Device for diagnosing battery and method therefor

Abstract

The present disclosure relates to a device for diagnosing a battery and a method therefor. The device for diagnosing a battery, according to one embodiment, can input, into a diagnosis model, a first impedance value measured while an alternating current signal of a first frequency is applied to a first battery cell to be diagnosed, and determine, on the basis of state data that is output from the diagnosis model, whether the first battery cell is defective. The diagnosis model can comprise: a first artificial neural network for predicting, on the basis of the first impedance value, a second impedance value while an alternating current signal of a second frequency is applied to the first battery cell; and a second artificial neural network for outputting, on the basis of the second impedance value, the state data indicating whether the first battery cell is defective.

Description

Battery diagnostic device and method thereof

[0001] The present disclosure relates to a battery diagnostic device and method thereof.

[0002] Secondary batteries, unlike non-rechargeable primary batteries, are rechargeable and dischargeable. Low-capacity secondary batteries are used in small, portable electronic devices such as smartphones, feature phones, laptops, digital cameras, and camcorders, while large-capacity secondary batteries are widely used as power sources and power storage devices for motors in hybrid and electric vehicles.

[0003] All-solid-state batteries are secondary batteries whose main components, including the electrolyte, are all solid. During the manufacturing process of all-solid-state batteries, the solid electrolyte layer is assembled into an integrated state with the positive and negative electrodes. Therefore, the short-circuit test method used before electrolyte injection in conventional liquid-electrolyte lithium secondary batteries is difficult to apply to all-solid-state batteries.

[0004] The above-described information disclosed in the background technology of this invention is only intended to improve understanding of the background of the present invention, and therefore may include information that does not constitute prior art.

[0005] The present invention provides a battery diagnostic device and method capable of diagnosing whether an all-solid-state battery is defective during the manufacturing process.

[0006] However, the technical problems to be solved by the present invention are not limited to the problems described above, and other problems not mentioned can be clearly understood by those skilled in the art from the description of the invention described below.

[0007] According to an embodiment for solving the above technical problem, a battery diagnosis device may include a storage device for storing a diagnosis model, a measuring device for measuring an impedance of a first battery cell to be diagnosed while an AC signal of a first frequency is applied to the first battery cell to obtain a first impedance value, and a control device for inputting the first impedance value into the diagnosis model and determining whether the first battery cell is defective based on status data output from the diagnosis model. The diagnosis model may include a first artificial neural network for predicting a second impedance value while an AC signal of a second frequency is applied to the first battery cell based on the first impedance value, and a second artificial neural network for outputting the status data indicating whether the first battery cell is defective based on the second impedance value.

[0008] The first frequency may be higher than the second frequency.

[0009] The first frequency may be a frequency of 1 Hz or more, and the second frequency may be a frequency of less than 1 Hz.

[0010] The battery diagnosis device may further include a diagnostic model construction device that constructs the diagnostic model. The diagnostic model construction device may train the first artificial neural network using third impedance values ​​measured while applying the AC signal of the first frequency to each of the plurality of second battery cells and fourth impedance values ​​measured while applying the AC signal of the second frequency to each of the plurality of second battery cells. The diagnostic model construction device may train the second artificial neural network using state data of each of the plurality of second battery cells and the fourth impedance values. The battery diagnosis device may construct the diagnostic model using the first artificial neural network and the second artificial neural network for which training has been completed.

[0011] The above measuring device can measure the impedance of the first battery cell by applying an AC signal of the first frequency while the first battery cell is pressurized.

[0012] The above first battery cell may be a Bi-Cell of an all-solid-state battery.

[0013] A battery diagnosis method according to one embodiment may include a step of obtaining a first impedance value by measuring the impedance of a first battery cell as a diagnosis target while an AC signal of a first frequency is applied to the first battery cell, a step of inputting the first impedance value into a diagnosis model, and a step of determining whether the first battery cell is defective based on status data output from the diagnosis model. The diagnosis model may include a first artificial neural network that predicts a second impedance value while an AC signal of a second frequency is applied to the first battery cell based on the first impedance value, and a second artificial neural network that outputs the status data indicating whether the first battery cell is defective based on the second impedance value.

[0014] In the above battery diagnosis method, the first frequency may be higher than the second frequency.

[0015] In the above battery diagnosis method, the first frequency may be a frequency of 1 Hz or more, and the second frequency may be a frequency of less than 1 Hz.

[0016] The battery diagnosis method may further include a step of training the first artificial neural network using third impedance values ​​measured while applying an AC signal of the first frequency to each of the plurality of second battery cells and fourth impedance values ​​measured while applying an AC signal of the second frequency to each of the plurality of second battery cells, a step of training the second artificial neural network using state data of each of the plurality of second battery cells and the fourth impedance values, and a step of constructing the diagnosis model using the first artificial neural network and the second artificial neural network for which training has been completed.

[0017] The obtaining step may include a step of applying an AC signal of the first frequency while pressurizing the first battery cell and measuring the impedance of the first battery cell.

[0018] In the above battery diagnosis method, the first battery cell may be a Bi-Cell of an all-solid-state battery.

[0019] According to the present invention, defect detection is possible even in the semi-finished state of an all-solid-state battery. Therefore, defective batteries can be sorted out before shipment, improving the yield of the final product.

[0020] However, the effects that can be obtained through the present invention are not limited to the effects described above, and other technical effects that are not mentioned can be clearly understood by those skilled in the art from the description of the invention described below.

[0021] The following drawings attached to this specification illustrate preferred embodiments of the present invention, and together with the detailed description of the invention described below, serve to further understand the technical idea of ​​the present invention, and therefore, the present invention should not be interpreted as being limited to matters described in such drawings.

[0022] Figure 1 schematically illustrates a battery diagnostic device according to one embodiment.

[0023] Figure 2 schematically illustrates a diagnostic model of a battery diagnostic device according to one embodiment.

[0024] Figure 3 schematically illustrates a method for constructing a diagnostic model according to one embodiment.

[0025] Figure 4 schematically illustrates a battery diagnostic method according to one embodiment.

[0026] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. Prior to this, terms or words used in this specification and claims should not be interpreted as limited to their typical or dictionary meanings, but should be interpreted with meanings and concepts that conform to the technical idea of ​​the present invention based on the principle that the inventor can appropriately define the concept of the term in order to explain his own invention in the best way. Therefore, it should be understood that the embodiments described in this specification and the configurations illustrated in the drawings are only some of the most preferred embodiments of the present invention and do not represent all of the technical idea of ​​the present invention, and various equivalents and modifications may exist at the time of filing this application. In addition, when used in this specification, "comprise" and "include" and / or "comprising" specify the presence of mentioned shapes, numbers, steps, operations, elements, components and / or groups thereof, and do not exclude the presence or addition of one or more other shapes, numbers, operations, elements, components and / or groups. Additionally, when describing embodiments of the present invention, “may” and “may be” may include “one or more embodiments of the present invention.”

[0027] Additionally, to facilitate understanding of the invention, the attached drawings are not drawn to scale and some components may be exaggerated in size. Furthermore, identical components may be assigned the same reference numbers in different embodiments.

[0028] The statement that two compared objects are "identical" means "substantially identical." Therefore, "substantially identical" may include deviations considered low in the art, such as deviations of less than 5%. Furthermore, uniformity of a parameter over a given region may imply uniformity on average.

[0029] Although terms like "first" and "second" are used to describe various components, these components are not limited by these terms. These terms are used merely to distinguish one component from another, and unless otherwise specified, a "first" component may also be a "second" component.

[0030] Throughout the specification, unless otherwise specifically stated, each element may be singular or plural.

[0031] Any configuration being placed "on (or under)" or "above (or below)" a component may mean not only that any configuration is placed in contact with the upper surface (or lower surface) of said component, but also that other configurations may intervene between said component and any configuration placed on (or below) said component.

[0032] Additionally, when it is described that a component is "connected," "coupled," or "connected" to another component, it should be understood that the components may be directly connected or connected to each other, but that other components may also be "interposed" between the components, or that each component may be "connected," "coupled," or "connected" through another component. Furthermore, when it is said that a part is electrically coupled to another part, this includes not only cases where they are directly connected, but also cases where they are connected with another element in between.

[0033] When reference is made throughout the specification to "A and / or B," this means A, B, or A and B, unless otherwise stated. In other words, "and / or" includes all or any combination of the listed items. When reference is made to "C through D," this means C or more and D or less, unless otherwise stated.

[0034] Figure 1 schematically illustrates a battery diagnostic device according to one embodiment.

[0035] Referring to FIG. 1, a battery diagnostic device (1) may include a diagnostic model building device (10), a diagnostic device (20), and an output device (30).

[0036] A diagnostic model building device (10) can build a diagnostic model for diagnosing whether an all-solid-state battery cell is defective. Here, the all-solid-state battery cell is a bi-cell of the all-solid-state battery, meaning a unit battery having a structure of anode / solid electrolyte layer / cathode / solid electrolyte layer / anode. The all-solid-state battery may be a laminated battery in which the bi-cell structure is repeated.

[0037] The diagnostic model building device (10) may include a storage device (11), a measuring device (12), and a control device (13).

[0038] The storage device (11) can store various data and information processed in the diagnostic model construction device (10). For example, the storage device (11) can store a diagnostic model (111), learning data used for learning the diagnostic model (111), etc.

[0039] The storage device (11) can also store a program for the operation of the control device (13) described later.

[0040] The measuring device (12) can measure the impedance of the all-solid-state battery cell while applying an AC signal between the cathode and anode of the all-solid-state battery cell. At this time, the all-solid-state battery cell is applied with a pressure jig (not shown) of 900 kgf / cm. 2 Or, it may be pressurized at a pressure of 2 MPa or less. That is, the measuring device (12) can apply an AC signal to the pressurized all-solid-state battery cell and measure the impedance of the all-solid-state battery cell in this state.

[0041] The control device (13) can build a machine learning-based diagnostic model (111) using the impedance value of each solid-state battery cell measured through the measuring device (12) and the status data of each solid-state battery cell.

[0042] Figure 2 schematically illustrates a diagnostic model of a battery diagnostic device according to one embodiment.

[0043] Referring to FIG. 2, the diagnostic model (111) may include multiple artificial neural networks (111-1, 111-2).

[0044] The first artificial neural network (111-1) can predict and output an impedance value of an all-solid-state battery cell measured when an AC signal of a first frequency is applied to the all-solid-state battery cell, from the impedance value of the all-solid-state battery cell measured when an AC signal of a first frequency is applied to the all-solid-state battery cell. The first frequency may be a higher frequency than the second frequency. The first frequency may be a frequency of 1 Hz or higher (for example, 1 Hz), and the second frequency may be a frequency of less than 1 Hz (for example, 0.1 Hz).

[0045] The second artificial neural network (111-2) can receive the prediction result of the first artificial neural network (111-1), and based on this, predict the state (normal or defective) of the corresponding solid-state battery cell and output the predicted state data.

[0046] The lower the frequency of the AC signal applied to an all-solid-state battery cell, the easier it is to distinguish between faulty and healthy cells using impedance measurements. However, the time required for impedance measurements increases exponentially as the frequency of the AC signal applied to the all-solid-state battery cell decreases.

[0047] Accordingly, in order to improve the discrimination ability of a defective solid-state battery cell while shortening the measurement time, the diagnostic model construction device (10) can construct a diagnostic model (111) including a first artificial neural network (111-1) for predicting an impedance value when an AC signal of a lower frequency than that applied during actual measurement is applied, and a second artificial neural network (111-2) for predicting the state of the solid-state battery cell using the impedance value predicted by the first artificial neural network (111-1).

[0048] The diagnostic model (111) constructed in this way can be used to diagnose defective conditions caused by internal short circuits, pin holes, deterioration of electrolyte or anode, insertion of internal foreign substances, etc. of an all-solid-state battery cell.

[0049] Again, referring to Figure 1, the control device (13) can control the overall operation of the diagnostic model building device (10).

[0050] The control device (13) can collect learning data for training the diagnostic model (111).

[0051] The control device (13) can control the measurement device (12) to collect impedance measurement values ​​measured while applying an AC signal of a first frequency to each of the plurality of all-solid-state battery cells. The control device (13) can also control the measurement device (12) to collect impedance measurement values ​​measured while applying an AC signal of a second frequency to each of the plurality of all-solid-state battery cells. The control device (13) can train the first artificial neural network (111-1) using the impedance measurement values ​​at the first frequency and the impedance measurement values ​​at the second frequency collected for the plurality of all-solid-state battery cells.

[0052] The control device (13) can also collect status data (indicating normal or defective) of each of the plurality of all-solid-state battery cells. The control device (13) can receive status data of each all-solid-state battery cell from a worker through an input device (not shown), or can receive status data of each all-solid-state battery cell from a worker terminal (not shown) through a communication device (not shown). The control device (13) can train a second artificial neural network (111-2) using the impedance measurement values ​​and status data collected for the plurality of all-solid-state battery cells at a second frequency.

[0053] When a diagnostic model (111) is constructed through the aforementioned learning, the control device (13) can store the constructed diagnostic model (111) in the storage device (11).

[0054] The diagnostic device (20) can diagnose the state of the solid-state battery cell, which is a diagnosis target, using the diagnostic model (111) constructed by the diagnostic model construction device (10).

[0055] The diagnostic device (20) may include a storage device (21), a measuring device (22), and a control device (23).

[0056] The storage device (21) can store various data and information processed in the diagnostic device (20). For example, the storage device (21) can store a diagnostic model (211) used for diagnosing the status of an all-solid-state battery cell. The diagnostic model (211) stored in the storage device (21) may be the same model as the diagnostic model (111) constructed by the aforementioned diagnostic model construction device (10).

[0057] The storage device (21) may also store a program for the operation of the control device (23) described later.

[0058] The measuring device (22) can measure the impedance of the solid-state battery cell while applying an AC signal of the first frequency between the cathode and the anode of the solid-state battery cell to be diagnosed. At this time, the solid-state battery cell is applied with a pressure jig (not shown) of 900 kgf / cm. 2 Or, it may be pressurized at a pressure of 2 MPa or less. That is, the measuring device (22) can apply an AC signal to the pressurized all-solid-state battery cell and measure the impedance of the all-solid-state battery cell in this state.

[0059] The control device (23) can control the overall operation of the diagnostic device (20).

[0060] The control device (23) can retrieve the diagnostic model (111) constructed by the diagnostic model construction device (10) in various ways and store it in the storage device (21). For example, the control device (23) can receive the diagnostic model (111) from the diagnostic model construction device (10) via a communication device (not shown). In addition, for example, the control device (23) can read the diagnostic model stored in the storage medium by the diagnostic model construction device (10) from the storage medium.

[0061] The control device (23) can receive an impedance measurement value measured while applying an AC signal of a first frequency to the solid-state battery cell to be diagnosed from the measurement device (22). Then, the control device (23) inputs the received impedance measurement value into the diagnosis model (211), and in response thereto, can diagnose the status (bad or normal) of the solid-state battery cell to be diagnosed based on status data output from the diagnosis model (211).

[0062] When an impedance measurement value at a first frequency is input, the diagnostic model (211) inputs it into a first artificial neural network (see drawing symbol 111-1 in FIG. 2), and the first artificial neural network can predict and output an impedance value at a second frequency in response to the input impedance measurement value at the first frequency. Then, the diagnostic model (211) inputs the impedance value at the second frequency predicted by the first artificial neural network into a second artificial neural network (see drawing symbol 111-2 in FIG. 2), and the second artificial neural network can output status data of an all-solid-state battery cell that is a diagnosis target in response to the input impedance value at the second frequency.

[0063] When the control device (23) completes the diagnosis of the state of the solid-state battery cell to be diagnosed using the diagnosis model (211), the diagnosis result can be output through an output device (30) such as a display.

[0064] The control unit (13) of the diagnostic model building device (10) and the control unit (23) of the diagnostic device (20) may each include at least one processor. The processor may refer to a data processing device having a physically structured circuit to perform a function expressed by a code or command included in a program, such as a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an application-specific integrated circuit (ASIC), or a field programmable gate array (FPGA).

[0065] Meanwhile, in FIG. 1, an example is illustrated in which the diagnostic model building device (10) and the diagnostic device (20) are implemented as separate devices, but the diagnostic model building device (10) and the diagnostic device (20) may be integrated into a single device. In this case, the functions performed by the control device (13) of the diagnostic model building device (10) may be performed by the control device (23) of the diagnostic device (20).

[0066] Hereinafter, a battery diagnosis method of a battery diagnosis device (1) according to one embodiment will be described with reference to FIGS. 3 and 4.

[0067] Fig. 3 schematically illustrates a method for constructing a diagnostic model of a battery diagnostic device according to one embodiment. The method of Fig. 3 can be performed by the diagnostic model constructing device (10) described with reference to Fig. 1.

[0068] The diagnostic model building device (10) can collect learning data for training the diagnostic model (111) (S11).

[0069] In step S11, the diagnostic model building device (10) can control the measuring device (12) to apply an AC signal of a first frequency and an AC signal of a second frequency to each of a plurality of all-solid-state battery cells, respectively, and measure the impedance value of each all-solid-state battery cell while the AC signal is applied. When the AC signal is applied, each all-solid-state battery cell may be pressurized to a predetermined pressure. The diagnostic model building device (10) can collect the impedance measurement value at the first frequency and the impedance measurement value at the second frequency measured by the measuring device (12) as learning data for training the first artificial neural network (111-1).

[0070] In step S11, the diagnostic model building device (10) can also obtain status data (indicating normal or defective) of each of the plurality of all-solid-state battery cells. The diagnostic model building device (10) can collect the impedance measurement values ​​at the second frequency measured for the plurality of all-solid-state battery cells by the measuring device (12) and the status data of each all-solid-state battery cell as learning data for training the second artificial neural network (111-2).

[0071] The diagnostic model building device (10) can train the first artificial neural network (111-1) and the second artificial neural network (111-2) using the learning data collected through step S11. The diagnostic model building device (10) can train the first artificial neural network (111-1) using the impedance measurement value measured in a state where an AC signal of a first frequency is applied to the all-solid-state battery cell and the impedance measurement value measured in a state where an AC signal of a second frequency is applied to the all-solid-state battery cell (S12). The diagnostic model building device (10) can train the second artificial neural network (111-2) using the impedance measurement value measured in a state where an AC signal of a second frequency is applied to the all-solid-state battery cell and the state data of the all-solid-state battery cell (S13).

[0072] When the learning of the first and second artificial neural networks (111-1, 111-2) is completed through steps S12 and S13, the diagnostic model building device (10) can build a diagnostic model (111) using the first and second artificial neural networks (111-1, 111-2) (S14).

[0073] In step S14, the diagnostic model building device (10) can build a diagnostic model (111) including a first artificial neural network (111-1) that receives an impedance measurement value measured at a first frequency and predicts and outputs an impedance value at a second frequency in response thereto, and a second artificial neural network (111-2) that receives an impedance value predicted by the first artificial neural network (111-1) and predicts and outputs a state of an all-solid-state battery cell in response thereto.

[0074] Fig. 4 schematically illustrates a diagnostic method of a battery diagnostic device according to one embodiment. The method of Fig. 4 can be performed by the diagnostic device (20) described with reference to Fig. 1.

[0075] The diagnostic device (20) can obtain an impedance measurement value measured by applying an AC signal of a first frequency to the solid-state battery cell to be diagnosed from the measuring device (22) (S21). In step S21, when the AC signal is applied, each solid-state battery cell may be pressurized to a predetermined pressure.

[0076] Then, the diagnostic device (20) can input the impedance measurement value obtained through step S21 into the diagnostic model (211) (S22). Here, the diagnostic model (211) may be the same model as the diagnostic model (111) constructed by the diagnostic model construction device (10).

[0077] When an impedance measurement value measured while an AC signal of a first frequency is applied to an all-solid-state battery cell is input, the diagnostic model (211) can predict and output the state (bad or normal) of the all-solid-state battery cell as a diagnosis target in response to this. When the impedance measurement value at the first frequency is input, the diagnostic model (211) inputs this to a first artificial neural network (see reference numeral 111-1 in FIG. 2), and the first artificial neural network can predict and output an impedance value at a second frequency in response to the input impedance measurement value at the first frequency. Then, the diagnostic model (211) inputs the impedance value at the second frequency predicted by the first artificial neural network to a second artificial neural network (see reference numeral 111-2 in FIG. 2), and the second artificial neural network can output state data of the all-solid-state battery cell as a diagnosis target in response to the input impedance value at the second frequency.

[0078] As described above, when the diagnostic model (211) predicts and outputs the state of the solid-state battery cell that is the target of diagnosis, the diagnostic device (20) can finally determine the state of the solid-state battery cell that is the target of diagnosis based on the output value of the diagnostic model (211) (S23). In addition, the diagnostic device (20) can output state data indicating the state of the solid-state battery cell that is the target of diagnosis through the output device (30) (S24).

[0079] Although the present invention has been described above with reference to 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 idea of ​​the present invention and the equivalent scope of the patent claims to be described below by a person having ordinary skill in the art to which the present invention pertains.

Claims

1. A storage device that stores the diagnostic model; A measuring device that obtains a first impedance value by measuring the impedance of a first battery cell while applying an AC signal of a first frequency to the first battery cell as a diagnosis target, and A control device that inputs the first impedance value into the diagnostic model and determines whether the first battery cell is defective based on status data output from the diagnostic model, The above diagnostic model is, A first artificial neural network that predicts a second impedance value when an AC signal of a second frequency is applied to the first battery cell based on the first impedance value, and A battery diagnostic device comprising a second artificial neural network that outputs status data indicating whether the first battery cell is defective based on the second impedance value.

2. In paragraph 1, A battery diagnostic device wherein the first frequency is higher than the second frequency.

3. In paragraph 2, A battery diagnostic device, wherein the first frequency is a frequency of 1 Hz or more, and the second frequency is a frequency of less than 1 Hz.

4. In paragraph 1, Further comprising a diagnostic model construction device for constructing the above diagnostic model, The above diagnostic model building device is, The first artificial neural network is trained using the third impedance values ​​measured while applying the AC signal of the first frequency to each of the plurality of second battery cells and the fourth impedance values ​​measured while applying the AC signal of the second frequency to each of the plurality of second battery cells. The second artificial neural network is trained using the state data of each of the plurality of second battery cells and the fourth impedance values, A battery diagnostic device that builds the diagnostic model using the first artificial neural network and the second artificial neural network for which learning has been completed.

5. In paragraph 1, The above measuring device is a battery diagnostic device that applies an AC signal of the first frequency while the first battery cell is pressurized and measures the impedance of the first battery cell.

6. In paragraph 1, A battery diagnostic device, wherein the first battery cell is a bi-cell of an all-solid-state battery.

7. As a battery diagnosis method of a battery diagnosis device, A step of obtaining a first impedance value by measuring the impedance of a first battery cell while applying an AC signal of a first frequency to the first battery cell as a diagnosis target, A step of inputting the above first impedance value into a diagnostic model, and A step of determining whether the first battery cell is defective based on status data output from the diagnostic model, The above diagnostic model is, A first artificial neural network that predicts a second impedance value when an AC signal of a second frequency is applied to the first battery cell based on the first impedance value, and A battery diagnosis method, comprising a second artificial neural network that outputs the status data indicating whether the first battery cell is defective based on the second impedance value.

8. In paragraph 7, A battery diagnostic method wherein the first frequency is higher than the second frequency.

9. In paragraph 8, A battery diagnosis method, wherein the first frequency is a frequency of 1 Hz or more, and the second frequency is a frequency of less than 1 Hz.

10. In paragraph 7, A step of training the first artificial neural network using third impedance values ​​measured while applying an AC signal of the first frequency to each of the plurality of second battery cells and fourth impedance values ​​measured while applying an AC signal of the second frequency to each of the plurality of second battery cells. A step of training the second artificial neural network using the state data of each of the plurality of second battery cells and the fourth impedance values, and A battery diagnosis method further comprising a step of constructing the diagnostic model using the first artificial neural network and the second artificial neural network for which learning has been completed.

11. In paragraph 7, The above acquisition steps are: A battery diagnosis method comprising the step of applying an AC signal of the first frequency while pressurizing the first battery cell and measuring the impedance of the first battery cell.

12. In paragraph 7, A battery diagnosis method, wherein the first battery cell is a bi-cell of an all-solid-state battery.

Images