Welding management device, welding management method, and welding management system

Abstract

According to some embodiments, a welding management device comprises: an interface configured to acquire electrical data collected during a welding process for a battery cell; and a controller configured to derive a state score indicating the welding state of the battery cell from the electrical data on the basis of a state diagnosis model, compare a first distribution of state scores for target battery cells and a second distribution of state scores of training data used to train the state diagnosis model, and thereby determine whether a data shift of the first distribution occurs, and adjust the state diagnosis model on the basis of differences between the first distribution and the second distribution when the data shift occurs.

Description

Welding management device, welding management method and welding management system

[0001] Cross-citation with related applications

[0002] This application claims the benefit of priority to Republic of Korea Patent Application No. 10-2024-0079446, filed June 19, 2024, the entire contents of which are incorporated herein by reference.

[0003] Technology field

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

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

[0006] A welding process may be performed to manufacture battery cells. For example, electric resistance welding may be performed to form the positive tab of a cylindrical battery cell. During the welding process, electrical data such as voltage, current, temperature, and resistance of the welding device may be collected, and the welding status of the battery cell may be determined based on this electrical data. However, if the status of the welding device changes, the electrical data used to determine the welding status may also change, which may reduce the accuracy of the welding status diagnosis.

[0007] One purpose of the embodiments disclosed in this document is to provide a welding management device, a welding management method, and a welding management system that can solve the problem of a decrease in accuracy in diagnosing a welding state based on electrical data due to a change in the state of a welding device.

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

[0009] According to some embodiments, a welding management device includes an interface configured to acquire electrical data collected during a welding process for a battery cell; and a controller configured to derive a condition score representing a welding condition of the battery cell from the electrical data based on a condition diagnosis model, compare a first distribution of condition scores for target battery cells with a second distribution of condition scores of learning data used to train the condition diagnosis model to determine whether a data shift of the first distribution occurs, and adjust the condition diagnosis model based on a difference between the first distribution and the second distribution when the data shift occurs.

[0010] According to some embodiments, the controller is configured to determine whether the data shift occurs based on a statistical hypothesis test for the first distribution and the second distribution.

[0011] According to some embodiments, the controller is configured to determine that the data shift has occurred when a significance probability (p-value) representing the probability that a null hypothesis that the first mean of the first distribution and the second mean of the second distribution are equal to each other is established is less than or equal to a threshold.

[0012] According to some embodiments, the controller is configured to adjust a threshold score used to determine a weld condition of the battery cell as defective when the data shift occurs.

[0013] According to some embodiments, the controller is configured to adjust the threshold score based on a difference between a first representative score among the condition scores for the target battery cells and a second representative score among the condition scores of the learning data.

[0014] According to some embodiments, the target battery cells include a reference number or more of battery cells that are most recently produced among a plurality of battery cells mass-produced by the welding process.

[0015] According to some embodiments, the data shift occurs due to replacement of consumables of a welding device performing the welding process.

[0016] According to some embodiments, a welding management method includes the steps of: obtaining electrical data collected during a welding process for a battery cell; deriving a condition score representing a welding condition of the battery cell from the electrical data based on a condition diagnosis model; comparing a first distribution of condition scores for target battery cells with a second distribution of condition scores of learning data used to train the condition diagnosis model to determine whether a data shift of the first distribution occurs; and adjusting the condition diagnosis model based on a difference between the first distribution and the second distribution when the data shift occurs.

[0017] According to some embodiments, the step of determining whether the data shift occurs comprises the step of determining whether the data shift occurs based on a statistical hypothesis test for the first distribution and the second distribution.

[0018] According to some embodiments, the step of determining whether the data shift occurs includes the step of determining that the data shift occurs when a significance probability (p-value) representing the probability that a null hypothesis assuming that the first mean of the first distribution and the second mean of the second distribution are equal to each other is established is less than or equal to a threshold value.

[0019] According to some embodiments, the step of adjusting the condition diagnosis model includes the step of adjusting a threshold score used to determine a weld condition of the battery cell as defective when the data shift occurs.

[0020] According to some embodiments, the step of adjusting the condition diagnosis model includes the step of adjusting the threshold score based on a difference between a first representative score among the condition scores for the target battery cells and a second representative score among the condition scores of the learning data.

[0021] According to some embodiments, the target battery cells include a reference number or more of battery cells that are most recently produced among a plurality of battery cells mass-produced by the welding process.

[0022] According to some embodiments, the data shift occurs due to replacement of consumables of a welding device performing the welding process.

[0023] According to some embodiments, a welding management system includes a welding device configured to perform a welding process for a battery cell; and a welding management device configured to obtain electrical data collected during the welding process, derive a condition score representing a welding condition of the battery cell from the electrical data based on a condition diagnosis model, compare a first distribution of condition scores for target battery cells with a second distribution of condition scores of learning data used to train the condition diagnosis model, determine whether a data shift of the first distribution occurs, and adjust the condition diagnosis model based on a difference between the first distribution and the second distribution when the data shift occurs.

[0024] According to the embodiments disclosed in this document, a welding management device, a welding management method, and a welding management system can be provided that can solve the problem of the accuracy of diagnosing a welding state based on electrical data deteriorating due to a change in the state of a welding device.

[0025] The technical effects according to the embodiments disclosed in this document are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art according to the disclosure of this document.

[0026] Figure 1 illustrates how a welding management system operates according to some embodiments.

[0027] FIG. 2 illustrates elements constituting a welding management device according to some embodiments.

[0028] FIG. 3 illustrates a process for adjusting a threshold score of a state diagnosis model in response to a data shift according to some embodiments.

[0029] Figure 4 illustrates state scores calculated before and after a data shift occurs according to some embodiments.

[0030] Figures 5 and 6 illustrate a form of over-inspection in which normal data is incorrectly judged as defective due to data shift according to some embodiments.

[0031] Figure 7 illustrates steps constituting a welding management method according to some embodiments.

[0032] Hereinafter, embodiments described in this document are described with reference to the attached drawings. However, this is not intended to limit the disclosure of this document to specific embodiments, and it should be understood that various modifications, equivalents, and / or alternatives of the embodiments described in this document are included.

[0033] The embodiments and terminology used in this document are not intended to limit the technical features described in this document to a specific embodiment, but should be understood to encompass various modifications, equivalents, or alternatives of the embodiment. In connection with the description of the drawings, similar reference numerals may be used to refer to similar or related components. The singular form of a noun corresponding to an item may include one or more of the item, unless the context clearly indicates otherwise.

[0034] In this document, the phrases "A or B", "at least one of A and B", "at least one of A or B", "A, B, or C", "at least one of A, B, and C", and "at least one of A, B, or C" can each include any one of the items listed together in that phrase, or all possible combinations thereof. Terms such as "first", "second", "first", "second", "A", "B", "(a)", or "(b)" may be used merely to distinguish the corresponding component from other corresponding components, and do not limit the corresponding components in any other respect (e.g., importance or order) unless specifically stated otherwise.

[0035] In this document, when a component (e.g., a first component) is referred to as being “connected,” “coupled,” or “connected,” with or without the terms “functionally” or “communicatively,” or “coupled” or “connected,” it means that the component can be connected to the other component directly (e.g., wired or wirelessly), or indirectly (e.g., via a third component).

[0036] The methods according to various embodiments disclosed in this document may be provided as a computer program product. The computer program product may be traded as a commodity between a seller and a buyer. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., compact disc read-only memory, CD-ROM), or may be distributed online (e.g., downloaded or uploaded) through an application store or directly between two driver devices. In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily generated in a machine-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0037] According to the embodiments disclosed in this document, each component (e.g., a module or a program) of the above-described components may include one or more entities, and some of the entities may be separated and placed in other components. According to the embodiments disclosed in this document, one or more components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to the integration. According to the embodiments disclosed in this document, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0038] Figure 1 illustrates how a welding management system operates according to some embodiments.

[0039] Referring to FIG. 1, the welding management system (100) may include a welding device (120) that performs a welding process for a battery cell (110) and a welding management device (130) that manages the welding process. However, the present invention is not limited thereto, and some components may be omitted from the welding management system (100), or other general-purpose components may be further included in the welding management system (100).

[0040] The battery cell (110) may be formed by a welding process. For example, the battery cell (110) may include a cylindrical battery cell. The cylindrical battery cell may include rivets, collectors, tabs, beading, etc. formed on the positive or negative electrode, and the detailed structure of such a cylindrical battery cell may be formed by welding.

[0041] A welding device (120) may perform a welding process to form a battery cell (110). The welding process may include a resistance welding process. During the resistance welding process, voltage and current may be input to the welding device (120), the welding device (120) may output voltage and current to the welding target, and the temperature, resistance, etc. of the welding target may be measured. The quality of the welding process for the battery cell (110) may be determined based on whether the electrical data during the welding process, such as the above, is outside an appropriate range. Based on this, a model for estimating the welding status based on the electrical data may be utilized.

[0042] The welding management device (130) can manage the welding process for the battery cell (110) performed by the welding device (120). The welding management device (130) can diagnose the welding status of the battery cell (110) using a status diagnosis model. For example, the welding management device (130) can determine whether the status diagnosis model for diagnosing the welding status is operating normally or whether correction of the status diagnosis model is required.

[0043] FIG. 2 illustrates elements constituting a welding management device according to some embodiments.

[0044] Referring to FIG. 2, the welding management device (130) may include an interface (131) and a controller (132). However, the present invention is not limited thereto, and some components may be omitted from the welding management device (130), or other general-purpose components may be further included in the welding management device (130).

[0045] The interface (131) may be configured to acquire electrical data collected during a welding process for a battery cell (110). For example, the interface (131) may include a sensor configured to measure electrical data, and the sensor may include a voltage sensor, a current sensor, a temperature sensor, a resistance sensor, or the like. Alternatively, a sensor configured to measure electrical data may be provided externally to the welding management device (130), and the interface (131) may include a communication unit configured to receive electrical data from an external source via a wire and / or wirelessly.

[0046] The controller (132) may include memory and a processor. The processor of the controller (132) may be implemented as an array of multiple logic gates for processing various operations or as a general-purpose microprocessor, and may be comprised of a single processor or multiple processors. For example, the processor may be implemented in the form of at least one of a microprocessor, a CPU, a GPU, and an AP.

[0047] The memory of the controller (132) can store various data, commands, mobile applications, computer programs, etc. The processor can process various operations by executing commands stored in the memory. For example, the memory can be implemented as a non-volatile device such as ROM, PROM, EPROM, EEPROM, flash memory, PRAM, MRAM, RRAM, FRAM, etc., or a volatile device such as DRAM, SRAM, SDRAM, PRAM, etc., and can be implemented in the form of an HDD, SSD, SD, Micro-SD, etc., or a combination thereof.

[0048] The controller (132) may be configured to derive a condition score representing the welding status of the battery cell (110) from the electrical data based on a condition diagnosis model. For example, the condition diagnosis model may derive a condition score representing the welding status of the battery cell (110) using an isolation forest (IF) algorithm with a binary tree partitioning method. The IF algorithm may be configured to derive the condition score based on a path distance from a root node. In an embodiment, a threshold score may be set to distinguish between normal welding and defective welding, and any battery cell having a condition score exceeding the threshold score may be determined to have a defective welding. In an embodiment, the condition diagnosis model may derive a condition score representing the welding status of the battery cell (110) using a local outlier factor (LOF) algorithm. The LOF algorithm may derive the condition score by utilizing local information of the electrical data to provide an indicator representing the degree of outlier.

[0049] The controller (132) may be configured to compare a first distribution of condition scores for target battery cells with a second distribution of condition scores of learning data used to train a condition diagnosis model to determine whether a data shift of the first distribution occurs. The target battery cells may include a first number or more of cells, and the first number may be, for example, 10, 20, 25, 30, 50, 100, 150, 200, 300, 500, or any other suitable number. The target battery cells may be set by the welding management device (130) or by a user of the welding management device (130). For example, the user may set the target battery cells when a decrease in the accuracy of the condition diagnosis model is suspected. Alternatively, the welding management device (130) may set cells before and after a point in time when the diagnostic performance of the welding condition decreases below a certain level as target battery cells. The learning data used to train the condition diagnosis model can be continuously updated while the welding management device (130) is operating. The current learning data can be used when determining the second distribution. According to an embodiment, the distribution of condition scores can be determined by statistical variables such as the mean, median, and standard deviation. In this case, if the difference between the statistical variables in the first distribution and the second distribution exceeds a threshold, it can be determined that a data shift has occurred. The threshold can be set and changed according to the performance requirements of the condition diagnosis model. For example, if the difference between the mean of the first distribution and the mean of the second distribution exceeds a threshold, it can be determined that a data shift has occurred.

[0050] The controller (132) may be configured to adjust the condition diagnosis model based on the difference between the first distribution and the second distribution when a data shift occurs. For example, if the difference between the mean of the first distribution and the mean of the second distribution exceeds a threshold, a data shift may be determined to have occurred, and the condition diagnosis model may be adjusted based on the average difference value. The model adjustment may include adjusting a threshold score for fault diagnosis.

[0051] According to an embodiment, the controller (132) may be configured to determine whether a data shift occurs based on a statistical hypothesis test for the first distribution and the second distribution. A statistical hypothesis test may refer to a statistical inference process that uses sample information to determine whether a hypothesis regarding the actual value of a population is valid. For example, if the hypothesis that the first distribution and the second distribution are identical is determined to be inappropriate, a data shift may be presumed to have occurred. A certain number of samples or more may be required to perform the statistical inference process. The certain number may be, for example, 50, and may be changed to another value as needed.

[0052] In an embodiment, the controller (132) may be configured to determine that a data shift occurs when a significance probability (p-value), which represents the probability that a null hypothesis that the first mean of the first distribution and the second mean of the second distribution are equal to each other, is established, is less than or equal to a threshold. If such a null hypothesis is rejected, it can be inferred that the first mean of the first distribution and the second mean of the second distribution are different from each other. The threshold used for rejecting the null hypothesis may be 0.05 corresponding to a confidence level of 95%, 0.01 corresponding to a confidence level of 99%, etc. In an embodiment, instead of the first mean and the second mean, other representative values, such as the first median of the first distribution and the second median of the second distribution, may be utilized.

[0053] According to an embodiment, the controller (132) may be configured to adjust a threshold score used to determine a welding condition of a battery cell (110) as defective when a data shift occurs. The data shift may shift the electrical data in the welding process by a certain amount overall, which may result in different data distributions being measured for the same welding process. The threshold score may be adjusted to prevent over-diagnosis, in which normal data is incorrectly diagnosed as defective due to changes in the data distribution. For example, when the standard deviation of the condition scores is σ and the mean is μ, the threshold score may be σ±3μ, and here, the value of the threshold score may be changed when a data shift occurs.

[0054] In an embodiment, the controller (132) may be configured to adjust the threshold score based on a difference between a first representative score among the condition scores for the target battery cells and a second representative score among the condition scores of the learning data. For example, if the mean of the first distribution is μ1, while the mean of the second distribution changes to μ2 in the target battery cells, the threshold score may change by μ2-μ1. Alternatively, for a median m, the threshold score may change by m2-m1. In an embodiment, a first standard deviation σ1 of the first distribution and a second standard deviation σ2 of the second distribution may be compared to determine whether it is appropriate for the second distribution to differ from the first distribution in the target battery cells, and whether a variation in the second distribution is temporary may be determined based on the comparison result. For example, the greater the difference between σ1 and σ2, the higher the likelihood that the variation in the second distribution is determined to be temporary.

[0055] In an embodiment, the target battery cells may include battery cells that are more than the most recently produced reference number among a plurality of battery cells mass-produced by the welding process. For example, if a welding process that previously exhibited a normal defect rate shows a very high defect rate after a certain point in time, it may be assumed that a factor affecting the defect determination of the welding process has recently occurred. To determine whether a change in the data distribution has occurred in recently diagnosed cells, battery cells that are more than the most recently produced reference number may be selected as the target battery cells. For example, the reference number may be 50, and the reference number may be increased or decreased for higher or lower accuracy.

[0056] In some embodiments, data shifts may occur due to replacement of consumables of a welding device (120) performing a welding process. For example, the welding rods of the welding device (120) may be periodically replaced as they are consumed as the welding process progresses. When old parts are replaced with new parts, a certain offset may occur in the measured values ​​of the electrical data. To prevent the condition diagnosis model from being retrained every time a consumable of the welding device (120) is replaced, threshold adjustments based on data distribution fluctuations may be performed.

[0057] FIG. 3 illustrates a process for adjusting a threshold score of a state diagnosis model in response to a data shift according to some embodiments.

[0058] Referring to FIG. 3, a flow (300) illustrating a process for adjusting a threshold score of a condition diagnosis model in response to a data shift may be illustrated. The flow (300) may include steps (310) through (380).

[0059] A welding process may begin at step (310). Electrical data may be collected during the welding process at step (320). The electrical data may include welding voltage, welding current, welding temperature, welding resistance, etc. At step (330), whether the welding condition is defective may be determined based on the electrical data. The determination of a welding defect may be performed using a condition score based on an isolation forest (IF) algorithm.

[0060] In step (350), it can be determined whether the number of target battery cells suspected of having data shift is 50 or more. If it is 50 or more, it can be determined whether a data shift has occurred in step (360). The occurrence of a data shift can be determined based on statistical characteristics of the distribution of target battery cells. If a data shift has occurred, the threshold score of the condition diagnosis model can be adjusted in step (370), and if it is determined in step (380) that the condition score of the battery cell exceeds the adjusted threshold score, the welding of the corresponding cell can be determined to be defective in step (390).

[0061] Figure 4 illustrates state scores calculated before and after a data shift occurs according to some embodiments.

[0062] Referring to FIG. 4, a graph (400) illustrating condition scores calculated before and after a data shift occurs may be illustrated. The horizontal and vertical axes of the graph (400) may represent the number of battery cells and the condition scores, respectively.

[0063] At point (410) of the graph (400), consumables such as welding rods of the welding device (120) may be replaced, but the critical score (420) of the condition diagnosis model may not be changed. In this case, condition scores calculated after point (410) may be offset by a certain value compared to condition scores calculated before point (410), which may result in unnecessary detection of a large number of defective cells.

[0064] Figures 5 and 6 illustrate a form of over-inspection in which normal data is incorrectly judged as defective due to data shift according to some embodiments.

[0065] Referring to FIG. 5, a graph (500) showing a first distribution before the consumables of the welding device (120) are replaced can be illustrated.

[0066] In the graph (500), the first distribution of the condition scores of the learning data of the condition diagnosis model may not overlap with the critical score, and therefore, only data that deviates significantly from the average may be judged as a welding defect.

[0067] Referring to FIG. 6, a graph (600) representing data shifts resulting from replacement of consumables of a welding device (120) may be illustrated.

[0068] As with the graph (400) of FIG. 4, the graph (600) may represent a second distribution of the condition scores of target battery cells comprised of the most recent 50 data, and the second distribution may shift by a certain offset compared to the first distribution due to the occurrence of data shift. In this case, if the critical score is not changed, over-diagnosis may occur, in which normal data is incorrectly diagnosed as defective, so adjusting the critical score may be necessary.

[0069] Figure 7 illustrates steps constituting a welding management method according to some embodiments.

[0070] Referring to FIG. 7, the welding management method (700) may include steps (710) to (740). However, the present invention is not limited thereto, and some steps may be omitted or other general steps may be added, and the steps of the welding management method (700) may be executed in a different order than the illustrated order.

[0071] The welding management method (700) may be composed of steps that are processed in a time-series manner in the welding management device (130). Therefore, even if the content is omitted below, the content described above for the welding management device (130) may be equally applied to the welding management method (700).

[0072] Steps (710) to (740) of the welding management method (700) can be performed by the interface (131) and controller (132) of the welding management device (130).

[0073] In step (710), the welding management device (130) may perform a step of acquiring electrical data collected during a welding process for a battery cell.

[0074] In step (720), the welding management device (130) may perform a step of deriving a condition score representing the welding condition of the battery cell from electrical data based on a condition diagnosis model.

[0075] In step (730), the welding management device (130) may perform a step of comparing a first distribution of condition scores for target battery cells and a second distribution of condition scores of learning data used to train a condition diagnosis model to determine whether a data shift of the first distribution occurs.

[0076] In step (740), the welding management device (130) may perform a step of adjusting the condition diagnosis model based on the difference between the first distribution and the second distribution when a data shift occurs.

[0077] According to an embodiment, the welding management method (700) may be implemented in the form of a computer program stored on a computer-readable storage medium. That is, the computer program may include instructions for implementing the welding management method (700), and the instructions of the program may be stored on the computer-readable storage medium. The computer program may include a mobile application.

[0078] According to an embodiment, the computer-readable storage medium may include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute computer program instructions such as ROMs, RAMs, flash memories, and the like. The computer program instructions may include machine language codes generated by a compiler and high-level language codes that can be executed by a computer using an interpreter, etc.

[0079] The terms "include," "comprise," or "have" used herein, unless otherwise specifically stated, imply that the corresponding component may be included, and therefore should be interpreted to include other components rather than to exclude other components. All terms, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiments disclosed in this document pertain, unless otherwise defined. Commonly used terms, such as terms defined in dictionaries, should be interpreted to be consistent with their contextual meaning in the relevant art, and shall not be interpreted in an idealized or overly formal sense, unless explicitly defined in this document.

[0080] The above description is merely an illustrative description of the technical idea disclosed in this document, and those skilled in the art to which the embodiments disclosed in this document pertain may make various modifications and variations without departing from the essential characteristics of the embodiments disclosed in this document. Therefore, the embodiments disclosed in this document are not intended to limit the technical idea of ​​the embodiments disclosed in this document, but to explain it, and the scope of the technical idea disclosed in this document is not limited by these embodiments. The protection scope of the technical idea disclosed in this document should be interpreted by the claims below, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of rights of this document.

[0081] [Explanation of symbols]

[0082] 100: Welding Management System 110: Battery Cell

[0083] 120: Welding device 130: Welding management device

[0084] 131: Interface 132: Controller

Claims

1. An interface configured to acquire electrical data collected during a welding process for a battery cell; and A condition score representing the welding condition of the battery cell is derived from the electrical data based on the condition diagnosis model, Comparing a first distribution of condition scores for target battery cells and a second distribution of condition scores of learning data used to train the condition diagnosis model to determine whether a data shift of the first distribution occurs, A welding management device comprising a controller configured to adjust the condition diagnosis model based on a difference between the first distribution and the second distribution when the data shift occurs.

2. In paragraph 1, A welding management device, wherein the controller is configured to determine whether the data shift occurs based on a statistical hypothesis test for the first distribution and the second distribution.

3. In paragraph 2, A welding management device, wherein the controller is configured to determine that the data shift occurs when the significance probability (p-value), which indicates the probability that the null hypothesis that the first mean value of the first distribution and the second mean value of the second distribution are equal to each other, is established, is less than or equal to a threshold value.

4. In paragraph 1, A welding management device, wherein the controller is configured to adjust a threshold score used to determine a welding condition of the battery cell as defective when the data shift occurs.

5. In paragraph 4, A welding management device, wherein the controller is configured to adjust the threshold score based on a difference between a first representative score among the condition scores for the target battery cells and a second representative score among the condition scores of the learning data.

6. In paragraph 1, A welding management device, wherein the target battery cells include battery cells that are more than the most recently produced standard number among a plurality of battery cells mass-produced by the welding process.

7. In paragraph 1, The above data shift is caused by replacement of consumables of a welding device performing the above welding process, a welding management device.

8. A step of acquiring electrical data collected during a welding process for a battery cell; A step of deriving a condition score representing the welding condition of the battery cell from the electrical data based on a condition diagnosis model; A step of comparing a first distribution of condition scores for target battery cells and a second distribution of condition scores of learning data used to train the condition diagnosis model to determine whether a data shift of the first distribution occurs; and A welding management method, comprising a step of adjusting the condition diagnosis model based on the difference between the first distribution and the second distribution when the data shift occurs.

9. In paragraph 8, The step of determining whether the above data shift occurs is: A welding management method comprising a step of determining whether the data shift occurs based on a statistical hypothesis test for the first distribution and the second distribution.

10. In paragraph 9, The step of determining whether the above data shift occurs is: A welding management method, comprising a step of determining that the data shift has occurred when the significance probability (p-value) indicating the probability that the null hypothesis that the first mean value of the first distribution and the second mean value of the second distribution are equal to each other is established is less than or equal to a critical value.

11. In paragraph 8, The step of adjusting the above condition diagnosis model is: A welding management method comprising a step of adjusting a threshold score used to determine a welding condition of the battery cell as defective when the above data shift occurs.

12. In paragraph 11, The step of adjusting the above condition diagnosis model is: A welding management method comprising a step of adjusting the threshold score based on a difference between a first representative score among the condition scores for the target battery cells and a second representative score among the condition scores of the learning data.

13. In paragraph 8, A welding management method, wherein the target battery cells include battery cells that are more than the most recently produced standard number among a plurality of battery cells mass-produced by the welding process.

14. In paragraph 8, A welding management method wherein the above data shift occurs due to replacement of consumables of a welding device performing the above welding process.

15. A welding device configured to perform a welding process on a battery cell; and A welding management system comprising a welding management device configured to acquire electrical data collected during the welding process, derive a condition score representing a welding condition of the battery cell from the electrical data based on a condition diagnosis model, compare a first distribution of condition scores for target battery cells with a second distribution of condition scores of learning data used to train the condition diagnosis model to determine whether a data shift of the first distribution occurs, and adjust the condition diagnosis model based on a difference between the first distribution and the second distribution when the data shift occurs.

Images