Determination method and determination device

Abstract

To provide a determination method and a determination device capable of non-destructively determining whether an active material of a secondary battery is reused.SOLUTION: A determination method comprises: acquiring a detection signal obtained by non-destructively detecting an internal state of a secondary battery to be determined, as a target detection signal; inputting the target detection signal to a classification model; and on the basis of the output result output from the classification model, determining whether the active material of the secondary battery to be determined can be reused. The classification model is a model that classifies the active material of the secondary battery into a category as to whether it can be reused, on the basis of the detection signal that non-destructively detects the internal state of the secondary battery.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a determination method and a determination device.

Background Art

[0002] There is known a technique of detecting an AE signal generated inside a lithium-ion secondary battery from an AE sensor and outputting a signal indicating deterioration of the lithium-ion secondary battery when the number of AE events generated within one charge-discharge cycle is equal to or greater than a threshold value (Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the technique described in Patent Document 1 is a technique for outputting a signal indicating deterioration of a secondary battery, and does not indicate whether the active material of the deteriorated secondary battery can be reused. Therefore, there is a problem that in order to determine whether the active material of the secondary battery can be reused, the secondary battery must be disassembled to check the internal state.

[0005] The problem to be solved by the present invention is to provide a determination method and a determination device that can non-destructively determine whether the active material of a secondary battery can be reused.

Means for Solving the Problems

[0006] The present invention acquires a detection signal obtained by non-destructively detecting the internal state of a secondary battery to be determined as a target detection signal, inputs the target detection signal into a classification model, and determines whether the active material of the secondary battery to be determined can be reused based on the output result output from the classification model. The classification model solves the above problems by being a model that classifies the active material of the secondary battery into categories of whether it can be reused or not based on a detection signal obtained by non-destructively detecting the internal state of the secondary battery.

Effect of the Invention

[0007] According to the present invention, it is possible to non-destructively determine whether the active material of a secondary battery can be reused.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Modes for Carrying Out the Invention

[0009] Hereinafter, a determination method and a determination device according to the present invention will be described based on the drawings. The determination method and the determination device according to the present invention non-destructively determine whether the active material of a used secondary battery can be reused.

[0010] <<First Embodiment>> With reference to FIG. 1, a determination method and a determination apparatus according to a first embodiment of the present invention will be described. FIG. 1 is a block diagram showing an example of the configuration of a determination system including a determination apparatus according to the first embodiment of the present invention. The determination system 1 includes a sensor 20, an input device 30, an output device 40, and a determination apparatus 100. Each device is connected so as to be able to exchange information with each other by wire or wirelessly. The determination system 1 is a system for determining whether the active material of a used secondary battery can be reused. In the determination system 1, the determination apparatus 100 acquires, as a target detection signal, a detection signal obtained by non-destructively detecting the internal state of the secondary battery to be determined from the sensor 20, and inputs the acquired target detection signal into a classification model. Then, according to the output result of the classification model, it is determined non-destructively whether the active material of the secondary battery to be determined can be reused. The determination apparatus 100 outputs the determination result to the output device 40. When the output device 40 receives the determination result from the determination apparatus 100, it outputs the determination result to the user. The user checks the determination result output from the output device 40 and performs processing according to the determination result. The user includes, for example, a recycling company that collects and reuses the active material, a used value appraiser, and the like. The secondary battery to be determined is a secondary battery for which it is determined whether its active material can be reused, and is a used secondary battery that has been used for some purpose. For example, the secondary battery to be determined is a secondary battery used in an electric vehicle. In addition, in the present embodiment, the secondary battery to be determined may be a secondary battery used for purposes other than an electric vehicle.

[0011] Further, the input device 30 and the output device 40 may be devices different from the determination device 100 as in the example of FIG. 1, or may be devices provided in the determination device 100. Further, the determination device 100 is not limited to being connected to the sensor 20, and the determination device 100 may be configured not to be connected to the sensor 20. In this case, for example, the user reads the data of the target detection signal from the sensor 20 and stores it in an external memory, and by connecting the external memory to the determination device 100, the determination device 100 acquires the data of the target detection signal from the external memory. Further, when the secondary battery is used in an electric vehicle, the determination device 100 may be a device mounted on a vehicle that is an electric vehicle, or may be an off-vehicle device such as a cloud server.

[0012] The sensor 20 is a sensor attached to the secondary battery to be determined, and non-destructively detects the internal state of the secondary battery to be determined. The internal state of the secondary battery is, for example, the state of the active material constituting the secondary battery, and includes destructive phenomena such as cracks and peeling occurring inside the secondary battery. For example, the sensor 20 non-destructively detects elastic waves (acoustic emissions) generated by destructive phenomena inside the secondary battery. The detected elastic waves are elastic waves in the range of several tens of kHz to several MHz. A detection signal (AE signal) indicating the elastic waves detected by the sensor 20 is output to the determination device 100. In the present embodiment, the sensor 20 outputs the detection signal detected by the secondary battery to be determined as the target detection signal. In the following description, the sensor 20 will be described as an AE sensor that detects acoustic emission (AE), but the sensor 20 is not limited to this as long as it is a sensor that non-destructively detects deterioration inside the secondary battery, for example, destructive phenomena such as cracks and peeling occurring in the active material.

[0013] In addition, in the present embodiment, a detection signal is used to generate a classification model described later. In this case, a sensor similar to the sensor 20 is attached to the secondary battery for data acquisition, and the internal state of the secondary battery for data acquisition is non-destructively detected by the sensor. The secondary battery for data acquisition is a secondary battery different from the secondary battery to be determined, and is used to acquire data in advance and generate a classification model before determining whether the active material of the secondary battery to be determined can be reused. While the secondary battery for data acquisition is being used, a detection signal (pre-detection signal) obtained by non-destructively detecting the internal state of the secondary battery for data acquisition is acquired. In the following description, the data of the detection signal (target detection signal) detected by the sensor 20 will be described. However, since the data of the detection signal (pre-detection signal) used for generating the classification model is the same, the description of the pre-detection signal will be omitted, and the description of the target detection signal detected by the sensor 20 will be appropriately referred to. That is, for each feature of the "pre-detection signal" described in the claims, the following description of the target detection signal will be appropriately referred to.

[0014] The sensor 20 detects elastic waves generated inside the secondary battery to be determined during use from the start to the end of use of the secondary battery to be determined. For example, the sensor 20 detects elastic waves at a certain period (for example, at 1-second intervals). The sensor 20 converts the detected elastic waves to obtain an AE signal. The data of the AE signal includes the frequency spectrum representing the AE signal. The frequency spectrum is obtained by decomposing the AE signal detected at each time point for each frequency and represents the intensity for each frequency. The horizontal axis represents the frequency, and the vertical axis represents the intensity (amplitude). The frequency spectrum is not limited to data including the intensity in all frequency bands within the detection range, and may be data limited to the intensity in the frequency band of 700 kHz or higher.

[0015] In addition, the data of the AE signals includes the data of a plurality of AE signals acquired during the use of the secondary battery. These AE signal data are, for example, all the AE signal data acquired at regular intervals during the use of the secondary battery, and are time-series continuous time-series data from the start to the end of the use of the secondary battery. The data of each of the plurality of acquired AE signals includes the frequency spectrum representing each AE signal.

[0016] In addition, the AE signal data may include the number of times the intensity of the AE signal becomes equal to or greater than a predetermined intensity threshold during the use from the start to the end of the use of the secondary battery. For example, for the AE signal at each time point, it is determined whether the intensity in any frequency band of the AE signal is equal to or greater than a predetermined intensity threshold. Then, among all the AE signals acquired during the use, the number of times it is determined that the intensity of the AE signal is equal to or greater than the predetermined intensity threshold is integrated. Also, it is not limited to determining whether the intensity in any frequency band of the AE signal is equal to or greater than a predetermined intensity threshold, and it may be determined whether the intensity in the frequency band of 700 kHz or higher of the AE signal is equal to or greater than a predetermined intensity threshold. In addition, the AE signal data may include data that stores, in time-series order, the frequency band at which the intensity of the AE signal reaches the maximum value at each time point during the period from the start to the end of the use of the secondary battery. Note that in the determination of whether it can be reused, each threshold can be arbitrarily determined by the user based on the intensity of the AE signal and the acquired data such as the number and size of cracks in the active material.

[0017] For example, FIG. 2 is a diagram showing an example of the frequency spectrum of the AE signal in the present embodiment. As shown in FIG. 2, the intensity of the AE signal detected at a certain point during the use of the secondary battery is recorded for each frequency band. In FIG. 2, the horizontal axis represents the frequency and the vertical axis represents the intensity (amplitude). In the example of FIG. 2, the intensity of the AE signal is equal to or greater than the intensity threshold in the frequency band of 700 kHz or higher. In the present embodiment, for each of all the AE signals acquired during the use, as shown in FIG. 2, it is determined whether the intensity of the AE signal is equal to or greater than the intensity threshold in any frequency band, and the number of times it is determined that the intensity of the AE signal is equal to or greater than the intensity threshold is recorded.

[0018] The input device 30 is a device that receives data input by a user. The data to be input is data used for generating a classification model. For example, a pre-detection signal and reuse-eligibility data are input to the input device 30. These data may be input separately, or a data set including these data may be input. The pre-detection signal is a detection signal obtained by non-destructively detecting the internal state of a secondary battery for data acquisition, and is, for example, an AE signal (hereinafter also referred to as a pre-AE signal) generated inside the secondary battery for data acquisition. The data of the pre-AE signal includes a frequency spectrum and / or the number of times the intensity of the AE signal has reached or exceeded an intensity threshold. The reuse-eligibility data is data indicating whether the active material of the secondary battery for data acquisition can be reused. When the generation of the classification model is executed by machine learning, the data set is teacher data in which the pre-AE signal and the reuse-eligibility data are associated with each other.

[0019] The output device 40 is a device for outputting information and providing information to the user. The output device 40 may be an in-vehicle device or a device mounted on a portable terminal such as a tablet terminal. For example, the output device 40 is a device for displaying image information and is composed of a liquid crystal display, a projector, etc. Further, the output device 40 may be a device for outputting audio information. For example, the output device 40 is composed of a speaker. In the present embodiment, when there is an input of a determination result indicating whether the active material of the secondary battery to be determined can be reused from the determination device 100, the output device 40 outputs image information and / or audio information including the determination result to the user. The user checks the determination result and performs processing corresponding to whether the active material of the secondary battery to be determined can be reused. For example, when the active material of the secondary battery can be reused, the user disassembles the secondary battery, peels it off from the electrode, removes the binder, auxiliary agent, etc., and then uses it again as an active material. When the active material of the secondary battery cannot be reused, the user disassembles the secondary battery and then bakes it to generate a black mass. From the black mass, a compound with a rare metal such as Ni or Co is generated by chemical solvent / extraction treatment and used as a synthesis raw material for the active material.

[0020] Next, the determination device 100 according to the present embodiment will be described. The determination device 100 determines whether the active material of the secondary battery to be determined can be reused. For example, when the secondary battery is used for applications such as electric vehicles, the used secondary battery becomes the secondary battery to be determined. The determination device 100 acquires an AE signal obtained by non-destructively detecting an elastic wave generated inside the secondary battery to be determined as a target AE signal, and inputs the target AE signal into a classification model. The determination device 100 determines whether the active material of the secondary battery to be determined can be reused based on the output result output from the classification model.

[0021] The determination device 100 includes a controller 10 and a storage device 50. The controller 10 determines whether the active material of a used secondary battery can be reused. The controller 10 is a computer including a ROM storing a program for determining whether the active material of a used secondary battery can be reused, a CPU functioning as an operation circuit that functions as the determination device 100 by executing the program stored in this ROM, and a RAM functioning as an accessible storage device.

[0022] The controller 10 includes, as functional blocks, an acquisition unit 110, a determination unit 111, and a model generation unit 112. The controller 10 executes each function through cooperation between software for realizing each of the above functions or executing each process and the above-described hardware. In the present embodiment, after dividing the functions of the controller 10 into three blocks, the functions of each functional block will be described. However, the functions of the controller 10 do not necessarily have to be divided into three blocks, and may be divided into two or fewer functional blocks, or four or more functional blocks. Also, although one controller 10 is illustrated in FIG. 1, the controller 10 may be a plurality of controllers. These controllers are connected via a CAN communication network. For example, although the controller 10 is assumed to have three functions, the present invention is not limited to this, and the three functions may be provided in separate controllers. The model generation unit 112 may be provided in a controller different from the controller including the acquisition unit 110 and the determination unit 111.

[0023] The acquisition unit 110 executes an acquisition process for acquiring various types of information. The acquisition unit 110 acquires necessary information in two scenarios. The two scenarios are a determination scenario for determining whether the active material of the secondary battery to be determined can be reused, and a generation scenario for generating a classification model. First, the acquisition process in the determination scenario will be described. In the determination scenario, the acquisition unit 110 acquires, as the target AE signal, the AE signal generated inside the secondary battery to be determined from the sensor 20 attached to the secondary battery to be determined. The acquired target AE signal includes all AE signals detected by the sensor 20 during the use of the secondary battery to be determined. Further, the acquired target AE signal may be all AE signals in the frequency band of 700 kHz or higher detected by the sensor 20 during the use of the secondary battery to be determined. Each target AE signal may include a frequency spectrum in all frequency bands, or may include a frequency spectrum in the frequency band of 700 kHz or higher. Further, the target AE signal may include the number of times the intensity of the target AE signal becomes equal to or higher than a predetermined intensity threshold during the use of the secondary battery to be determined. Further, the target AE signal may include the number of times the intensity of the target AE signal becomes equal to or higher than a predetermined intensity threshold in the frequency band of 700 kHz or higher during the use of the secondary battery to be determined. The acquired target AE signal is used in the determination process by the determination unit 111.

[0024] Next, the acquisition process in the generation scenario will be described. In the generation scenario, the acquisition unit 110 acquires a pre-detection signal and reusability data. The pre-detection signal is a detection signal obtained by non-destructively detecting the internal state of the secondary battery for data acquisition, for example, an AE signal generated inside the secondary battery for data acquisition. The reusability data is data indicating whether the active material of the secondary battery for data acquisition can be reused. The pre-AE signal and the reusability data are acquired from the input device 30. Note that the pre-AE signal and the reusability data are not limited to being acquired separately, and a data set including the pre-AE signal and the reusability data may be acquired. For example, the data set is teacher data in which the pre-AE signal and the reusability data are associated. When the pre-AE signal and the reusability data are acquired separately, the acquisition unit 110 combines the acquired pre-AE signal and the reusability data to generate a data set. Note that the acquisition unit 110 is not limited to acquiring these pieces of information via the input device 30, and may acquire these pieces of information via a communication device (not shown).

[0025] The determination unit 111 executes a determination process for determining whether the active material of the secondary battery to be determined can be reused. In the determination process, the determination unit 111 inputs the target AE signal acquired by the acquisition unit 110 into the classification model, and determines whether the active material of the secondary battery to be determined can be reused based on the output result output from the classification model. The classification model outputs output data including a classification as to whether the active material can be reused as output data corresponding to the input data when the target AE signal is input as the input data.

[0026] The classification model is a model that classifies the active material of the secondary battery to be determined into a classification as to whether it can be reused based on the target AE signal. The classification model is trained so that reusability data is associated with the pre-AE signal. Thereby, when the target AE signal is input, the classification model outputs output data including a classification as to whether the input target AE signal can be reused. The classification model will be described later.

[0027] In this embodiment, when the category output from the classification model is a category indicating that it can be reused, the determination unit 111 determines that the active material of the secondary battery to be determined can be reused. Further, when the category output from the classification model is a category indicating that it cannot be reused, the determination unit 111 determines that the active material of the secondary battery to be determined cannot be reused.

[0028] The determination unit 111 notifies the user according to the determination result. For example, when the determination unit 111 determines that the active material of the secondary battery to be determined can be reused, it outputs a control instruction to the output device 40 to output image information and / or audio information indicating that the active material of the secondary battery to be determined can be reused to the user. When the determination unit 111 determines that the active material of the secondary battery to be determined cannot be reused, it outputs a control instruction to the output device 40 to output image information and / or audio information indicating that the secondary battery to be determined cannot be reused to the user.

[0029] The model generation unit 112 executes a model generation process for generating a classification model. In the model generation process, the model generation unit 112 generates a classification model by statistically processing a data set including pre-AE signals and reuse-ability data. In the following description, a method using machine learning will be described as an example of a method for statistically processing a data set. Examples of machine learning include, for example, the k-nearest neighbor method, decision tree, logistic regression, support vector machine, naive Bayes classification, random forest, and neural network. Note that in this embodiment, the method for generating the classification model is not limited to machine learning. For example, the model generation unit 112 may use methods such as t-test, cross-tabulation and chi-square test, statistical threshold and rule-based classification.

[0030] The pre-AE signal is a detection signal obtained by non-destructively detecting acoustic emissions (AE) generated inside a secondary battery for data acquisition. The recyclability data is data indicating whether the active material of the secondary battery for data acquisition can be recycled. The secondary battery for data acquisition is a secondary battery used for data acquisition in the generation scenario of the classification model. For example, a secondary battery prepared in an experiment or the like, or a secondary battery mounted on an electric vehicle is used as the secondary battery for data acquisition, and a sensor for non-destructively detecting the internal state of the secondary battery is attached. A pre-AE signal is acquired from the secondary battery for data acquisition from the start to the end of its use. Then, after the use of the secondary battery is completed, it is determined whether the active material of the used secondary battery can be recycled, and recyclability data is generated according to the determination result.

[0031] The recyclability data is generated by analyzing an electrode image including the active material of the secondary battery for data acquisition. That is, whether the active material of the secondary battery for data acquisition can be recycled is determined based on the state of the active material of the secondary battery for data acquisition. The analysis may be performed by human visual inspection or by image recognition processing of the electrode image by a computer. For example, based on the number of cracks per unit particle contained in the active material and / or the average length of the cracks measured from the electrode image, it is determined whether the active material of the secondary battery for data acquisition can be recycled, and recyclability data is generated. The electrode image is an SEM image and / or a CT image obtained by disassembling the cell of the secondary battery.

[0032] Specifically, when the number of cracks per unit particle contained in the active material is equal to or greater than a predetermined crack number threshold, the reusability data is generated as data indicating that the active material of the secondary battery for data acquisition cannot be reused. When the number of cracks per unit particle contained in the active material is less than the predetermined crack number threshold, the reusability data is generated as data indicating that the active material of the secondary battery for data acquisition can be reused. Further, when the average length of the cracks contained in the active material is equal to or greater than a predetermined length threshold, the reusability data is generated as data indicating that the active material of the secondary battery for data acquisition cannot be reused. When the average length of the cracks contained in the active material is less than the predetermined length threshold, the reusability data is generated as data indicating that the active material of the secondary battery for data acquisition can be reused.

[0033] The model generation unit 112 performs learning of a neural network using teacher data in which pre-AE signals and reuse-eligibility data are associated, and generates a classification model. The classification model is configured by a neural network including an input layer to which input data including pre-AE signal data is input, and an output layer that outputs output data including reuse-eligibility data. The classification model is learned to determine whether or not the active material of the secondary battery can be reused, using the teacher data associating the input data and the output data. When input data is input to the input layer, the learned classification model causes the computer to function so as to output, from the output layer, output data corresponding to the input data. In the present embodiment, the model generation unit 112 performs learning to classify whether or not the active material of the secondary battery can be reused, based on a detection signal non-destructively detected from the secondary battery, by machine learning using a data set as teacher data, and generates a learned model as the classification model. Specifically, the neural network learns so that the error between the estimated value estimated from the pre-AE signal and the true value given by the reuse-eligibility data becomes small. The learned neural network is stored in the storage device 50 as the classification model. Note that in the present embodiment, the neural network is an example, and as examples of machine learning, in addition to the neural network, for example, the k-nearest neighbor method, decision tree, logistic regression, support vector machine, naive Bayes classification, and random forest are used.

[0034] The storage device 50 is a storage medium that stores various types of information. The storage device 50 stores a classification model. In the present embodiment, when the determination device 100 generates a classification model, the generated classification model is stored in the storage device 50. The determination device 100 determines whether or not the active material of the secondary battery to be determined can be reused using the classification model pre-stored in the storage device 50 in a determination scenario. Note that, in the present embodiment, it is assumed that the classification model is generated by the determination device 100. However, the present invention is not limited to this, and the determination device 100 may acquire a learned classification model via the input device 30 and store it in the storage device 50. Further, the storage device 50 is not limited to being provided in the determination device 100, and may be a device outside the determination device 100.

[0035] Next, an example of a control procedure for executing the determination method according to the present embodiment will be described with reference to FIG. 3. FIG. 3 is a diagram showing an example of a flowchart of control processing of the determination method executed by the determination device according to the present embodiment. In the present embodiment, when the use of the secondary battery to be determined ends, the determination device 100 starts the control flow from step S101.

[0036] In step S101, the determination device 100 acquires a target detection signal from the sensor 20. The acquired target detection signal is a detection signal acquired by the sensor 20 during the use of the secondary battery to be determined. The target detection signal is, for example, an acoustic emission signal (target AE signal) obtained by non-destructively detecting elastic waves generated inside the secondary battery to be determined. In step S102, the determination device 100 inputs the input data including the target detection signal acquired in step S101 into the classification model. The classification model outputs output data indicating whether it can be reused according to the input data. In step S103, the determination device 100 determines whether the active material of the secondary battery to be determined can be reused. For example, the determination device 100 determines whether the active material of the secondary battery to be determined can be reused based on the output data output from the classification model. When the classification of the output data is a classification indicating that it can be reused, the determination device 100 determines that the active material of the secondary battery to be determined can be reused. When the classification of the output data is a classification indicating that it cannot be reused, the determination device 100 determines that the active material of the secondary battery to be determined cannot be reused.

[0037] If it is determined that the active material of the secondary battery to be determined can be reused, the determination device 100 proceeds to step S104. If it is determined that the active material of the secondary battery to be determined cannot be reused, the determination device 100 proceeds to step S105. In step S104, the determination device 100 causes the output device 40 to output a determination result indicating that the active material of the secondary battery to be determined can be reused. Specifically, the determination device 100 transmits a control instruction to the output device 40 to output image information and / or audio information indicating that the active material of the secondary battery to be determined can be reused. In step S105, the determination device 100 causes the output device 40 to output a determination result indicating that the active material of the secondary battery to be determined cannot be reused. Specifically, the determination device 100 transmits a control instruction to the output device 40 to output image information and / or audio information indicating that the active material of the secondary battery to be determined cannot be reused. The output device 40 outputs the determination result according to the control instruction.

[0038] Next, an example of a control procedure for generating the classification model according to the present embodiment will be described with reference to FIG. 4. FIG. 4 is a diagram showing an example of a flowchart of a control process for generating the classification model according to the present embodiment.

[0039] In step S201, the determination device 100 acquires, as a pre-detection signal, a detection signal obtained by non-destructively detecting the internal state of the secondary battery for data acquisition. In step S202, the determination device 100 acquires reuse-availability data indicating whether or not the active material of the secondary battery for data acquisition can be reused. In step S203, the determination device 100 generates a classification model based on the pre-detection signal and the reuse-availability data acquired in steps S201 and S202. For example, the determination device 100 performs statistical processing for determining whether or not the active material of the used secondary battery can be reused using teacher data in which the pre-detection signal and the reuse-availability data are associated, and generates a classification model. The generated classification model is stored in the storage device 50.

[0040] As described above, in the determination method and the determination device according to the present embodiment, the controller acquires, as a target detection signal, a detection signal obtained by non-destructively detecting the internal state of the secondary battery to be determined, inputs the target detection signal to the classification model, and determines whether or not the active material of the secondary battery to be determined can be reused based on the output result output from the classification model. The classification model is a model that classifies the active material of the secondary battery into categories of whether or not it can be reused based on a detection signal obtained by non-destructively detecting the internal state of the secondary battery. Thereby, it is possible to non-destructively determine whether or not the active material of the secondary battery can be reused.

[0041] Also, in the determination method and determination device according to the present embodiment, the controller acquires a detection signal obtained by non-destructively detecting the internal state of the secondary battery for data acquisition as a pre-detection signal, acquires reusability data indicating whether the active material of the secondary battery for data acquisition can be reused, generates a data set including the pre-detection signal and the reusability data, and statistically processes the data set to generate a classification model. Thereby, a classification model reflecting the relationship between the detection signal obtained by non-destructively detecting the internal state of the secondary battery and whether the active material of the secondary battery can be reused can be generated.

[0042] Also, in the determination method and determination device according to the present embodiment, the controller generates a classification model by performing machine learning based on the data set. Thereby, a classification model reflecting the non-linear correlation between the detection signal obtained by non-destructively detecting the internal state of the secondary battery and whether the active material of the secondary battery can be reused can be generated.

[0043] Also, in the determination method and determination device according to the present embodiment, the pre-detection signal is an acoustic emission signal generated inside the secondary battery for data acquisition. Thereby, a signal caused by a destructive phenomenon inside the secondary battery can be acquired non-destructively.

[0044] Also, in the determination method and determination device according to the present embodiment, the acoustic emission signal is all the acoustic emission signals detected during the use of the secondary battery for data acquisition from the start of use to the end of use. Thereby, a signal caused by a destructive phenomenon inside the secondary battery generated during the use of the secondary battery can be acquired non-destructively.

[0045] In addition, in the determination method and determination device according to the present embodiment, the acoustic emission signal is all acoustic emission signals in a frequency band of 700 kHz or higher detected during the use from the start to the end of the use of the secondary battery for data acquisition. It is known that signals derived from cracks and peeling of the active material appear in the frequency band of 700 kHz or higher. By limiting the range of the acquired signals, a more accurate classification model can be generated.

[0046] In addition, in the determination method and determination device according to the present embodiment, the acoustic emission signal includes the number of times the intensity of the acoustic emission signal becomes equal to or greater than a predetermined intensity threshold during the use from the start to the end of the use of the secondary battery for data acquisition. By using the integrated value in this way, the computational load involved in generating the classification model can be reduced.

[0047] In addition, in the determination method and determination device according to the present embodiment, the acoustic emission signal includes the number of times the intensity of the acoustic emission signal becomes equal to or greater than a predetermined intensity threshold in the frequency band of 700 kHz or higher during the use from the start to the end of the use of the secondary battery for data acquisition. Thereby, a more accurate classification model can be generated, and the computational load involved in the generation can be reduced.

[0048] In addition, in the determination method and determination device according to the present embodiment, the data on whether reuse is possible is generated by analyzing an electrode image including the active material of the secondary battery for data acquisition. Thereby, data indicating whether the active material of the secondary battery can be reused can be generated from the image information including the active material of the secondary battery.

[0049] Also, in the determination method and determination device according to the present embodiment, the reuse-eligibility data is generated as data indicating that the active material of the secondary battery for data acquisition cannot be reused when the number of cracks per unit particle contained in the active material, measured from the electrode image, is equal to or greater than a predetermined crack number threshold, and is generated as data indicating that the active material of the secondary battery for data acquisition can be reused when the number of cracks is less than the predetermined crack number threshold. Thereby, by using the quantitative value obtained from the image information including the active material of the secondary battery, the computational load related to the generation of the reuse-eligibility data can be reduced.

[0050] Also, in the determination method and determination device according to the present embodiment, the reuse-eligibility data is generated as data indicating that the active material of the secondary battery for data acquisition cannot be reused when the average length of the cracks contained in the active material, measured from the electrode image, is equal to or greater than a predetermined length threshold, and is generated as data indicating that the active material of the secondary battery for data acquisition can be reused when the average length of the cracks is less than the predetermined length threshold. Thereby, by using the quantitative value obtained from the image information including the active material of the secondary battery, the computational load related to the generation of the reuse-eligibility data can be reduced.

[0051] Also, in the determination method and determination device according to the present embodiment, a sensor for non-destructively detecting the internal state of the secondary battery to be determined is attached to the secondary battery to be determined, and the controller acquires, from the sensor, a target detection signal detected during the use of the secondary battery to be determined. Thereby, a signal generated while the secondary battery to be determined is being used can be acquired.

[0052] <<Second Embodiment>> In the first embodiment, the determination device 100 has been described as including the model generation unit 112. However, the present invention is not limited to this, and the model generation unit 112 may be provided in a device different from the determination device 100. Hereinafter, an example in which the determination system 1 includes a model generation device 200 including the model generation unit 112 separately from the determination device 100 will be described. FIG. 5 is a block diagram showing an example of the configuration of a determination system including a determination device according to the second embodiment of the present invention. As shown in FIG. 5, the determination system 1 includes a sensor 20, an output device 40, a determination device 100, and a model generation device 200. The second embodiment is different from the first embodiment in that the determination device 100 does not have the function related to the model generation unit 112, and the model generation device 200 is provided in the determination system 1. The other configurations are the same as those of the first embodiment described above. In the following description, the same configurations and control processes as those of the first embodiment will be omitted, but the description of the first embodiment will be appropriately incorporated into the omitted description.

[0053] The model generation device 200 includes a controller 21. The controller 21 generates a classification model. The controller 21 is a computer including a ROM in which a program for generating the classification model is stored, a CPU as an operation circuit that functions as the model generation device 200 by executing the program stored in the ROM, and a RAM that functions as an accessible storage device.

[0054] The controller 21 includes, as functional blocks, an acquisition unit 210 and a model generation unit 211. The controller 21 executes each function through the cooperation of software for realizing each of the above functions or executing each process and the above-described hardware. In the present embodiment, after dividing the functions of the controller 21 into two blocks, the functions of each functional block will be described. However, the functions of the controller 21 do not necessarily have to be divided into two blocks, and may be divided into one functional block or three or more functional blocks.

[0055] The acquisition unit 210 executes an acquisition process for acquiring information necessary for generating a classification model. In the acquisition process, the acquisition unit 210 acquires, as a pre-detection signal, a detection signal obtained by non-destructively detecting the internal state of a secondary battery for data acquisition, and acquires reusability data indicating whether the active material of the secondary battery for data acquisition can be reused. The model generation unit 211 generates a classification model based on the pre-detection signal and the reusability data acquired by the acquisition unit 210 from an input device (not shown). When the model generation unit 211 generates a classification model, it transmits the generated classification model to the determination device 100. When the determination device 100 receives the classification model, it stores the received classification model in the storage device 50.

[0056] Note that the embodiments described above are described to facilitate the understanding of the present invention, and are not described to limit the present invention. Therefore, each element disclosed in the above embodiments is intended to include all design changes and equivalents belonging to the technical scope of the present invention.

Explanation of reference numerals

[0057] 1…Determination system 100…Determination device 10…Controller 110…Acquisition unit 111…Determination unit 112…Model generation unit 50…Storage device 20…Sensor 30…Input device 40…Output device

Claims

1. A determination method executed by a controller for determining whether the active material of a used secondary battery can be reused, comprising: The controller: acquires, as a target detection signal, a detection signal obtained by non-destructively detecting the internal state of the secondary battery to be determined; inputs the target detection signal into a classification model; determines whether the active material of the secondary battery to be determined can be reused based on the output result output from the classification model; The classification model is a model that classifies the active material of the secondary battery into a category of whether it can be reused based on a detection signal obtained by non-destructively detecting the internal state of the secondary battery.

2. The determination method according to claim 1, comprising: The controller: acquires, as a pre-detection signal, a detection signal obtained by non-destructively detecting the internal state of the secondary battery for data acquisition; acquires reuse-eligibility data indicating whether the active material of the secondary battery for data acquisition can be reused; generates a dataset including the pre-detection signal and the reuse-eligibility data and statistically processes the dataset to generate the classification model.

3. The determination method according to claim 2, comprising: The controller: generates the classification model by performing machine learning based on the dataset.

4. The determination method according to claim 2 or 3, comprising: The pre-detection signal is an acoustic emission signal generated inside the secondary battery for data acquisition.

5. The determination method according to claim 4, comprising: The acoustic emission signal is all acoustic emission signals detected during the use of the secondary battery for data acquisition from the start of use to the end of use.

6. The determination method according to claim 5, comprising: The acoustic emission signal is all acoustic emission signals in a frequency band of 700 kHz or higher detected during the use of the secondary battery for data acquisition from the start of use to the end of use.

7. The determination method according to claim 5, comprising: The acoustic emission signal includes the number of times the intensity of the acoustic emission signal becomes equal to or greater than a predetermined intensity threshold during the use of the secondary battery for data acquisition from the start of use to the end of use.

8. The determination method according to claim 7, comprising: The method for determination includes the number of times the intensity of the acoustic emission signal is equal to or greater than a predetermined intensity threshold value in a frequency band of 700 kHz or higher during the use of the secondary battery for data acquisition from the start to the end of use.

9. The method for determination according to claim 2 or 3, wherein the data indicating whether reuse is possible is generated by analyzing an electrode image including an active material of the secondary battery for data acquisition.

10. The method for determination according to claim 9, wherein the data indicating whether reuse is possible is generated as data indicating that the active material of the secondary battery for data acquisition cannot be reused when the number of cracks per unit particle contained in the active material measured from the electrode image is equal to or greater than a predetermined crack number threshold value, and the data indicating whether reuse is possible is generated as data indicating that the active material of the secondary battery for data acquisition can be reused when the number of cracks is less than the predetermined crack number threshold value.

11. The method for determination according to claim 9, wherein the data indicating whether reuse is possible is generated as data indicating that the active material of the secondary battery for data acquisition cannot be reused when the average length of cracks contained in the active material measured from the electrode image is equal to or greater than a predetermined length threshold value, and the data indicating whether reuse is possible is generated as data indicating that the active material of the secondary battery for data acquisition can be reused when the average length of the cracks is less than the predetermined length threshold value.

12. The method for determination according to any one of claims 1 to 3, wherein a sensor for non-destructively detecting the internal state of the secondary battery to be determined is attached to the secondary battery to be determined, and the controller acquires the target detection signal detected during the use of the secondary battery to be determined from the sensor.

13. A determination device including a controller for determining whether the active material of a used secondary battery can be reused, wherein the controller acquires, as a target detection signal, a detection signal obtained by non-destructively detecting the internal state of the secondary battery to be determined, inputs the target detection signal to a classification model, and determines whether the active material of the secondary battery to be determined can be reused based on an output result output from the classification model. The classification model is a determination device that classifies the active material of the secondary battery into categories of whether it can be reused or not based on a detection signal obtained by non-destructively detecting the internal state of the secondary battery.

Images