Battery deterioration state estimation method and battery deterioration state estimation device

Abstract

The controller (23) constructs a model using, as explanatory variables, measured value information including an AC resistance value measured for learning, and use history information acquired for learning, and using, as a target variable, a deterioration state measured for learning. Using the constructed model, the controller (43) estimates, as a target variable, the deterioration state of the battery (42) using, as explanatory variables, measured value information including the AC resistance value measured for estimation and use history information acquired for estimation.

Description

Technical Field

[0001] The present invention relates to a method for estimating the deterioration state of a battery and a device for estimating the deterioration state of a battery. Background Art

[0002] In Patent Document 1, the following is proposed: The correspondence relationship between the temperature history, the input / output resistance, and the full charge capacity of a secondary battery is stored in advance, and the full charge capacity of the secondary battery is estimated based on the temperature history and the input / output resistance of the secondary battery.

[0003] Prior Art Documents

[0004] Patent Documents

[0005] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2018-132369 Summary of the Invention

[0006] Problems to be Solved by the Invention

[0007] If only the temperature history is considered, it is impossible to know how the storage battery is used. Therefore, if only the temperature history is considered, it is difficult to accurately estimate the deterioration state of the storage battery.

[0008] The problem of the present invention is to improve the estimation accuracy of the deterioration state of the storage battery.

[0009] Solutions to the Problems

[0010] According to one aspect of the present invention, a learning controller performs the following processes: measuring the AC resistance value of a learning target battery mounted on a vehicle by electrochemical impedance spectroscopy, and obtaining measurement value information including at least the real-axis component of the AC resistance value measured for learning. The learning controller performs the following processes: obtaining usage history record information of the learning target battery for learning. The learning controller performs the following processes: measuring the deterioration state of the learning target battery for learning. The learning controller performs the following processes: constructing a model in which the measurement value information measured for learning and the usage history record information obtained for learning are used as explanatory variables, and the deterioration state measured for learning is used as a target variable. An estimation controller performs the following processes: measuring the AC resistance value of an estimation target battery by electrochemical impedance spectroscopy, and obtaining measurement value information including at least the real-axis component of the AC resistance value measured for estimation. The estimation controller performs the following processes: obtaining usage history record information of the estimation target battery for estimation. The estimation controller performs the following processes: using the model, and using the measurement value information measured for estimation and the usage history record information obtained for estimation as explanatory variables to estimate the deterioration state of the estimation target battery as a target variable.

[0011] Effects of the Invention

[0012] According to the present invention, by considering the usage history information of the storage battery, the estimation accuracy of the deterioration state of the storage battery can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a block diagram showing a battery deterioration state estimation device.

[0014] Figure 2 is a flowchart showing a learning process.

[0015] Figure 3 is a Nyquist diagram of the AC resistance value.

[0016] Figure 4 is a schematic diagram of e-dimensional data.

[0017] Figure 5 is a schematic diagram of m-dimensional data.

[0018] Figure 6 is a schematic diagram of a design matrix.

[0019] Figure 7 is a schematic diagram of N-dimensional data.

[0020] Figure 8 is a flowchart showing an estimation process.

[0021] Figure 9 is a diagram showing a second embodiment.

[0022] Figure 10 is a diagram showing a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0023] Hereinafter, embodiments of the present invention will be described based on the drawings. In addition, each drawing is a schematic drawing and is sometimes different from reality. Further, the following embodiments illustrate devices and methods for embodying the technical idea of the present invention, and do not specifically limit the structure to the following structure. That is, various changes can be made to the technical idea of the present invention within the scope of protection described in the claims.

[0024] 《First Embodiment》

[0025] 《Structure》

[0026] Figure 1 is a block diagram showing a battery deterioration state estimation device.

[0027] The battery degradation state estimation device 11 includes a vehicle 21 as a learning object and a vehicle 41 as an estimation object. The vehicle 21 is an electric vehicle or a hybrid vehicle capable of driving a motor for traveling, and includes a storage battery 22 (learning object storage battery) and a controller 23 (learning controller). The storage battery 22 is, for example, a lithium-ion battery used for vehicle driving, but may also be a sodium-ion battery or an all-solid-state battery.

[0028] An EIS sensor 31, a current sensor 32, a voltage sensor 33, and a temperature sensor 34 are provided in the storage battery 22. The EIS sensor 31 measures the AC resistance value Z of the storage battery 22 by electro-chemical impedance spectroscopy (EIS: Electro-chemical Impedance Spectroscopy). The current sensor 32 measures the current I of the storage battery 22. The voltage sensor 33 measures the voltage E of the storage battery 22. The temperature sensor 34 measures the temperature Tb of the storage battery 22.

[0029] An external air temperature sensor 35, a positioning device 36, and a charge / discharge controller 37 are provided in the vehicle 21. The external air temperature sensor 35 measures the external air temperature To. The positioning device 36 obtains the position information of the vehicle 21 by GPS. When charging the storage battery 22, the charge / discharge controller 37 obtains the charging type information of whether it is normal charging or rapid charging, and when discharging the storage battery 22, the charge / discharge controller 37 obtains the discharge type information of whether it is discharge based on traveling or discharge based on power supply to, for example, home or smart grid other than traveling. The charging type information and the discharge type information can be obtained via the controller 23 of the vehicle 21 described later, the management device of the charger, the management device of the power supply target, etc.

[0030] The controller 23 is constituted by, for example, a microcomputer, and includes a measurement value information acquisition unit 24, a history information acquisition unit 25, a degradation state measurement unit 26, and a model construction unit 27. The measurement value information acquisition unit 24 acquires the measurement value information including at least the real-axis component Zr of the AC resistance value Z acquired by the EIS sensor 31. The history information acquisition unit 25 acquires the usage history information of the storage battery 22. The degradation state measurement unit 26 measures the degradation state (SOH: States Of Health) of the storage battery 22 by an existing method. The model construction unit 27 constructs a model for each type of storage battery single cell, with the measurement value information measured for learning and the usage history information acquired for learning as explanatory variables, and the degradation state SOH measured for learning as the target variable.

[0031] The vehicle 41 is equipped with a storage battery 42 (the storage battery to be estimated) and a controller 43 (the controller for estimation). The type of the single-cell battery of the storage battery 42 is the same as that of the storage battery 22 mounted on the vehicle 21 as the object of learning. An EIS sensor 51, a current sensor 52, a voltage sensor 53, and a temperature sensor 54 are provided in the storage battery 42. An outside air temperature sensor 55, a positioning device 56, and a charge-discharge controller 57 are provided in the vehicle 41. The controller 43 includes a measured value information acquisition unit 44, a history information acquisition unit 45, a model storage unit 46, and a degradation state estimation unit 47. For parts common to the structure of the vehicle 21, detailed description is omitted. The model storage unit 46 stores the model constructed by the model construction unit 27. The degradation state estimation unit 47 uses the model, takes the measured value information measured for estimation and the usage history information acquired for estimation as explanatory variables, and estimates the state of health SOH of the storage battery 42 as the target variable.

[0032] Next, the learning process executed by the controller 23 of the vehicle 21 as the object of learning will be described. This process is executed at regular intervals. However, since the degradation of the storage battery 22 does not progress every day, it may be executed, for example, once a week or once a month, but of course it is not limited to this.

[0033] Figure 2 It is a flowchart showing the learning process.

[0034] First, in step S101, the AC impedance, that is, the AC resistance value Z of the storage battery 22 is measured by the EIS sensor 31. The measurement of the AC resistance value Z can be carried out at any timing, but it is desirable to carry it out in a quiet state after parking for a specified time without performing driving discharge and charging to improve the measurement accuracy.

[0035] Figure 3 It is a Nyquist diagram of the AC resistance value Z.

[0036] Here, with the real-axis component Zr of the AC resistance value Z as the horizontal axis and the imaginary-axis component Zx of the AC resistance value Z as the vertical axis, the AC resistance value Z corresponding to the frequency fa is plotted. The measured AC resistance value Z is set to a frequency band of 100 Hz or more where there is no large change in the real-axis component Zr.

[0037] Return to Figure 2In the next step S102, e-dimensional data (first dimensional data) having e feature quantities is generated based on the measured value information measured for learning. e is a natural number greater than or equal to 1. The measured value information includes at least the real axis component Zr of the AC resistance value Z. The measured value information may also include any one of the imaginary axis component Zx of the AC resistance value Z, the difference ΔZ between the DC resistance value DCR and the real axis component Zr, the state of charge SOC (State Of Charge), and the open circuit voltage OCV as other feature quantities.

[0038] Figure 4 It is a schematic diagram of e-dimensional data.

[0039] One feature quantity is schematically represented by a square, and multiple feature quantities are arranged in a row and drawn. Here, as an example, e-dimensional data having three feature quantities is assumed.

[0040] Return to Figure 2 In the next step S103, the usage history information of the battery 22 is obtained. The usage history information includes the maximum temperature Tb of the battery during charging and discharging. MAX , the retention time Tstemp of the battery temperature segment during parking, the retention time Tssoc of the state of charge segment during parking, the state of charge SOC at the start of charging and discharging, the battery temperature Tb at the start of charging and discharging, the charge and discharge current Qc, and the charge and discharge type. The usage history information is stored as time-series data, with discharge and charge each being an event, and each event having one or more characteristic quantities. Discharge includes not only discharge due to driving but also discharge due to power supply to the home or smart grid. Cooling performance varies depending on the module's mounting location, and deterioration progresses when exposed to high temperatures. Therefore, the battery temperature may also be classified by the module's mounting location, statistics such as the maximum, average, and variance related to the outside air temperature To, the average latitude information in the vehicle's location information, and regional classification using the mode value of latitude and longitude. Battery temperature segments may be, for example, low, medium, and high temperature segments. The charge and discharge current Ic is the total value during the event. A high frequency of power supply indicates a high frequency of charge and discharge cycles. Furthermore, if the variance of the SOC at the start of charging is high, it means that the SOC range (swing width) used is wide, and the frequency of charge and discharge cycles remains high. This charge and discharge history also serves as an indicator of the state of degradation (SOH). The usage history information can also include information on the proportion of discharge current to total charge current based on the aforementioned power supply and the type of cell of the battery 22 being learned.

[0041] In the next step S104, m-dimensional data (second-dimensional data) with m feature quantities is generated based on the usage history information obtained for learning. m is a natural number of 1 or more. Specifically, the m-dimensional data is generated by performing statistical processing on the usage history information held as time-series data. For example, statistics including sum, maximum value, minimum value, average value, quartile value, mode value, and variance are used. Regarding the feature quantities, in model learning, a general feature quantity selection unit is used to select feature quantities with high importance through, for example, a decision tree model, PFI, or to select them through a lasso regression model.

[0042] Figure 5 It is a schematic diagram of the m-dimensional data.

[0043] One feature quantity is schematically represented by a square, and multiple feature quantities are arranged in a row for depiction. Here, as an example, it is assumed to be m-dimensional data with 5 feature quantities.

[0044] Return to Figure 2 the description of. In the next step S105, N pieces of e + m-dimensional data (third-dimensional data) obtained by combining the e-dimensional data generated for learning and the m-dimensional data generated for learning are collected to generate a design matrix. N is a natural number of e + m or more and is a large number of sample data required for machine learning. In order to improve the learning accuracy, N needs to be set to a large value. It is also possible to perform standardization processing such as an operation to make the average zero and the variance 1, an operation to make the median zero and the interquartile range 1, an operation to limit the range of the variable to a specific interval, or further perform processing based on non-linear transformation by logarithmic conversion on the design matrix.

[0045] Figure 6 It is a schematic diagram of the design matrix.

[0046] By arranging the e-dimensional data generated for learning and the m-dimensional data generated for learning in a row to form e + m-dimensional data, and arranging this e + m-dimensional data in a column, a design matrix with the number of rows set to N and the number of columns set to e + m is generated.

[0047] Return to Figure 2 the description of. In the next step S106, the state of health (SOH) of the storage battery 22 is measured for learning. Here, as an existing method, for example, the state of health (SOH) of the storage battery 22 is measured by measuring the cumulative discharge amount based on the fully charged state.

[0048] In the next step S107, N-dimensional data (fourth-dimensional data) is generated based on the N measured SOHs of the degradation states for learning. The number of N-dimensional data corresponds to the number of rows of the design matrix.

[0049] Figure 7 is a schematic diagram of the N-dimensional data.

[0050] A degradation state SOH is schematically represented by a square, and a large number of degradation states SOH are arranged in a column for depiction.

[0051] Return to Figure 2 the description of. In the next step S108, a model is constructed with the design matrix generated for learning as the explanatory variable and the N-dimensional data generated for learning as the target variable, and the model is stored. The model is constructed for each type of battery cell. Usually, the SOH of the degradation state is expressed in the range of SOH = {SOH|0 < SOH < 1}. Therefore, it is also possible to construct a learning of a regression model with z, which is obtained by performing the logit transformation shown in Equation (1) below, that is, z = {z|-∞ < z < ∞}, as the target variable. Thereby, it is possible to avoid the situation where the value obtained by estimating the SOH of the degradation state using the model becomes less than 0 or exceeds 1. Here, a Gaussian process regression model using a kernel function is adopted, but it is not limited thereto. As a machine learning algorithm, in addition to this, lasso regression, elastic net regression, support vector regression, random forest, neural network, etc. can also be adopted.

[0052] [Equation (1)]

[0053]

[0054] In addition, when the number of e+m-dimensional data and the number of N-dimensional data are less than N, it directly returns to the main program. Therefore, it becomes a loop process that repeats the processing of steps S101 to S107 until the number of e+m-dimensional data and the number of N-dimensional data reach N. At the time when the number of e+m-dimensional data and the number of N-dimensional data reach N, an initial model is constructed. After that, the learned model is updated by increasing the number of e+m-dimensional data and N-dimensional data, or updating the e+m-dimensional data and N-dimensional data.

[0055] The above is the learning process executed by the controller 23. Among them, the processing of step S101 corresponds to the measured value information acquisition unit 24, and the processing of step S103 corresponds to the history information acquisition unit 25. In addition, the processing of step S106 corresponds to the degradation state measurement unit 26, and the processing of step S108 corresponds to the processing of the model construction unit 27.

[0056] Next, the estimation process executed by the controller 43 of the vehicle 41 to be estimated will be described. This process is executed at regular intervals. However, since the deterioration of the battery 42 does not progress every day, it may be executed, for example, once a week or once a month, but of course it is not limited to this.

[0057] Figure 8 is a flowchart showing the estimation process.

[0058] First, in step S111, the EIS sensor 51 measures the AC resistance value Z, which is the AC impedance of the battery 42. The concept is the same as the process in step S101 during learning.

[0059] In the next step S112, e-dimensional data (first-dimensional data) with e characteristic quantities is generated based on the measured value information measured for estimation. The concept is the same as the process in step S102 during learning.

[0060] In the next step S113, the usage history information of the battery 42 is obtained. The concept is the same as the process in step S103 during learning.

[0061] In the next step S114, m-dimensional data (second-dimensional data) with m characteristic quantities is generated based on the usage history information obtained for estimation. The concept is the same as the process in step S104 during learning.

[0062] In the next step S115, the model stored in the model storage unit 46 is used to estimate the deterioration state SOH of the battery 42. That is, as shown in Equation 2 below, the e + m-dimensional data (third-dimensional data) obtained by combining the e-dimensional data generated for estimation and the m-dimensional data generated for estimation is used as the explanatory variable, and the deterioration state SOH of the battery 42 is estimated as the target variable.

[0063] [Equation 2]

[0064]

[0065] In the case of performing a logit transformation during learning, by performing the inverse logit transformation shown in Equation 3 below on the estimated value from the learned regression model, the deterioration state SOH expressed in the range of SOH = {SOH|0 < SOH < 1} can be obtained.

[0066] [Equation 3]

[0067]

[0068] The above is the estimation process executed by the controller 43. Among them, the process of step S111 corresponds to the measured value information acquisition unit 44, the process of step S113 corresponds to the history information acquisition unit 45, and the process of step S115 corresponds to the deterioration state estimation unit 47.

[0069] 《Effect》

[0070] Next, the main effects of the first embodiment will be described.

[0071] (1) The controller 23 executes the following processes: measuring the AC resistance value Z of the storage battery 22 for vehicle drive by electrochemical impedance spectroscopy, and acquiring the measured value information including at least the real-axis component Zr of the AC resistance value Z measured for learning. The controller 23 executes the following processes: acquiring the usage history information of the storage battery 22 for learning; and measuring the state of health (SOH) of the deterioration of the storage battery 22 for learning. The controller 23 executes the following processes: constructing a model in which the measured value information measured for learning and the usage history information acquired for learning are used as explanatory variables, and the state of health (SOH) of the deterioration measured for learning is used as the target variable. Although the deterioration mode varies depending on the usage method of the storage battery 22, a model with improved estimation accuracy of the state of health (SOH) of the deterioration can be constructed by performing machine learning while considering the usage history information of the storage battery 22. The controller 43 executes the following processes: measuring the AC resistance value Z of the storage battery 42 for vehicle drive by electrochemical impedance spectroscopy, and acquiring the measured value information including at least the real-axis component Zr of the AC resistance value Z measured for estimation. The controller 43 executes the following processes: acquiring the usage history information of the storage battery 42 for estimation. The controller 43 executes the following processes: using the model, using the measured value information measured for estimation and the usage history information acquired for estimation as explanatory variables, and estimating the state of health (SOH) of the deterioration of the storage battery 42 as the target variable. Although the deterioration mode varies depending on the usage method of the storage battery 42, the estimation accuracy of the state of health (SOH) of the deterioration of the storage battery 42 can be improved by considering the usage history information of the storage battery 42.

[0072] (2) The measured value information includes at least one of the imaginary-axis component Zx of the AC resistance value Z, the difference ΔZ between the DC resistance value DCR and the real-axis component Zr, the state of charge (SOC), and the open-circuit voltage (OCV) in addition to the real-axis component Zr. By considering, in addition to the real-axis component Zr, the characteristic quantities that can represent the change in the Nyquist diagram, the estimation accuracy of the state of health (SOH) of the deterioration can be further improved. In addition, as the interfacial resistance, a value equivalent to the electrode resistance can be simulated. In addition, the influence of electron / ion conduction caused by the volume change of the positive and negative electrodes can be considered.

[0073] (3) The usage history information includes the maximum battery temperature Tb during charge and discharge. MAX ..., the residence time Tstemp of the battery temperature section during parking, the residence time Tssoc of the charge state section during parking, the charge state SOC at the start of charge and discharge, the battery temperature Tb at the start of charge and discharge, the charge and discharge current Qc, and at least one of the charge and discharge types. Thus, it is possible to consider factors that affect the degradation state SOH, use further limited data to reduce the calculation load, and improve the estimation accuracy of the degradation state SOH.

[0074] (4) The AC resistance value Z is in a frequency band of 100 Hz or higher. Thus, it is possible to measure the electrolyte resistance component in the AC resistance value Z that has a high correlation with the degradation state SOH.

[0075] (5) The controller 23 performs the following processes: generating e-dimensional data with e feature quantities based on the measured value information measured for learning; and generating m-dimensional data with m feature quantities based on the usage history information obtained for learning. The controller 23 performs the following processes: collecting N e+m-dimensional data obtained by combining the e-dimensional data generated for learning and the m-dimensional data generated for learning to generate a design matrix, where N is e+m or more. The controller 23 performs the following processes: generating N-dimensional data based on the N degradation state SOCs measured for learning. The controller 23 constructs a model using the design matrix generated for learning and the N-dimensional data generated for learning. Thus, it is possible to perform machine learning considering the usage history information of the battery 22 to construct a model that improves the estimation accuracy of the degradation state SOH. The controller 43 performs the following processes: generating e-dimensional data with e feature quantities based on the measured value information measured for estimation; and generating m-dimensional data with m feature quantities based on the usage history information obtained for estimation. The controller 43 uses the model to estimate the degradation state SOH based on the e+m-dimensional data obtained by combining the e-dimensional data generated for estimation and the m-dimensional data generated for estimation. Thus, it is possible to improve the estimation accuracy of the degradation state SOH of the battery 42 by considering the usage history information of the battery 42.

[0076] The controller 23 constructs a model for each type of battery cell. This is because the influence on the degradation state SOH corresponding to various feature quantities varies for each type of battery cell. Therefore, by constructing a model for each type of battery cell, it is possible to construct a model that improves the estimation accuracy.

[0077] The controller 23 is installed in the vehicle 21 to be learned, and the controller 43 is installed in the vehicle 41 to be estimated. Thus, the original data can be directly referred to, and the operation using data with high time resolution and information resolution can be performed during the vehicle use. Therefore, the deterioration state SOH can be estimated with high accuracy.

[0078] Next, a comparative example will be described.

[0079] Here, a structure for estimating the deterioration state SOH of the battery by considering the AC resistance value Z measured by the EIS sensor and the temperature history of the battery is set as a comparative example. Even when the temperature history is considered, when the usage mode of frequent charge and discharge is performed, the correlation between the AC resistance value Z and the deterioration state SOH changes. This is related to the following situations: sodium electrolytic deposition occurs, or the ion / electron conduction path is cut off and the active material is isolated, or the reaction surface becomes inactivated. Therefore, only from the temperature history, it is impossible to know how the battery is used. Thus, it is difficult to accurately estimate the deterioration state SOC of the battery only by considering the temperature history. In addition, even if calibration is performed using the reference calibration curve utilizing the AC resistance value Z, multiple factors such as the charge state SOC, temperature history, and load cycle will complexly affect the deterioration of the battery. Therefore, the performance based on the univariate regression model is insufficient.

[0080] 《Second Embodiment》

[0081] 《Structure》

[0082] The second embodiment is a method of installing the controller 43 for estimating the deterioration state SOH in an external server. For the other structures, they are the same as those of the foregoing first embodiment. Therefore, for the common parts, detailed descriptions are omitted.

[0083] Figure 9 It is a diagram showing the second embodiment.

[0084] Here, an external server 61 is provided, and the controller 43 is installed in the external server 61. A communication device 62 is provided in the vehicle 41, and the communication device 62 communicates with the external server 61 via the Internet. A communication device 63 is provided in the external server 61, and the communication device 63 communicates with the vehicle 41 via the Internet. The controller 43 receives various information measured by the vehicle 41 to estimate the deterioration state SOH and sends the estimation result to the vehicle 41.

[0085] 《Effect》

[0086] Next, the main effects of the second embodiment will be described.

[0087] (1) The controller 43 is installed in an external server 61 capable of communicating with the vehicle 41. Thereby, the computational load and data capacity performed by the vehicle 41 can be reduced, and the estimation accuracy of the degradation state SOH can be improved. In addition, since a model updated in a cloud environment can be used, the estimation accuracy of the degradation state SOH can be improved.

[0088] For other effects brought about by the common structure, it is the same as in the foregoing first embodiment.

[0089] 《Variant Example》

[0090] In the second embodiment, a structure in which only the controller 43 for estimating the degradation state SOH is installed in an external server has been described, but it is not limited thereto. That is, the controller 23 for constructing a model may also be installed in an external server capable of communicating with the vehicle 21. Thereby, a model that reduces the computational load or data capacity performed by the vehicle 21 and improves the estimation accuracy of the degradation state SOH can be constructed. In addition, since the design matrix generated for learning and the N-dimensional data generated for learning can be updated in a cloud environment, a model that improves the estimation accuracy of the degradation state SOH can be constructed. In addition, since the measurement value information and usage history information provided from multiple vehicles 21 can be referred to collectively, a model that improves the estimation accuracy of the degradation state SOH can be constructed.

[0091] 《Third Embodiment》

[0092] 《Structure》

[0093] The third embodiment is a method of estimating the degradation state SOH of the battery 42 after being removed from the vehicle 41 by installing the controller 43 for estimation in an external server. For the other structures, it is the same as in the foregoing first embodiment, and thus detailed description of the common parts is omitted.

[0094] Figure 10 It is a diagram showing the third embodiment.

[0095] Here, an external server 71 is provided, and a controller 43 is installed in the external server 71. An EIS sensor 72 is installed in the battery 42 after being removed from the vehicle 41. A measured value information acquisition unit 44 acquires measured value information including at least the real-axis component Zr of the AC resistance value Z acquired by the EIS sensor 72. The controller 43 includes a vehicle data storage unit 73. The vehicle data storage unit 73 stores usage history information of the battery 22. A history information acquisition unit 45 acquires the usage history information of the battery 22 from the vehicle data storage unit 73. In addition, the AC resistance value Z is measured by the EIS sensor 72, but it is not limited thereto. That is, if the AC resistance value Z finally measured by the vehicle 41 can be acquired from the vehicle data storage unit 73, the AC resistance value Z can be used. In addition, of course, for a used-up battery without the EIS sensor 72 installed, the resistance can be measured by an EIS measuring device set in an offline (non-vehicle-mounted state, so-called outside the vehicle) manner.

[0096] 《Effect》

[0097] Next, the main effects of the third embodiment will be described.

[0098] (1) The controller 43 estimates the deterioration state SOH of the battery 42 after being removed from the vehicle 41. Thereby, for a semi-old battery 42, the estimation accuracy of the deterioration state SOH can also be improved. As a result, the reuse of the battery 42 can be promoted.

[0099] For other effects brought about by the common structure, they are the same as those of the first embodiment described above.

[0100] As described above, the description is made with reference to a limited number of embodiments. However, the scope of the claims is not limited thereto, and changes based on the disclosed embodiments above will be obvious to those skilled in the art.

[0101] Description of Reference Numerals

[0102] 11: Battery degradation state estimation device; 21: Vehicle; 22: Storage battery; 23: Controller; 24: Measured value information acquisition unit; 25: History record information acquisition unit; 26: Degradation state measurement unit; 27: Model construction unit; 31: EIS sensor; 32: Current sensor; 33: Voltage sensor; 34: Temperature sensor; 35: External air temperature sensor; 36: Positioning device; 37: Charge and discharge controller; 41: Vehicle; 42: Storage battery; 43: Controller; 44: Measured value information acquisition unit; 45: History record information acquisition unit; 46: Model storage unit; 47: Degradation state estimation unit; 51: EIS sensor; 52: Current sensor; 53: Voltage sensor; 54: Temperature sensor; 55: External air temperature sensor; 56: Positioning device; 57: Charge and discharge controller; 61: External server; 62: Communication device; 63: Communication device; 71: External server; 72: EIS sensor; 73: Vehicle data storage unit.

Claims

1. A method for estimating the deterioration state of a battery, characterized in that, Cause the learning controller to perform the following processes: Measure the AC resistance value of the battery to be learned for vehicle drive by electrochemical impedance spectroscopy, and obtain measurement value information including at least the real-axis component of the AC resistance value measured for learning; Obtain the usage history record information of the battery to be learned for learning; Measure the degradation state of the battery to be learned for learning; and Construct a model that uses the measurement value information measured for learning and the usage history record information obtained for learning as explanatory variables, and uses the degradation state measured for learning as the target variable, Cause the estimation controller to perform the following processes: Measure the AC resistance value of the battery to be estimated for vehicle drive by electrochemical impedance spectroscopy, and obtain measurement value information including at least the real-axis component of the AC resistance value measured for estimation; Obtain the usage history record information of the battery to be estimated for estimation; and Use the model to estimate the degradation state of the battery to be estimated as the target variable, using the measurement value information measured for estimation and the usage history record information obtained for estimation as explanatory variables.

2. The battery degradation state estimation method according to claim 1, wherein The measurement value information includes at least one of the imaginary-axis component of the AC resistance value, the difference between the DC resistance value and the real-axis component, the state of charge, and the open-circuit voltage, in addition to the real-axis component.

3. The battery degradation state estimation method according to claim 1, wherein The usage history record information includes at least one of the maximum battery temperature during charge and discharge, the residence time in the battery temperature section during parking, the residence time in the state of charge section during parking, the state of charge at the start of charge and discharge, the battery temperature at the start of charge and discharge, the charge and discharge current, and the charge and discharge type.

4. The battery degradation state estimation method according to claim 1, wherein The AC resistance value is in a frequency band of 100 Hz or higher.

5. The battery degradation state estimation method according to claim 1, wherein The learning controller performs the following processes: Generate first-dimensional data with a predetermined number of feature quantities based on the measurement value information measured for learning; Generate second-dimensional data with a predetermined number of feature quantities based on the usage history record information obtained for learning; Collect a plurality of third-dimensional data obtained by combining the first-dimensional data generated for learning and the second-dimensional data generated for learning to generate a design matrix; and Generate fourth-dimensional data based on a plurality of the degradation states measured for learning, Construct the model using the design matrix generated for learning and the fourth-dimensional data generated for learning, The estimation controller performs the following processes: Generate first-dimensional data with a predetermined number of feature quantities based on the measurement value information measured for estimation; and Generate second-dimensional data with a predetermined number of feature quantities based on the usage history information obtained for estimation. Using the model, estimate the deterioration state based on the third-dimensional data obtained by combining the first-dimensional data generated for estimation and the second-dimensional data generated for estimation.

6. The method for estimating the battery deterioration state according to claim 1, characterized in that: The learning controller constructs the model for each type of individual battery of the storage battery.

7. The method for estimating the battery deterioration state according to claim 1, characterized in that: At least one of the learning controller and the estimation controller is installed in a vehicle.

8. The method for estimating the battery deterioration state according to claim 1, characterized in that: At least one of the learning controller and the estimation controller is installed in an external server capable of communicating with a vehicle.

9. The method for estimating the battery deterioration state according to claim 1, characterized in that: The estimation controller estimates the deterioration state of the estimation target storage battery removed from the vehicle.

10. A battery deterioration state estimation device, characterized in that, A learning controller is provided, and the learning controller performs the following processing: Measure the AC resistance value of the learning target storage battery for vehicle drive by electrochemical impedance spectroscopy, and obtain measurement value information including at least the real-axis component including the AC resistance value measured for learning; Obtain the usage history information of the learning target storage battery for learning; Measure the deterioration state of the learning target storage battery for learning; and Construct a model that uses the measurement value information measured for learning and the usage history information obtained for learning as explanatory variables, and uses the deterioration state measured for learning as the target variable. The battery deterioration state estimation device further includes an estimation controller, and the estimation controller performs the following processing: Measure the AC resistance value of the estimation target storage battery for vehicle drive by electrochemical impedance spectroscopy, and obtain measurement value information including at least the real-axis component including the AC resistance value measured for estimation; Obtain the usage history information of the estimation target storage battery for estimation; and Using the model, use the measurement value information measured for estimation and the usage history information obtained for estimation as explanatory variables to estimate the deterioration state of the estimation target storage battery as the target variable.

Images