SOH prediction device and method

Abstract

An SOH prediction device according to an embodiment of the present invention comprises: a data acquisition unit configured to acquire an observed histogram indicating a correspondence relationship among a voltage value, a current value, a temperature value, and a frequency value of a battery during an observation period; and a control unit configured to generate first to n-th histograms corresponding to first to n-th periods subsequent to the observation period, on the basis of the observed histogram and a reference temperature dataset, and predict a target SOH indicating an SOH of the battery at an end time point of the n-th period from the first to n-th histograms and an observed SOH indicating the SOH of the battery at an end time point of the observation period, on the basis of a preset degradation prediction model, wherein n is a natural number of 2 or more.

Description

SOH prediction device and method

[0001] The present invention relates to an SOH prediction device and method for predicting the SOH of a battery using a pre-learned degradation prediction model.

[0002] This application claims priority based on Korean Application No. 10-2025-0001430 filed on January 6, 2025, and all contents disclosed in the specification of said application are incorporated into this application.

[0003]

[0004] 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.

[0005] 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.

[0006] Battery State of Health (SOH) prediction technology is an essential requirement for ensuring battery stability and efficiency by evaluating battery performance and remaining lifespan. It is widely utilized in various application fields, such as electric vehicles, Energy Storage Systems (ESS), and portable electronic devices, and has established itself as one of the core functions of Battery Management Systems (BMS).

[0007] Conventional SOH prediction technologies have relied on physical models based on the physical and chemical properties of batteries. Physical models offer the advantage of providing high prediction accuracy by mathematically modeling electrochemical reactions, thermal behavior, and physical changes within the battery. However, physical models have the disadvantage of requiring complex calculations, resulting in long computation times, and are difficult to apply in environments requiring real-time processing, such as on-board or cloud-based BMSs. Furthermore, physical models are optimized for specific battery configurations or operating conditions, which limits their versatility across diverse environments.

[0008] To overcome these limitations, data-driven State of Health (SOH) prediction models are gaining attention. Data-driven models predict battery status by learning from battery measurement data (e.g., voltage, current, temperature, etc.), offering the advantages of faster computation speeds and simpler implementation compared to physical models. However, existing data-driven models have often yielded results with lower prediction accuracy than physical models, which has limited their practical applications.

[0009] Therefore, it is necessary to develop a new SOH prediction model that combines the high prediction accuracy of physical models with the fast computation speed of data-driven models.

[0010]

[0011] The present invention was devised to solve the above-mentioned problems and aims to provide an SOH prediction device and method capable of predicting a target SOH at a specific future point in time based on past actual observation data and a pre-trained degeneration prediction model.

[0012] 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.

[0013]

[0014] A SOH prediction device according to one aspect of the present invention may include: a data acquisition unit configured to acquire an actual observation histogram representing the correspondence between voltage, current, temperature, and frequency values ​​of a battery during an actual observation period; and a control unit configured to generate a first to nth histogram corresponding to a first to nth period following the actual observation period based on the actual observation histogram and a reference temperature dataset, and to predict a target SOH representing the SOH of the battery at the end of the nth period from the first to nth histograms and the actual observation SOH representing the SOH of the battery at the end of the actual observation period based on a preset degradation prediction model. n is a natural number greater than or equal to 2.

[0015] The above reference temperature dataset may include first to n reference temperature values ​​corresponding to the first to n periods.

[0016] The control unit may be configured to generate the first to nth histograms based on the comparison results between the representative temperature value of the actual observed histogram and each of the first to nth reference temperature values.

[0017] The control unit may be configured to generate the reference temperature dataset based on a plurality of temperature datasets collected from a plurality of different batteries over a past data collection period having a time length greater than or equal to the total time length of the first to nth periods.

[0018] The control unit above may be configured to determine a j-th temperature correction value, which is the difference between the representative temperature value and the j-th reference temperature among the first to n-th reference temperature values. j is a natural number less than or equal to n.

[0019] The above control unit may be configured to generate the j-th histogram among the first to n-th histograms by correcting each temperature value of the actual observation histogram based on the j-th temperature correction value.

[0020] The above degradation prediction model may be trained by multiple training data sets collected from multiple different batteries during multiple reference periods.

[0021] Each of the above multiple training data sets may include, as features, a reference histogram representing the correspondence between the voltage value, current value, temperature value, and frequency value of another battery during a reference period prior to the actual observation period, and the SOH of the other battery at the start of the reference period, and may include the SOH of the other battery at the end of the reference period as a label.

[0022] The time length of each of the above multiple reference periods may be equal to the total time length of the first to nth periods.

[0023] The time length of each of the first to n periods above may be the same as the time length of the actual observation period above.

[0024] The above degeneration prediction model can be composed of a CNN (Convolutional Neural Network) based model.

[0025] The control unit may be configured to predict the degradation rate of the battery over the first to nth periods based on the actual observed SOH and the target SOH.

[0026] A battery pack according to another aspect of the present invention may include the SOH prediction device.

[0027] A vehicle according to another aspect of the present invention may include the SOH prediction device.

[0028] A server according to another aspect of the present invention may include the SOH prediction device.

[0029] A method for predicting SOH according to another aspect of the present invention may include: obtaining an actual observation histogram representing the correspondence between voltage, current, temperature, and frequency values ​​of a battery during an actual observation period; generating a first to nth histogram corresponding to a first to nth period following the actual observation period based on the actual observation histogram and a reference temperature dataset; and predicting a target SOH representing the SOH of the battery at the end of the nth period from the actual observation SOH representing the SOH of the battery at the end of the actual observation period based on the first to nth histograms and the actual observation SOH representing the SOH of the battery at the end of the actual observation period, based on a preset degradation prediction model. n is a natural number greater than or equal to 2.

[0030]

[0031] According to one aspect of the present invention, based on past actual observation data and a pre-trained degeneration prediction model, the target SOH at a specific point in the future can be predicted with high reliability.

[0032] In addition, according to one aspect of the present invention, by using a 3D-CNN-based degradation prediction model, the SOH of a battery can be predicted more accurately and reliably.

[0033] 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.

[0034]

[0035] 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.

[0036] FIG. 1 is a schematic diagram illustrating an SOH prediction device according to one embodiment of the present invention.

[0037] Figure 2 is a schematic diagram illustrating one example of an actual observation histogram.

[0038] FIG. 3 is a drawing referenced to explain the actual observation period and the first to nth periods.

[0039] Figures 4 and 5 are schematic diagrams illustrating the structure of a degeneration prediction model.

[0040] FIG. 6 is a schematic diagram illustrating a battery pack according to another embodiment of the present invention.

[0041] FIG. 7 is a schematic drawing illustrating an automobile according to another embodiment of the present invention.

[0042] FIG. 8 is a schematic diagram illustrating a server according to another embodiment of the present invention.

[0043] FIG. 9 is a schematic diagram illustrating a method for predicting SOH according to another embodiment of the present invention.

[0044] FIG. 10 is a schematic diagram illustrating the sub-steps of step S920 of FIG. 9.

[0045]

[0046] 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.

[0047] 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.

[0048] 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.

[0049] 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.

[0050] 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.

[0051] 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.

[0052] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings.

[0053] FIG. 1 is a schematic diagram illustrating another SOH prediction device (100) in one embodiment of the present invention.

[0054] Referring to FIG. 1, the SOH prediction device (100) may include a data acquisition unit (110) and a control unit (120). The SOH prediction device (100) may further include a storage unit (130).

[0055] The data acquisition unit (110) may be configured to acquire an actual observation histogram showing the correspondence between the voltage value, current value, temperature value, and frequency value of the battery (see reference numeral 11 in FIG. 6) during the actual observation period.

[0056] 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.

[0057] Figure 2 is a schematic diagram illustrating one example of an actual observation histogram.

[0058] The actual observation histogram may be the result of VIT data points obtained during the actual observation period being classified into any one of the first to second VIT intervals. Here, each VIT data point may be a three-dimensional data value defined by voltage, current, and temperature values ​​indexed at the same measurement timing.

[0059] Each of the first to D VIT intervals is a combination of any one of the first to A voltage intervals (A is a natural number greater than or equal to 2), any one of the first to B current intervals (B is a natural number greater than or equal to 2), and any one of the first to C temperature intervals (C is a natural number greater than or equal to 2), and thus D may be equal to the product of A, B, and C.

[0060] The first to A voltage ranges may be a total voltage range allowed to the battery divided into A parts, the first to B current ranges may be a total current range allowed to the battery divided into B parts, and the first to C temperature ranges may be a total temperature range allowed to the battery divided into C parts. For example, when the total voltage range = 3.0 to 5.0 [V] and A=4, the first voltage range = 3.0 or more to less than 3.5 [V], the second voltage range = 3.5 or more to less than 4.0 [V], the third voltage range = 4.0 or more to less than 4.5 [V], and the fourth voltage range = 4.5 or more to less than 5.0 [V]. As another example, if the total current range is -20 to +20 [A] and B=4, the first current range may be -20 or more to less than -10 [A], the second current range may be -10 or more to less than 0 [A], the third current range may be 0 or more to less than +10 [A], and the fourth current range may be +10 or more to less than +20 [A]. As yet another example, if the total temperature range is -30 to +50 [℃] and C=5, the first temperature range may be -30 or more to less than -20 [℃], the second temperature range may be -20 or more to less than -5 [℃], the third temperature range may be -5 or more to less than +15 [℃], the fourth temperature range may be +15 or more to less than +35 [℃], and the fifth temperature range may be +35 or more to less than +50 [℃]. For reference, if A = 4, B = 4, and C = 5 as exemplified in Fig. 2, then D = 80.

[0061] As illustrated in FIG. 2, the actual observation histogram can be represented in the form of a 3D matrix. Specifically, referring to FIG. 2, the row number of the actual observation histogram may represent the voltage interval number, the column number may represent the current interval number, and the depth number may represent the temperature interval number.

[0062] Accordingly, each coordinate of the 3D matrix representing the actual observation histogram can correspond to any one of the first to second VIT intervals. Additionally, each value of the coordinates of the 3D matrix representing the actual observation histogram can represent the number (frequency value) of VIT data points classified into the VIT interval corresponding to that coordinate.

[0063] Referring to FIG. 2, where a is a natural number less than or equal to A, b is a natural number less than or equal to B, and c is a natural number less than or equal to C, the coordinates (a, b, c) can represent the {AB(c-1) + (a-1)B + b} VIT interval. For example, if the voltage, current, and temperature values ​​of a certain VIT data point are 3.2 [V], the current is -18 [A], and -25 [℃], then that VIT data point can be classified into the first VIT interval corresponding to the coordinates (1,1,1). For another example, if the voltage, current, and temperature values ​​of another VIT data point are 4.6 [V], the current is +17 [A], and +43 [℃], then that VIT data point can be classified into the 80th VIT interval corresponding to the coordinates (4,4,5).

[0064] Let us assume that a total of E VIT data points were obtained during the actual observation period, and that F of these VIT data points were classified into the {AB(c-1) + (a-1)B + b} VIT interval corresponding to the coordinates (a, b, c). Then, F or F / E can be assigned as the frequency value to the coordinates (a, b, c) of the actual observation histogram.

[0065] The actual observation period refers to the period during which the battery's voltage, current, and temperature are measured. The actual observation period can be pre-set. For example, the actual observation period can be pre-set to the period from June 1, 2024 to June 30, 2024.

[0066] In one embodiment, the data acquisition unit (110) can directly measure the voltage, current, and temperature of the battery. Specifically, the data acquisition unit (110) can measure the positive voltage and the negative voltage through a pair of voltage sensing lines connected to the positive and negative electrodes of the battery, respectively. Then, the data acquisition unit (110) can measure the voltage across the two ends of the battery based on the voltage difference between the measured positive voltage and the negative voltage. The data acquisition unit (110) can measure the current of the battery by being connected to the battery through a current measurement unit. For example, the current measurement unit may be a current sensor or a shunt resistor that is provided in the battery's charge / discharge path to measure the battery's current. Here, the battery's charge / discharge path may be a high-current path where a charging current is applied to the battery or a discharging current is output from the battery. The data acquisition unit (110) can measure the temperature of the battery using a temperature sensor. That is, the data acquisition unit (110) can directly measure the voltage, current, and temperature of the battery and generate an actual observation histogram based on the measurement results.

[0067] In another embodiment, the data acquisition unit (110) can receive an actual observation histogram from the outside. That is, the data acquisition unit (110) can receive an actual observation histogram from the outside by being connected via wired and / or wireless so as to be able to communicate with the outside. For example, the data acquisition unit (110) can receive an actual observation histogram from the outside using CAN (Controller Area Network) communication or CAN-FD (CAN with Flexible Data rate) communication. As another example, the data acquisition unit (110) can receive an actual observation histogram from the outside using Zigbee, Bluetooth, WIFI, or a mobile communication network. Of course, as long as it supports communication between the data acquisition unit (110) and the outside, the type of communication protocol is not particularly limited.

[0068] The data acquisition unit (110) may be connected via wired and / or wireless means to communicate with the control unit (120). The data acquisition unit (110) may transmit the acquired actual observation histogram to the control unit (120). The control unit (120) may receive the actual observation histogram from the data acquisition unit (110).

[0069] The control unit (120) may be configured to generate first to n histograms corresponding to first to n periods following the actual observation period, based on an actual observation histogram and a reference temperature data set. n is a natural number greater than or equal to 2.

[0070] The reference temperature data set may include first to n reference temperature values ​​corresponding to the first to n periods. The period from the start point of the first period to the end point of the n period may be referred to as the target future period.

[0071] Periods 1 through n refer to consecutive periods following the actual observation period. Each of periods 1 through n can be set as a time unit (e.g., day, week, month) for predicting the target SOH based on actual observation data. For example, if the time unit is 'day', the start date of period 1 may be the day following the end date of the actual observation period, and the start date of period 2 may be the day following the end date of period 1.

[0072] Specifically, the control unit (120) can generate the first to nth histograms by correcting the temperature values ​​of the actual observation histogram using a reference temperature data set. At this time, under the assumption that the usage pattern of the battery during the actual observation period remains the same in each of the first to nth periods, it is assumed that the pattern of voltage and current values ​​measured during the actual observation period remains the same in each of the first to nth periods.

[0073] The control unit (120) can correct the temperature values ​​of the actual observation histogram by referring to the reference temperature values ​​of the first to nth periods included in the reference temperature data set in order to reflect both the temperature change due to battery usage and the temperature change due to the external environment. For example, the control unit (120) can correct the actual observation histogram by reflecting the external temperature change over time. Specifically, the first to nth histograms can be generated that reflect high-temperature environments (summer) or low-temperature environments (winter) that have a significant impact on the operating temperature and performance of the battery. Through this, the first to nth histograms can be generated that reflect the temperature conditions corresponding to each period while maintaining the same voltage and current value patterns as the actual observation data.

[0074] The temperature correction process can be performed based on the first to nth reference temperature values ​​and the temperature values ​​included in the actual observation histogram. The control unit (120) can correct the temperature values ​​of the actual observation data by reflecting the battery's usage environment and external environmental factors through the temperature correction process. Based on the corrected data, the control unit (120) can generate the first to nth histograms representing the correspondence between future voltage, current, temperature, and frequency values.

[0075] The control unit (120) may be configured to predict a target SOH representing the SOH of the battery at the end of the nth period based on the actual observed SOH representing the SOH of the battery at the end of the actual observation period and the first to nth histograms using a preset degeneration prediction model.

[0076] FIG. 3 is a drawing referenced to explain the actual observation period and the first to nth periods.

[0077] Referring to Fig. 3, the actual observation period refers to a period during which data representing changes in voltage, current, and temperature is collected by monitoring the state of the battery. The SOH calculated at the end of the actual observation period is defined as the actual observation SOH, and the histogram generated based on the data collected during the actual observation period is defined as the actual observation histogram.

[0078] Periods 1 through 9 represent consecutive intervals following the actual observation period. The battery SOH at the end of periods 1 through 9 is defined as the target SOH. In particular, period 9 is the last interval in time, and the battery SOH at the end of period 9 is defined as the target SOH. Furthermore, histograms 1 through 9 are individually associated with periods 1 through 9. That is, assuming j is a natural number less than or equal to n, the j-th histogram is associated with period j. For example, the first histogram is a histogram representing the correspondence between the voltage, current, temperature, and frequency values ​​predicted for period 1.

[0079] Specifically, the control unit (120) can predict the target SOH by inputting the first to nth histograms and the actual observed SOH into the degeneration prediction model.

[0080] The degradation prediction model is a model that utilizes deep learning technology. Deep learning is a type of machine learning algorithm and is a technology based on an artificial neural network composed of multiple layers. Deep learning-based degradation prediction models can effectively learn the complex non-linear relationships inherent in large-scale data, thereby enabling the prediction of a battery's State of Health (SOH) with high reliability.

[0081] Specifically, the degradation prediction model may be a deep learning model trained by inputting a histogram representing the correspondence between voltage values, current values, temperature values, and frequency values, and the SOH of the battery at a specific point in time as a training data set.

[0082] For example, the degradation prediction model may be a model trained by inputting a reference histogram representing the correspondence between voltage, current, temperature, and frequency values ​​during a reference period and the SOH of the battery at the end of the reference period as a training data set.

[0083] That is, the SOH prediction device (100) can predict the target SOH at a specific point in the future with high reliability based on past actual observation data and a pre-learned degeneration prediction model.

[0084] Meanwhile, the data acquisition unit (110) and / or control unit (120) provided in the SOH prediction 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 data acquisition unit (110) and / or control unit (120) may be implemented as a set of program modules. In this case, the program modules may be stored in memory and executed by the data acquisition unit (110) and / or control unit (120). The memory may be located inside or outside the data acquisition unit (110) and / or control unit (120) and may be connected to the data acquisition unit (110) and / or control unit (120) by various well-known means.

[0085] Additionally, the SOH prediction device (100) may further include a storage unit (130). The storage unit (130) may store data or programs necessary for each component of the SOH prediction 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 known to be able to record, erase, update, and read data. As an example, the information storage means may include RAM (Random Access Memory), Flash Memory (Read Only Memory), ROM (Read Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), Register, etc. Additionally, the storage unit (130) may store program codes that define processes executable by the data acquisition unit (110) and / or the control unit (120).

[0086] Specifically, the storage unit (130) can store information necessary for the control unit (120) to predict the target SOH of the battery. The control unit (120) can access the storage unit (130) to obtain information necessary for the control unit (120) to predict the target SOH of the battery. For example, the actual observation histogram obtained by the data acquisition unit (110) is stored in the storage unit (130), and the control unit (120) can access the storage unit (130) to obtain the stored actual observation histogram. The degradation prediction model may be recorded in the storage unit (130).

[0087] In one embodiment, the time length of each of the first to nth periods can be set to be the same as the time length of the actual observation period. Through this, the control unit (120) can generate the first to nth histograms, which are prediction data, more consistently based on the actual observation histogram generated from the actual observation data. Specifically, the equality of time lengths can contribute to improving the accuracy and reliability of the prediction data by unifying the temporal standard during the data generation process.

[0088]

[0089] The control unit (120) may be configured to generate the first to nth histograms based on the comparison results between the representative temperature value of the actual observation histogram and each of the first to nth reference temperature values.

[0090] Here, the representative temperature value refers to a representative value calculated from multiple temperature values ​​included in the actual observation histogram. For example, the representative value may be the mean, median, mode, or a value based on user-defined rules.

[0091] The first to nth reference temperature values ​​are values ​​that serve as standards for temperature correction and can be set according to a predetermined standard. Additionally, the first to nth reference temperature values ​​can constitute a reference temperature data set.

[0092] For example, the control unit (120) may be configured to generate a reference temperature data set based on a plurality of temperature data sets collected from a plurality of different batteries over a past data collection period having a time length greater than or equal to the total time length of the first to nth periods.

[0093] Here, multiple other batteries may consist of batteries of the same type as the target battery for SOH prediction. The multiple other batteries may include batteries with varying degrees of degradation, manufacturing environments, and usage conditions. When the target battery is included in the multiple other batteries, the temperature data of the target battery is directly utilized in the process of generating a reference temperature data set, thereby generating a reference temperature data set that more effectively reflects the usage conditions and environment of the target battery. On the other hand, when the target battery is not included in the multiple other batteries, the temperature data of the other batteries is utilized to generate a more generalized reference temperature data set.

[0094] Multiple temperature data sets refer to data sets containing temperature values ​​measured from multiple different batteries during a past data collection period. Each temperature data set contains temperature values ​​measured from individual batteries and may consist of temporally continuous data. Temperature data sets may include various temperature values ​​that reflect the battery usage environment and usage conditions during a specific period. Each temperature value in the multiple temperature data sets may be indexed with measurement timing information (e.g., month, day, hour, minute, second).

[0095] The control unit (120) can classify multiple temperature values ​​included in multiple temperature data sets into first to x data sets according to a predetermined rule. x is a natural number greater than or equal to n. For example, if the time unit for classification is 'day', x may be 365. In this case, the first data set contains temperature values ​​measured on January 1st regardless of the year, the 32nd data set contains temperature values ​​measured on February 1st regardless of the year, and the 365th data set contains temperature values ​​measured on December 31st regardless of the year. As another example, if the time unit for classification is 'month', x may be 12.

[0096] Specifically, the control unit (120) can determine the first to n groups corresponding to each of the first to n periods from the first to x data sets. The control unit (120) can classify each data set having time information belonging to the time range of the j period among the first to x data sets into the j group. For example, when m = 365 and the first period = January 10 to February 9, the first group may include all temperature values ​​of the 10 to 40 data sets.

[0097] Additionally, the control unit (120) can set a representative value of the temperature values ​​included in the j-th group as the j-th reference temperature value. Here, the representative value may be an average value, a median value, a mode value, or a value based on a user-defined rule. Preferably, the representative temperature value of the actual observation histogram and the first to n-th reference temperature values ​​may be set according to the same standard.

[0098] The control unit (120) can determine the first to nth temperature correction values ​​based on a representative temperature value and the first to nth reference temperature values.

[0099] Specifically, the control unit (120) may be configured to determine a j-th temperature correction value, which is the difference between a representative temperature value and a j-th reference temperature value among the first to n-th reference temperature values. That is, specifically, the control unit (120) may compare the representative temperature value and the j-th reference temperature value to calculate the difference between the two values ​​and determine the calculated difference as the j-th temperature correction value.

[0100] The control unit (120) may be configured to correct each temperature value of the actual observation histogram based on the j-th temperature correction value to generate the j-th histogram among the first to n-th histograms.

[0101] Specifically, the control unit (120) can compare a representative temperature value with a reference temperature value and determine a method of correction calculation based on the comparison result.

[0102] For example, if the j-th reference temperature value exceeds the representative temperature value, the control unit (120) can perform an operation of adding the j-th temperature correction value to each temperature value of the actual observation histogram. Then, the control unit (120) can generate the j-th histogram composed of the corrected temperature values. As another example, if the j-th reference temperature value is less than the representative temperature value, the control unit (120) can perform an operation of subtracting the j-th temperature correction value from each temperature value of the actual observation histogram. Then, the control unit (120) can generate the j-th histogram composed of the corrected temperature values. As yet another example, if the j-th reference temperature value is the same as the representative temperature value, the actual observation histogram can be set as the j-th histogram. In this case, the correction operation may not be performed.

[0103]

[0104] The degradation prediction model may be trained by multiple training data sets collected from multiple different batteries over multiple reference periods.

[0105] Each of the multiple training data sets may include a reference histogram representing the correspondence between the voltage, current, and temperature values ​​of another battery during a reference period prior to the actual observation period, and the SOH of another battery at the start of the reference period as features, and the SOH of another battery at the end of the reference period as a label.

[0106] Here, the reference period refers to a specific period set to collect data used for training the degeneration prediction model, and is the interval prior to the actual observation period.

[0107] Preferably, the time length of each of the multiple reference periods may be equal to the total time length of the first to nth periods. By setting the time scales of the training data set and the actual observation data set to be similar, the degeneration prediction model can maintain consistency between the data used to train the model and the data used for prediction, thereby increasing the accuracy of the prediction.

[0108] Meanwhile, the battery used to generate the training data set and the battery used to generate the reference temperature data set may be the same or different.

[0109] A reference period and a training data set may be assigned to each of the plurality of batteries. For example, a first to k-th reference period corresponding to the first to k-th batteries (i.e., a plurality of different batteries) may be set, and a first to k-th training data set may be configured based on the first to k-th reference periods. k is a natural number greater than or equal to 2.

[0110] Specifically, data collected during the m-th reference period for the m-th battery, which is one of the first to k-th batteries, may constitute the m-th learning data set. The m-th learning data set may include the m-th reference histogram showing the correspondence between voltage values, current values, temperature values, and frequency values ​​during the m-th reference period, the SOH of the m-th battery at the start of the m-th reference period, and the SOH of the m-th battery at the end of the m-th reference period.

[0111] More specifically, the histogram of reference m and the SOH of battery m at the start of reference period m are input as features into the degradation prediction model, and the SOH of battery m at the end of reference period m can be input as a label into the degradation prediction model.

[0112] Figures 4 and 5 are schematic diagrams illustrating the structure of a degeneration prediction model.

[0113] Referring to Fig. 4, the input data of the degeneration prediction model is the first to nth histograms and the actual observed SOH, and the output result is the target SOH.

[0114] The degeneration prediction model can be composed of a CNN-based model. For example, the degeneration prediction model can be composed of a 3D-CNN-based deep learning model.

[0115] The degeneration prediction model can be configured to predict the target SOH by receiving the first to nth histograms and the actual observed SOH as input data.

[0116] Referring to Fig. 5, the data processing process of the degeneration prediction model using 3D-CNN can be schematically observed.

[0117] The degeneration prediction model may include an input layer, a convolutional layer, a fully connected layer, and an output layer. The input layer may pass the first to the nth histogram as input data to the convolutional layer. Each convolutional layer may extract feature maps through convolution operations on the input data and reduce the size of the feature maps through pooling operations. The feature maps extracted from the last convolutional layer may be converted into the form of a one-dimensional vector and connected to the fully connected layer. The fully connected layer may perform various functions, such as classification or regression, identical to those of a general neural network model. In the output layer, the output of a prediction result (i.e., target SOH) based on a softmax function, etc., may be performed.

[0118] The degeneration prediction model can progressively reduce data and extract feature maps of the input data by repeatedly performing 3D convolution operations, batch normalization, ReLU (Recified Linear Unit) activation function operations, and / or MaxPooling operations on the input first to nth histograms. The feature maps extracted from the convolution layer and pooling layer of the CNN can be connected and integrated through a fully connected layer.

[0119] Feature data generated in a fully connected layer can be combined with real-observed SOHs. Here, combining means including real-observed SOHs as additional input values ​​in the nodes of the fully connected layer, which increases the number of input nodes in the fully connected layer.

[0120] Data with added actual observed SOH is processed through a fully connected layer, batch normalization, and / or ReLU activation function operations, and a target SOH can finally be output. As the principles and operation of 3D-CNN are already widely known, further explanation thereof will be omitted.

[0121] The SOH prediction device (100) can predict the SOH of a battery more accurately and reliably by using a 3D-CNN-based degradation prediction model. Specifically, the 3D-CNN effectively learns complex non-linear relationships between voltage, current, and temperature within the input data and can automatically extract feature maps. Through this, the state change of the battery can be precisely analyzed to predict the target SOH more accurately and reliably.

[0122]

[0123] The control unit (120) can be configured to predict the degradation rate of the battery over the first to nth periods based on the actual observed SOH and the target SOH.

[0124] Here, the degradation rate refers to the rate or speed at which the SOH of the battery decreases during the first to nth periods.

[0125] Specifically, the control unit (120) can calculate the difference between the actual observed SOH and the target SOH. Then, the control unit (120) can calculate the degradation rate by dividing the calculated SOH difference by the total time length of the first to nth periods. The degradation rate numerically represents the trend of SOH reduction during the future usage period of the battery.

[0126] The predicted degradation rate can be utilized for long-term performance evaluation and management of the battery. For example, the control unit (120) can compare the degradation rate with a preset threshold and set usage conditions for the battery based on the comparison result. Specifically, if the degradation rate exceeds the threshold, the control unit (120) can adjust the charge / discharge pattern or change the usage conditions. As another example, the control unit (120) can monitor the pattern of change in the degradation rate and, if an abnormal change in the degradation rate is predicted, set usage conditions for the battery.

[0127] The SOH prediction device (100) can improve the safety and reliability of the battery by predicting performance degradation in long-term use based on the degradation rate and setting usage conditions for the optimized battery based on this.

[0128] The SOH prediction device (100) according to the present invention can be applied to a BMS. That is, the BMS according to the present invention may include the SOH prediction device (100) described above. In this configuration, at least some of the components of the SOH prediction device (100) may be implemented by supplementing or adding the functions of the components included in the conventional BMS. For example, the data acquisition unit (110) and the control unit (120) of the SOH prediction device (100) may be implemented as components of the BMS.

[0129] FIG. 6 is a schematic diagram illustrating a battery pack (10) according to another embodiment of the present invention.

[0130] The SOH prediction device (100) according to the present invention may be provided in a battery pack (10). That is, the battery pack (10) according to the present invention may include the SOH prediction device (100) described above and one or more battery cells. In addition, the battery pack (10) may further include electrical components (relays, fuses, etc.) and a case, etc.

[0131] The positive terminal of the battery (11) can be connected to the positive terminal (P+) of the battery pack (10), and the negative terminal of the battery (11) can be connected to the negative terminal (P-) of the battery pack (10).

[0132] The measuring unit (12) can be connected to the first sensing line (SL1), the second sensing line (SL2), and the third sensing line (SL3). Specifically, the measuring unit (12) can be connected to the positive terminal of the battery (11) through the first sensing line (SL1) and to the negative terminal of the battery (11) through the second sensing line (SL2). The measuring unit (12) can measure the voltage of the battery (11) based on the voltage measured at each of the first sensing line (SL1) and the second sensing line (SL2).

[0133] Additionally, the measuring unit (12) can be connected to a current measuring unit (A) through a third sensing line (SL3). 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 charging amount by measuring the charging current of the battery (11) through the third sensing line (SL3). Furthermore, the measuring unit (12) can calculate the discharging amount by measuring the discharging current of the battery (11) through the third sensing line (SL3).

[0134] The data acquisition unit (110) can be connected via wired and / or wireless means to communicate with the measurement unit (12). The data acquisition unit (110) can receive voltage, current, and / or temperature information of the battery (11) from the measurement unit (12).

[0135] FIG. 7 is a schematic drawing of a vehicle (1) according to another embodiment of the present invention.

[0136] Referring to FIG. 7, the battery pack (10) described above with reference to FIG. 6 may be included in a vehicle (1), such as an electric vehicle (EV) or a hybrid vehicle (HV). The battery pack (10) can drive the vehicle (1) by supplying power to a motor through an inverter provided in the vehicle (1). Here, the battery pack (10) may include a SOH prediction device (100). In this case, the SOH prediction device (100) may be an on-board device included in the vehicle (1).

[0137] FIG. 8 is a schematic diagram illustrating a server (2) according to another embodiment of the present invention.

[0138] Referring to FIG. 8, the SOH prediction device (100) according to the present invention may be provided in a server (2). The server (2) can predict the target SOH of the battery by providing high-performance computing resources and data storage functions.

[0139] The SOH prediction device (100) provided in the server (2) can be linked with a plurality of BMS (3), user terminals (4) and / or vehicle control systems (5), etc., to perform integrated management of a battery system including a plurality of batteries. The server (2) can be connected via wired and / or wireless connections to enable communication with a plurality of BMS (3) and / or user terminals (4).

[0140] The server (2) is linked with the BMS (3) and can transmit the SOH prediction result, the battery degradation rate and / or the diagnosis result in real time to the corresponding BMS (3). Alternatively, if the battery condition is diagnosed as abnormal, the server (2) can transmit a warning or control signal to the corresponding BMS (3).

[0141] The server (2) can be linked with the user terminal (4) to allow the user to remotely monitor the status of the battery. The user can check the status of the battery in real time using a dedicated application.

[0142] FIG. 9 is a schematic diagram illustrating a method for predicting SOH according to another embodiment of the present invention. FIG. 10 is a schematic diagram illustrating the sub-steps of step S920 of FIG. 9.

[0143] Each step of the SOH prediction method can be performed by the SOH prediction device (100). For convenience of explanation, details that overlap with previously described content will be omitted or briefly explained below.

[0144] Referring to FIGS. 1 to 9, the SOH prediction method includes steps S910, S920, and S930.

[0145] Step S910 is a step of acquiring an actual observation histogram showing the correspondence between the voltage value, current value, and temperature value of the battery during the actual observation period, and can be performed by the data acquisition unit (110).

[0146] Step S920 can be performed by the control unit (120) as a step of generating a first to nth histogram corresponding to a first to nth period following the actual observation period, based on an actual observation histogram and a reference temperature data set.

[0147] Referring to FIG. 10, step S920 may include step S1010, step S1020, step S1030 and step S1040.

[0148] Step S1010 is a step of generating a reference temperature data set based on a plurality of temperature data sets, which can be performed by a control unit (120). For example, the control unit (120) may be configured to generate a reference temperature data set based on a plurality of temperature data sets collected from a plurality of different batteries over a past data collection period having a time length greater than or equal to the total time length of the first to nth periods.

[0149] Step S1020 is a step of comparing the representative temperature value of the actual observation histogram with each of the first to n reference temperature values, and can be performed by the control unit (120).

[0150] Step S1030 is a step of determining the first to nth temperature correction values ​​by calculating the difference between the representative temperature value of the actual observation histogram and the first to nth reference temperature values, and can be performed by the control unit (120).

[0151] Step S1040 is a step of generating the first to nth histograms by correcting the temperature values ​​of the actual observation histograms based on the first to nth temperature correction values, and can be performed by the control unit (120).

[0152] Specifically, the control unit (120) can compare a representative temperature value with a reference temperature value and determine a method of correction calculation based on the comparison result.

[0153] For example, if the j-th reference temperature value exceeds the representative temperature value, the control unit (120) can perform an operation of adding the j-th temperature correction value to each temperature value of the actual observation histogram. Then, the control unit (120) can generate the j-th histogram composed of the corrected temperature values. As another example, if the j-th reference temperature value is less than the representative temperature value, the control unit (120) can perform an operation of subtracting the j-th temperature correction value from each temperature value of the actual observation histogram. Then, the control unit (120) can generate the j-th histogram composed of the corrected temperature values. As yet another example, if the j-th reference temperature value is the same as the representative temperature value, the actual observation histogram can be set as the j-th histogram. In this case, the correction operation may not be performed.

[0154] Step S930 is a step of predicting a target SOH representing the SOH of the battery at the end of the nth period from the actual observed SOH representing the SOH of the battery at the end of the actual observation period and the first to nth histograms based on a preset degradation prediction model, and can be performed by a control unit (120).

[0155] Specifically, the control unit (120) can predict the target SOH by inputting the first to nth histograms and the actual observed SOH into the degeneration prediction model.

[0156] The degradation prediction model may be a deep learning model trained by inputting a histogram representing the correspondence between voltage values, current values, temperature values, and frequency values, and the SOH of the battery at a specific time point as a training data set.

[0157] For example, the degradation prediction model may be a model trained by inputting a reference histogram representing the correspondence between voltage, current, temperature, and frequency values ​​during a reference period and the SOH of the battery at the end of the reference period as a training data set.

[0158]

[0159] The control unit (120) may be configured to change preset usage conditions for the battery based on the predicted target SOH or degradation rate.

[0160] Specifically, the control unit (120) can appropriately change the preset usage conditions to correspond to the state of the battery.

[0161] Here, the usage conditions may include at least one of the maximum allowable temperature, upper limit SOC, lower limit SOC, and upper limit C-rate.

[0162] In one embodiment, the control unit (120) compares the target SOH with a preset threshold SOH and can change the usage conditions based on the comparison result.

[0163] Here, the critical SOH can be defined as the SOH that serves as a criterion for determining the point at which the likelihood of battery performance or safety degradation increases. The critical SOH can be theoretically and / or experimentally pre-set.

[0164] For example, if the target SOH is below the critical SOH, the control unit (120) can perform control to lower at least one of the maximum allowable temperature, upper limit SOC, and upper limit C-rate. As another example, if the target SOH is below the critical SOH, the control unit (120) can perform control to raise the lower limit SOC.

[0165] In addition, the degree of change in usage conditions can be adjusted in proportion to the degree of battery degradation. Specifically, the adjustment level of the applied usage conditions can be determined according to a control protocol that indicates the correspondence between SOH and the degree of adjustment of usage conditions.

[0166] The control protocol includes adjustment coefficients or adjustment maps set for each SOH range, and can be configured such that, for example, normal conditions are maintained without adjustment when the target SOH is 90% or higher, the C-rate is reduced by 10% when the target SOH is 80% or higher but less than 90%, and the C-rate is reduced by 20% when the target SOH is less than 80%. The control protocol may be predefined based on experimental data, simulation results, or system-specific requirements.

[0167] In another embodiment, the degeneration rate can be compared with a preset threshold rate, and the usage conditions can be changed based on the comparison result.

[0168] Here, the critical rate can be defined as the State of Health (SOH) serving as a criterion for determining the point at which the likelihood of accelerated battery degradation increases. Accelerated degradation may refer to a state where the rate of SOH reduction increases rapidly, making it highly likely that lifespan reduction or safety issues will occur. The critical rate can be theoretically and / or experimentally predetermined.

[0169] For example, if the degradation rate is above a critical rate, the control unit (120) can perform control to lower at least one of the maximum allowable temperature, upper limit SOC, and upper limit C-rate. As another example, if the degradation rate is above a critical rate, the control unit (120) can perform control to raise the lower limit SOC.

[0170] In another embodiment, the control unit (120) may be configured to output an alarm if the target SOH is below the threshold SOH or if the degradation rate is above the threshold rate. That is, the control unit (120) can notify the outside of the battery status by immediately outputting an alarm. For example, the control unit (120) may output an alarm notifying the battery status to an alarm unit (not shown), a display unit (not shown), a user terminal (not shown), and a server, etc., connected via wired and / or wireless communication.

[0171] According to one embodiment of the present invention, the SOH prediction device (100) can increase the expected lifespan of the battery and prevent safety accidents caused by abnormal degradation of the battery by taking appropriate measures according to the prediction result.

[0172] 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.

[0173] 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.

[0174] 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.

[0175] 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.

[0176] 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.

[0177] In addition, the program may be provided as part of a computer program product. Computer program products may be traded between a seller and a buyer as goods.

[0178] 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.

[0179] 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.

[0180] 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.

[0181] 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.

[0182]

[0183] (Explanation of symbols)

[0184] 1: Car

[0185] 2: Server

[0186] 3: BMS

[0187] 4: User terminal

[0188] 10: Battery pack

[0189] 11: Battery

[0190] 12: Measurement section

[0191] 100: SOH Prediction Device

[0192] 110: Data acquisition unit

[0193] 120: Control unit

[0194] 130: Storage section

Claims

1. A data acquisition unit configured to acquire an actual observation histogram representing the correspondence between the voltage, current, temperature, and frequency values ​​of the battery during an actual observation period; and Based on the above actual observation histogram and reference temperature dataset, the first to nth histograms corresponding to the first to nth periods following the actual observation period are generated, and An SOH prediction device comprising a control unit configured to predict a target SOH representing the SOH of the battery at the end of the n-th period from the actual observed SOH representing the SOH of the battery at the end of the actual observation period and the first to n-th histograms based on a preset degradation prediction model, wherein n is a natural number greater than or equal to 2.

2. In Paragraph 1, The above reference temperature dataset includes first to n reference temperature values ​​corresponding to the first to n periods, and The above control unit is, A SOH prediction device configured to generate the first to nth histograms based on the comparison results between the representative temperature value of the above actual observation histogram and each of the first to nth reference temperature values.

3. In Paragraph 2, The above control unit is, SOH prediction device configured to generate the reference temperature dataset based on a plurality of temperature datasets collected from a plurality of different batteries over a past data collection period having a time length greater than or equal to the total time length of the first to nth periods.

4. In Paragraph 2, The above control unit is, A j-th temperature correction value is determined, which is the difference between the above representative temperature value and the j-th reference temperature among the above first to n-th reference temperature values, and A SOH prediction device configured to generate the j-th histogram among the first to n-th histograms by correcting each temperature value of the actual observation histogram based on the j-th temperature correction value, wherein j is a natural number less than or equal to n.

5. In Paragraph 1, The above degeneration prediction model is, It is learned by multiple training data sets collected from multiple different batteries during multiple reference periods, and A SOH prediction device in which each of the above plurality of training data sets includes, as features as a reference histogram representing the correspondence relationship between the voltage value, current value, temperature value, and frequency value of another battery during a reference period prior to the actual observation period, and the SOH of the other battery at the start of the reference period, and the SOH of the other battery at the end of the reference period as a label.

6. In Paragraph 5, A SOH prediction device characterized in that the time length of each of the plurality of reference periods is equal to the total time length of the first to nth periods.

7. In Paragraph 1, A SOH prediction device characterized in that the time length of each of the first to n periods is the same as the time length of the actual observation period.

8. In Paragraph 1, The above degeneration prediction model is a SOH prediction device composed of a CNN (Convolutional Neural Network) based model.

9. In Paragraph 1, The above control unit is, An SOH prediction device configured to predict the degradation rate of the battery over the first to nth periods based on the actual observed SOH and the target SOH.

10. A battery pack comprising a SOH prediction device according to any one of claims 1 to 9.

11. An automobile comprising a SOH prediction device according to any one of paragraphs 1 through 9.

12. A server comprising an SOH prediction device according to any one of paragraphs 1 through 9.

13. A step of obtaining an actual observation histogram showing the correspondence between the voltage, current, temperature, and frequency of the battery during the actual observation period; Based on the above actual observation histogram and reference temperature dataset, the step of generating a first to nth histogram corresponding to a first to nth period following the actual observation period; and A method for predicting an SOH, wherein n is a natural number greater than or equal to 2, comprising the step of predicting a target SOH representing the SOH of the battery at the end of the n-th period from the actual observed SOH representing the SOH of the battery at the end of the actual observation period and the first to n-th histograms based on a preset degradation prediction model.

Images