of FIG. 2, it is assumed that the charging end voltages of the first to fourth batteries (B1, B2, B3, and B4) are measured for the first to fourth cycles. The model learning unit (120) can calculate a correlation coefficient C1 for the first cycle by considering the correlation between the voltages (V11, V12, V13, and V14) for the first cycle of the first to fourth batteries (B1, B2, B3, and B4) and the target values ​​of the target cycles of the first to fourth batteries (B1, B2, B3, and B4). The model learning unit (120) can calculate a correlation coefficient C2 for the second cycle by considering the correlation between the voltages (V21, V22, V23, and V24) for the second cycle of the first to fourth batteries (B1, B2, B3, and B4) and the target values ​​of the target cycles of the first to fourth batteries (B1, B2, B3, and B4). The model learning unit (120) can calculate a correlation coefficient C3 for the third cycle by considering the correlation between the voltages (V31, V32, V33, and V34) for the third cycle of the first to fourth batteries (B1, B2, B3, and B4) and the target values ​​of the target cycle of the first to fourth batteries (B1, B2, B3, and B4). The model learning unit (120) can calculate a correlation coefficient C4 for the fourth cycle by considering the correlation between the voltages (V41, V42, V43, and V44) for the fourth cycle of the first to fourth batteries (B1, B2, B3, and B4) and the target values ​​of the target cycle of the first to fourth batteries (B1, B2, B3, and B4). That is, the model learning unit (120) can calculate a correlation coefficient for each indicator for each cycle.

[0077] The model learning unit (120) can be configured to determine a target indicator among multiple indicators based on multiple correlation coefficients calculated for each of multiple indicators.

[0078] For example, the target indicator can be determined as one of a plurality of indicators. The model learning unit (120) can determine the target indicator that best reflects the state of the battery among the plurality of indicators by considering a plurality of correlation coefficients calculated for the plurality of indicators.

[0079] As another example, the target indicator may be determined as at least one of a plurality of indicators. The model learning unit (120) may determine one or more target indicators among the plurality of indicators that reflect the state of the battery at a certain level or higher by considering a plurality of correlation coefficients calculated for the plurality of indicators.

[0080] First, the model learning unit (120) may be configured to determine a reference coefficient among a plurality of correlation coefficients corresponding to each of a plurality of indicators. Here, the reference coefficient refers to a single coefficient value determined for each of the plurality of indicators. That is, any one of the plurality of correlation coefficients corresponding to a given indicator may be determined as the reference coefficient representing that indicator.

[0081] For example, assuming that there are k cycles, k correlation coefficients corresponding to the P index can be calculated. This is because a correlation coefficient corresponding to the P index is determined for each cycle. And, the model learning unit (120) can determine any one of the k correlation coefficients as a reference coefficient representing the P index.

[0082] Specifically, the model learning unit (120) may be configured to determine the maximum correlation coefficient among a plurality of correlation coefficients as the reference coefficient. For example, the model learning unit (120) may determine the maximum value among a plurality of correlation coefficients corresponding to a plurality of cycles as the reference coefficient.

[0083] FIGS. 3 to 5 schematically illustrate the correlation coefficients for the first to third indicators per cycle. Specifically, FIG. 3 schematically illustrates a plurality of correlation coefficients for the first indicator. FIG. 4 schematically illustrates a plurality of correlation coefficients for the second indicator. FIG. 5 schematically illustrates a plurality of correlation coefficients for the third indicator. In the embodiments of FIGS. 3 to 5, the number of cycles is the same.

[0084] For example, in the embodiment of FIG. 3, the model learning unit (120) may determine t1, which corresponds to the maximum value among a plurality of correlation coefficients for the first indicator, as the reference coefficient of the first indicator. In the embodiment of FIG. 4, the model learning unit (120) may determine t2, which corresponds to the maximum value among a plurality of correlation coefficients for the second indicator, as the reference coefficient of the second indicator. In the embodiment of FIG. 5, the model learning unit (120) may determine t3, which corresponds to the maximum value among a plurality of correlation coefficients for the third indicator, as the reference coefficient of the third indicator.

[0085] And, the model learning unit (120) can be configured to determine a target indicator by comparing a plurality of determined reference coefficients with each other.

[0086] Specifically, the model learning unit (120) can compare a plurality of determined reference coefficients with each other and determine a target indicator among a plurality of indicators based on the comparison result. For example, the model learning unit (120) can determine at least one indicator among a plurality of indicators as a target indicator.

[0087] In the embodiments of FIGS. 3 to 5, the reference coefficient of the first indicator is t1, the reference coefficient of the second indicator is t2, and the reference coefficient of the third indicator is t3. The model learning unit (120) compares t1, t2, and t3 with each other and can determine a target indicator among the first to third indicators based on the comparison result. For example, the model learning unit (120) can determine the second indicator, which has the largest corresponding reference coefficient, as the target indicator.

[0088] The model learning unit (120) can be configured to train a diagnostic model using the target values ​​of multiple batteries and the indicator values ​​of each of the multiple batteries corresponding to the target indicators.

[0089] Specifically, the target indicator determined by the model learning unit (120) is the indicator most closely related to the state of the target cycle of the plurality of batteries. Accordingly, the model learning unit (120) can train a diagnostic model using the indicator values ​​of the plurality of batteries corresponding to the target indicator.

[0090] Here, the process of training the diagnostic model may adopt a conventional method of training a model. For example, the model training unit (120) may classify some of the multiple batteries as training batteries and the rest as verification batteries. The model training unit (120) may input the target value and indicator value of the training battery into the diagnostic model and compare the result value output from the diagnostic model with the corresponding target value. The model training unit (120) may train the diagnostic model by modifying various parameters or conditions of the diagnostic model based on the comparison result. Then, the model training unit (120) may input the target value and indicator value of the verification battery into the trained diagnostic model and obtain the verification result value output from the diagnostic model. Finally, the model training unit (120) may verify the trained diagnostic model by comparing the obtained verification result value with the target value. By repeating this process, the model training unit (120) may train the diagnostic model so that the verification accuracy is above a certain level.

[0091] A diagnostic model generating device (100) according to one embodiment of the present invention has the advantage of being able to generate a learned diagnostic model that can more accurately estimate the state of the target cycle of the battery.

[0092]

[0093] The battery information acquisition unit (110) and / or model learning unit (120) provided in the diagnostic model generation device (100) may optionally include a processor, an ASIC (application-specific integrated circuit), another chipset, a logic circuit, a register, a communication modem, a data processing device, etc., known in the art to execute various control logics performed in the present invention. Additionally, when the control logic is implemented in software, the battery information acquisition unit (110) and / or model learning unit (120) may be implemented as a set of program modules.

[0094] The storage unit (130) can store data or programs necessary for each component of the diagnostic model generation device (100) to perform operations and functions, or data generated during the process of performing operations and functions. The storage unit (130) is not limited in its type as long as it is a known information storage means capable of recording, erasing, updating, and reading data. As an example, the information storage means may include RAM, flash memory, ROM, EEPROM, registers, etc. Additionally, the storage unit (130) can store program codes in which processes executable by each component of the diagnostic model generation device (100) are defined.

[0095]

[0096] The model learning unit (120) can be configured to calculate the correlation coefficient of each of the multiple indicators by considering the linear relationship between the target value of the multiple batteries and the indicator value of each of the multiple indicators.

[0097] Specifically, the model learning unit (120) can calculate a correlation coefficient for a cycle by considering the linear correlation between the target value and the indicator value of a plurality of batteries for each cycle. For example, the closer the correlation coefficient calculated by the model learning unit (120) is to 1, the stronger the correlation between the target value and the indicator value, and the closer it is to 0, the weaker the correlation between the target value and the indicator value. That is, the model learning unit (120) can quantify the correlation between the target value and the indicator value by calculating a linear correlation coefficient.

[0098] Preferably, the model learning unit (120) can calculate the cycle-by-cycle correlation coefficient of each of the multiple indicators by calculating the Pearson correlation coefficient between the target value and the indicator value of the multiple batteries.

[0099] Here, the Pearson correlation coefficient can be calculated in the range of -1 to 1. Specifically, the Pearson correlation coefficient is calculated to be closer to 1 when there is a strong positive correlation between the two variables (a relationship where if one variable increases, the other variable also increases), closer to 0 when there is no correlation between the two variables, and closer to -1 when there is a strong negative correlation between the two variables (a relationship where if one variable increases, the other variable decreases). The model learning unit (120) can calculate the correlation coefficient corresponding to each cycle by taking the absolute value of the calculated Pearson correlation coefficient.

[0100] For example, in the embodiment of FIG. 2, it is assumed that the SOH of the target cycle of the first battery (B1) is SOH1, the SOH of the target cycle of the second battery (B2) is SOH2, the SOH of the target cycle of the third battery (B3) is SOH3, and the SOH of the target cycle of the fourth battery (B4) is SOH4. The model learning unit (120) can calculate the correlation coefficient of the first cycle as C1 by considering the correspondence relationship (SOH1-V11, SOH2-V12, SOH3-V13 and SOH4-V14) between the target values ​​(SOH1, SOH2, SOH3 and SOH4) of the target cycle of the first to fourth batteries (B1, B2, B3 and B4) and the indicator values ​​(V11, V12, V13 and V14) of the first cycle. And, the model learning unit (120) can calculate the correlation coefficient of the second cycle as C2 by considering the correspondence relationship (SOH1-V21, SOH2-V22, SOH3-V23 and SOH4-V24) between the target values ​​(SOH1, SOH2, SOH3 and SOH4) of the target cycles of the first to fourth batteries (B1, B2, B3 and B4) and the indicator values ​​(V21, V22, V23 and V24) of the second cycle. And, the model learning unit (120) can calculate the correlation coefficient of the third cycle as C3 by considering the correspondence relationship (SOH1-V31, SOH2-V32, SOH3-V33 and SOH4-V34) between the target values ​​(SOH1, SOH2, SOH3 and SOH4) of the target cycles of the first to fourth batteries (B1, B2, B3 and B4) and the indicator values ​​(V31, V32, V33 and V34) of the third cycle. And, the model learning unit (120) can calculate the correlation coefficient of the fourth cycle as C4 by considering the correspondence relationship (SOH1-V41, SOH2-V42, SOH3-V43 and SOH4-V44) between the target values ​​(SOH1, SOH2, SOH3 and SOH4) of the target cycles of the first to fourth batteries (B1, B2, B3 and B4) and the indicator values ​​(V41, V42, V43 and V44) of the fourth cycle.Here, C1, C2, C3, and C4 are the absolute values ​​of the Pearson correlation coefficients for the corresponding cycles.

[0101] The diagnostic model generating device (100) determines a target indicator among multiple indicators by considering only the degree of correlation between the target value and the indicator value, without considering the directionality (positive correlation or negative correlation) of the correlation between the target value and the indicator value, so that an indicator with a high degree of correlation with the target value can be determined as the target indicator. In addition, the diagnostic model generating device (100) can generate a learned diagnostic model that can more accurately diagnose the state of the battery of the target cycle by training the diagnostic model using the target indicator.

[0102]

[0103] Below, various embodiments are described in which the model learning unit (120) determines a target indicator among multiple indicators.

[0104] Specifically, the model learning unit (120) may be configured to compare the magnitudes of multiple reference coefficients and determine at least one target indicator based on the comparison result. That is, the model learning unit (120) can determine a target indicator among multiple indicators by comparing the values ​​of reference coefficients representing each of the multiple indicators.

[0105] In one embodiment, the model learning unit (120) may be configured to determine the maximum reference coefficient among a plurality of reference coefficients and to determine the indicator corresponding to the maximum reference coefficient as the target indicator.

[0106] A reference coefficient can represent the degree of correlation between the corresponding indicator and the target value of the battery. Accordingly, the model learning unit (120) can determine the indicator with the highest correlation to the target value, that is, the indicator with the largest reference coefficient, as the target indicator. For example, in FIGS. 3 to 5, the reference coefficient of the first indicator is t1, the reference coefficient of the second indicator is t2, and the reference coefficient of the third indicator is t3. The model learning unit (120) can determine the second indicator, which has the largest reference coefficient among the first to third indicators, as the target indicator.

[0107] In another embodiment, the model learning unit (120) may be configured to determine a reference coefficient among a plurality of reference coefficients that is greater than or equal to a preset reference value, and to determine an indicator corresponding to the determined reference coefficient as a target indicator. Here, the reference value may be preset as a value that serves as a criterion for determining the corresponding indicator as a target indicator.

[0108] For example, the reference value may be set experimentally and / or theoretically in advance. In the embodiments of FIGS. 3 to 5, it is assumed that the reference value is set to a value greater than t3 and less than t1. In this case, the model learning unit (120) may determine the first indicator and the second indicator, in which the reference coefficient is greater than or equal to the reference value, as target indicators.

[0109] As another example, the model learning unit (120) may obtain a coefficient range including multiple reference coefficients and set a value corresponding to a preset ratio (e.g., 95%) in the coefficient range as a reference value. For example, assume that the preset ratio is α, the maximum value among the multiple reference coefficients is Cmax, and the minimum value is Cmin. The model learning unit (120) may set the reference value by calculating the formula "Cmin+{(Cmax-Cmin)×α}" or "Cmax-{(Cmax-Cmin)×(1-α)}". Then, the model learning unit (120) may determine an indicator among multiple indicators in which the value of the reference coefficient is greater than or equal to the reference value as a target indicator.

[0110] As another example, the model learning unit (120) can determine a preset number of reference coefficients in order of largest value among a plurality of reference coefficients, and set the minimum reference coefficient among the determined reference coefficients as the reference value. For example, in the embodiments of FIGS. 3 to 5, it is assumed that the preset number is 2. The model learning unit (120) can determine 2 reference coefficients (t1 and t2) among a plurality of reference coefficients (t1, t2, t3). And, the model learning unit (120) can set the minimum reference coefficient (t1) among the determined reference coefficients (t1 and t2) as the reference value.

[0111] A diagnostic model generating device (100) according to one embodiment of the present invention has the advantage of being able to determine a target indicator that has a high correlation with a target value among a plurality of indicators in various ways.

[0112]

[0113] In one embodiment, the model learning unit (120) can determine the final target indicator by considering a combination of target indicators determined in a plurality of ways.

[0114] For example, the model learning unit (120) can determine only the overlapping indicators among the target indicators determined according to multiple methods as the final target indicators.

[0115] As another example, the model learning unit (120) can determine all target indicators determined according to multiple methods as the final target indicators.

[0116] The diagnostic model generating device (100) has the advantage of being able to train a diagnostic model using a final target indicator that is more related to the target value of the battery's target cycle by synthesizing target indicators determined according to various methods to determine a final target indicator.

[0117]

[0118] FIG. 6 is a schematic diagram illustrating a battery diagnostic device (200) according to another embodiment of the present invention.

[0119] Referring to FIG. 6, the battery diagnostic device (200) may include a memory (210) and a processor (220).

[0120] The memory (210) may be configured to store a diagnostic model learned by the diagnostic model generation device (100). And, the processor (220) may be configured to access the memory (210) and use the learned diagnostic model.

[0121] The processor (220) can be configured to obtain a diagnostic indicator value for a target indicator of the diagnostic battery.

[0122] For example, the processor (220) may be connected via wired and / or wireless means to communicate with the outside. And, the processor (220) may obtain a diagnostic indicator value for a target indicator of a diagnostic battery from the outside.

[0123] The diagnostic battery is a battery of the same type as the plurality of batteries, but is an independent battery that is not included in the plurality of batteries. That is, the plurality of batteries are reference batteries used to learn the diagnostic model by the diagnostic model generation device (100), and the diagnostic battery is a battery that is the subject of state diagnosis.

[0124] Preferably, the diagnostic indicator value obtained by the processor (220) may be a value for a target indicator obtained in a plurality of cycles. For example, the processor (220) may obtain a diagnostic indicator value of a target indicator for the first to 100 cycles of the diagnostic battery.

[0125] The processor (220) can be configured to diagnose the state of the target cycle of the diagnostic battery by inputting the diagnostic indicator value into a learned diagnostic model.

[0126] Specifically, the diagnostic model is trained to output a target value of the target cycle based on the input data when it receives an indicator value of the target indicator. Accordingly, the processor (220) inputs the diagnostic indicator value into the trained diagnostic model and can estimate the target value of the target cycle of the diagnostic battery based on the result output from the trained diagnostic model.

[0127] Meanwhile, the diagnostic indicator value obtained by the processor (220) is a value corresponding to a plurality of cycles, and the plurality of cycles are cycles prior to the target cycle. For example, if the plurality of cycles are the first to the 100th cycles, the target cycle is a cycle of the 101st cycle or higher. That is, the processor (220) can diagnose the state of the battery in a future cycle from the diagnostic indicator value for the target indicator of a past cycle.

[0128] For example, assume that the target value is the SOH of the target cycle, multiple diagnostic indicator values ​​for the first to the 100th cycles have been obtained, and the target cycle is the 300th cycle. The processor (220) inputs the multiple diagnostic indicator values ​​into a learned diagnostic model and can estimate the SOH of the diagnostic battery for the 300th cycle based on the input result.

[0129] A battery diagnostic device (200) according to one embodiment of the present invention can estimate the future state of a battery by using a learned diagnostic model. According to the battery diagnostic device (200), since the future state of the battery is predicted, appropriate charge and discharge control for the battery can be performed to increase the lifespan of the battery. As a result, the expected lifespan of the battery can be significantly increased.

[0130]

[0131] The processor (220) provided in the battery diagnostic device (200) may optionally include an application-specific integrated circuit (ASIC), other chipsets, logic circuits, registers, communication modems, data processing devices, etc., known in the art, to execute various control logics performed in the present invention. Additionally, when the control logic is implemented in software, the processor (220) may be implemented as a set of program modules.

[0132] The memory (210) can store data or programs necessary for each component of the battery diagnostic device (200) to perform operations and functions, or data generated during the process of performing operations and functions. The memory (210) is not limited in its type as long as it is a known information storage means known to be able to record, erase, update, and read data. As an example, the information storage means may include RAM, flash memory, ROM, EEPROM, registers, etc. Additionally, the memory (210) can store program codes that define processes executable by each component of the battery diagnostic device (200).

[0133]

[0134] The battery diagnostic device (200) according to the present invention may be applied to a Battery Management System (BMS). That is, the BMS according to the present invention may include the battery diagnostic device (200) described above. In this configuration, at least some of the components of the battery diagnostic device (200) may be implemented by supplementing or adding the functions of the components included in a conventional BMS. For example, the memory (210) and processor (220) of the battery diagnostic device (200) may be implemented as components of the BMS.

[0135] In one embodiment, the battery diagnostic device (200) can control the charging and / or discharging of the battery based on the diagnostic results of the battery. That is, the processor (220) can control the charging and / or discharging of the battery by utilizing the diagnostic results so that the charging and / or discharging of the battery can proceed with optimized charging and / or discharging.

[0136] In another embodiment, the battery diagnostic device (200) may change various state conditions set for the battery based on the battery diagnostic results to prevent further degradation of the battery. For example, the processor (220) may set at least one of the battery's upper charge limit SOC, upper charge limit voltage, upper charge limit C-rate, lower discharge limit SOC, lower discharge limit voltage, upper discharge limit C-rate, and upper limit temperature. For convenience of explanation, the state conditions that can be set by the battery diagnostic device (200) have been listed above, but it should be noted that any state condition that can be set using the battery diagnostic results to delay battery degradation may be applied without limitation.

[0137]

[0138] In addition, the battery diagnostic device (200) according to the present invention may be provided in a battery pack. That is, the battery pack according to the present invention may include the battery diagnostic device (200) described above and one or more battery cells. In addition, the battery pack may further include electrical components (relays, fuses, etc.) and a case, etc.

[0139] FIG. 7 is a schematic diagram illustrating a battery pack (10) according to another embodiment of the present invention.

[0140] The positive terminal of the battery (11) may be connected to the positive terminal (P+) of the battery pack (10), and the negative terminal of the battery (11) may be connected to the negative terminal (P-) of the battery pack (10). In the embodiment of FIG. 7, the battery (11) is shown as a single battery cell, but the battery (11) may include a plurality of battery cells connected in series and / or parallel.

[0141] The measuring unit (12) is electrically connected to the battery (11) and can measure the voltage of the battery (11).

[0142] Additionally, the measuring unit (12) may be electrically connected to a current measuring unit (A). For example, the current measuring unit (A) may be an ammeter or a shunt resistor capable of measuring the charging current and discharging current of the battery (11). The measuring unit (12) can calculate the charge amount of the battery (11) by measuring the charging current of the battery (11) through the current measuring unit (A). Furthermore, the measuring unit (12) can calculate the discharge amount of the battery (11) by measuring the discharge current of the battery (11) through the current measuring unit (A).

[0143] The processor (220) can be connected to communicate with the measurement unit (12). The processor (220) can obtain a diagnostic indicator value of the battery (11) for a target indicator from the measurement unit (12). The processor (220) can estimate the target value of the battery (11) in a target cycle using the diagnostic indicator value of the battery (11) and a learned diagnostic model.

[0144] An external device may be connected to the positive terminal (P+) and the negative terminal (P-) of the battery pack (10). For example, the external device may be a charging device or a load. Also, the positive terminal of the battery (11), the positive terminal (P+) of the battery pack (10), the external device, the negative terminal (P-) of the battery pack (10), and the negative terminal of the battery (11) may be electrically connected.

[0145]

[0146] FIG. 8 is a schematic drawing illustrating a vehicle (800) according to another embodiment of the present invention.

[0147] Referring to FIG. 8, a battery pack (10) according to an embodiment of the present invention may be included in a vehicle (800), such as an electric vehicle (EV) or a hybrid vehicle (HV). The battery pack (10) can drive the vehicle (800) by supplying power to a motor through an inverter provided in the vehicle (800). Here, the battery pack (10) may include a battery diagnostic device (200). That is, the vehicle (800) may include a battery diagnostic device (200). In this case, the battery diagnostic device (200) may be an on-board device included in the vehicle (800).

[0148]

[0149] FIG. 9 is a schematic diagram illustrating a battery diagnostic method according to another embodiment of the present invention.

[0150] Referring to FIG. 9, the method for generating a diagnostic model may include a battery information acquisition step (S100), a correlation coefficient calculation step (S200), a target indicator determination step (S300), and a learning step (S400).

[0151] Preferably, each step of the diagnostic model generation method can be performed by a battery diagnostic device (200). For convenience of explanation, details that overlap with previously described content will be omitted or briefly explained below.

[0152] The battery information acquisition step (S100) is a step of acquiring indicator values ​​for each of a plurality of batteries for a plurality of preset indicators at each of a plurality of cycles, and acquiring target values ​​for a plurality of batteries of a target cycle, and can be performed by the battery information acquisition unit (110).

[0153] The battery information acquisition unit (110) can acquire battery information for a plurality of cycles and battery information for a target cycle for a plurality of batteries from an external source connected to enable communication. Here, the target cycle is a cycle after the plurality of cycles. For example, the plurality of cycles may be the first to the 100th cycles, and the target cycle may be the 300th cycle.

[0154] The correlation coefficient calculation step (S200) is a step of calculating the correlation coefficient between the target value of a plurality of batteries and each of a plurality of indicators for each of a plurality of cycles, and can be performed by the model learning unit (120).

[0155] The model learning unit (120) can calculate the correlation coefficient between the target value of a plurality of batteries and each indicator in each of a plurality of cycles. For example, the model learning unit (120) can calculate the Pearson correlation coefficient between the target value of a plurality of batteries and the indicator value of each indicator in each of a plurality of cycles.

[0156] The target indicator determination step (S300) is a step of determining a target indicator among multiple indicators based on multiple correlation coefficients calculated for each of multiple indicators, and can be performed by the model learning unit (120).

[0157] The model learning unit (120) can determine a reference coefficient among a plurality of correlation coefficients corresponding to each of a plurality of indicators. For example, the model learning unit (120) can determine the maximum correlation coefficient among a plurality of correlation coefficients for each indicator as the reference coefficient.

[0158] In addition, the model learning unit (120) can determine a target indicator by comparing a plurality of determined reference coefficients with each other. For example, the model learning unit (120) can determine at least one of the plurality of indicators as the target indicator.

[0159] The learning step (S400) is a step of learning a diagnostic model using the target values ​​of multiple batteries and the indicator values ​​of each of the multiple batteries corresponding to the target indicators, and can be performed by the model learning unit (120).

[0160] The model learning unit (120) can generate a learned diagnostic model that can be used to predict the state of the target cycle of the diagnostic battery by training the diagnostic model using the target values ​​of the multiple batteries and the indicator values ​​of each of the multiple batteries corresponding to the target indicators.

[0161]

[0162] The embodiments of the present invention described above are not limited to implementation through devices and methods, but may also be implemented through a program that realizes a function corresponding to the configuration of the embodiments of the present invention or a recording medium on which such a program is recorded. Such implementation can be easily achieved by a person skilled in the art to which the present invention pertains, based on the description of the embodiments described above.

[0163] Another embodiment of the present invention may provide a computer-readable recording medium having a program recorded thereon for executing the various embodiments described above on a computer.

[0164] A program may be implemented as hardware components, software components, and / or a combination of hardware and software components. A program may be executed by any system capable of executing computer-readable instructions.

[0165] Software may include computer programs, code, instructions, or a combination thereof, and may configure a processing unit to operate as desired or command the processing unit independently or collectively.

[0166] Software can be implemented as a computer program containing instructions stored on a computer-readable storage medium. Examples of computer-readable storage media include magnetic storage media (e.g., ROM (read-only memory), RAM (random-access memory), floppy disks, hard disks, etc.) and optical reading media (e.g., CD-ROMs, DVDs (Digital Versatile Discs)). Computer-readable storage media can be distributed across networked computer systems, allowing computer-readable code to be stored and executed in a distributed manner. The storage medium is readable by a computer, stored in memory, and can be executed by a processor.

[0167] Computer-readable recording media may be provided in the form of non-transitory recording media. Here, 'non-transitory storage media' simply means that it is a tangible device and does not contain a signal (e.g., electromagnetic waves), and the term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily. For example, 'non-transitory storage media' may include a buffer in which data is stored temporarily.

[0168] In addition, the program may be provided by being included in a computer program product. A computer program product may be traded between a seller and a buyer as a product.

[0169] A computer program product may include a software program or a computer-readable recording medium on which the software program is stored. For example, a computer program product may include a product in the form of a software program that is distributed electronically through a manufacturer of an electronic device or an electronic market (e.g., a downloadable application). For electronic distribution, at least a portion of the software program may be stored on a recording medium or temporarily created. In this case, the recording medium may be a server of the manufacturer of the electronic device, a server of the electronic market, or a recording medium of a relay server that temporarily stores the software program.

[0170] Although the present invention has been described above by limited embodiments and drawings, the present invention is not limited thereto, and it is obvious that various modifications and variations are possible within the scope of the technical spirit of the present invention and the equivalent scope of the claims described below by those skilled in the art to which the present invention belongs.

[0171] Furthermore, since the present invention described above allows for various substitutions, modifications, and changes within the scope of the technical concept of the present invention to those skilled in the art without departing from the technical spirit of the present invention, it is not limited by the aforementioned embodiments and attached drawings, but rather all or part of each embodiment may be selectively combined to allow for various modifications.

[0172] (Explanation of symbols)

[0173] 10: Battery pack

[0174] 11: Battery

[0175] 12: Measurement section

[0176] 100: Diagnostic model generation device

[0177] 110: Battery Information Acquisition Unit

[0178] 120: Model learning section

[0179] 130: Storage section

[0180] 200: Battery Diagnostic Device

[0181] 210: Memory

[0182] 220: Processor

[0183] 800: Car

Claims

1. A battery information acquisition unit configured to acquire indicator values ​​for each of a plurality of batteries for a plurality of preset indicators in each of a plurality of cycles, and to acquire target values ​​for the plurality of batteries in a target cycle; and A diagnostic model generation device comprising a model learning unit configured to calculate a correlation coefficient between a target value of a plurality of batteries and each of a plurality of indicators for each of the plurality of cycles, determine a target indicator among a plurality of indicators based on the correlation coefficient calculated for each of the plurality of indicators, and train a diagnostic model using the target value of a plurality of batteries and the indicator value of each of the plurality of batteries corresponding to the target indicator.

2. In Paragraph 1, The above model learning unit is, A diagnostic model generating device configured to calculate the correlation coefficient for each of the plurality of indicators by considering the linear relationship between the target value of the plurality of batteries and the indicator value of each of the plurality of indicators.

3. In Paragraph 1, The above model learning unit is, A diagnostic model generating device configured to determine a reference coefficient among a plurality of correlation coefficients corresponding to each of the plurality of indicators, and to determine the target indicator by comparing the determined plurality of reference coefficients with each other.

4. In Paragraph 3, The above model learning unit is, A diagnostic model generating device configured to compare the magnitudes of a plurality of reference coefficients and determine at least one target indicator based on the comparison result.

5. In Paragraph 4, The above model learning unit is, A diagnostic model generating device configured to determine the maximum reference coefficient among the plurality of reference coefficients and to determine the indicator corresponding to the maximum reference coefficient as the target indicator.

6. In Paragraph 4, The above model learning unit is, A diagnostic model generating device configured to determine a plurality of reference coefficients among the plurality of reference coefficients that are greater than or equal to a preset reference value, and to determine an indicator corresponding to the determined plurality of reference coefficients as the target indicator.

7. In Paragraph 3, The above model learning unit is, A diagnostic model generating device configured to determine the maximum correlation coefficient among the plurality of correlation coefficients as the reference coefficient.

8. In Paragraph 1, The above target cycle is, A diagnostic model generation device set to a cycle after the above plurality of cycles.

9. A memory configured to store a diagnostic model learned by a diagnostic model generating device according to any one of claims 1 to 8; and A battery diagnostic device comprising a processor configured to diagnose the state of the target cycle of the diagnostic battery by acquiring a diagnostic indicator value for the target indicator of the diagnostic battery and inputting the diagnostic indicator value into the learned diagnostic model.

10. In Paragraph 9, The above processor is, A battery diagnostic device configured to estimate the target value of the target cycle of the above-mentioned diagnostic battery.

11. A battery pack including a battery diagnostic device according to paragraph 9.

12. An automobile including a battery diagnostic device pursuant to paragraph 9.

13. A battery information acquisition step of acquiring indicator values ​​for each of a plurality of batteries for a plurality of preset indicators in each of a plurality of cycles, and acquiring target values ​​for the plurality of batteries in a target cycle; A correlation coefficient calculation step for calculating the correlation coefficient between the target value of the plurality of batteries and each of the plurality of indicators for each of the plurality of cycles; A target indicator determination step for determining a target indicator among a plurality of indicators based on a plurality of correlation coefficients calculated for each of the plurality of indicators; and A method for generating a diagnostic model comprising a learning step of training a diagnostic model using the target values ​​of the plurality of batteries and the indicator values ​​of each of the plurality of batteries corresponding to the target indicators.

14. A battery information acquisition step of acquiring indicator values ​​for each of a plurality of batteries for a plurality of preset indicators in each of a plurality of cycles, and acquiring target values ​​for the plurality of batteries in a target cycle; A correlation coefficient calculation step for calculating the correlation coefficient between the target value of the plurality of batteries and each of the plurality of indicators for each of the plurality of cycles; A target indicator determination step for determining a target indicator among a plurality of indicators based on a plurality of correlation coefficients calculated for each of the plurality of indicators; and A computer-readable recording medium storing a computer program for executing a diagnostic model generation method comprising a learning step of training a diagnostic model using target values ​​of the plurality of batteries and indicator values ​​of each of the plurality of batteries corresponding to the target indicators.

Images