Computer readable medium storing a parameter fitting program, parameter fitting method, and parameter fitting device

Abstract

Traditional methods require extensive calculations to fit the parameters of the hysteresis voltage model. The parameter fitting procedure of this invention executes the following steps: Model building process (S1), which defines a hysteresis voltage model that forms an equivalent circuit using resistance parameter R and capacitance parameter C; Impulse response measurement process (S2), which obtains at least one voltage parameter from the measured value of the pulse voltage of the secondary battery; Resistance parameter setting process (S3), which calculates the value of resistance parameter R based on the voltage parameter; and Capacitance parameter optimization process (S4), which applies a current signal identical to the pulse current to the hysteresis voltage model to obtain a voltage response value and optimizes the value of capacitance parameter C in the hysteresis voltage model so that the voltage response value matches the value of the voltage parameter obtained in the impulse response measurement process. The hysteresis voltage model includes the resistance parameter R calculated in the resistance parameter setting process (S3).

Description

Technical Field

[0001] The present invention relates to a computer-readable medium storing a parameter fitting program, a parameter fitting method, and a parameter fitting apparatus for fitting parameters of a hysteresis voltage model that estimates the hysteresis voltage generated during the charging and discharging of a secondary battery. Background Technology

[0002] In secondary batteries, there is a hysteresis characteristic where the voltage at the start of a voltage rise during charging or discharging differs from the voltage at the end of a voltage drop. Furthermore, in the control of secondary batteries, a state estimation model is used to estimate the charging rate, etc., based on the internal state of the secondary battery. Moreover, a hysteresis voltage model is included as a part of the state estimation model to estimate the hysteresis characteristic of the secondary battery. Examples of techniques for estimating parameters within this hysteresis voltage model are disclosed in Japanese Patent Application Publication Nos. 2017-198542 and 2019-185899.

[0003] The parameter estimation device disclosed in Japanese Patent Application Publication No. 2017-198542 is a parameter estimation device for a battery. It estimates the parameters of an equivalent circuit model of a battery having an overvoltage model and a hysteresis model. The device is characterized by inputting a first current having a first amplitude and a second current having a second amplitude smaller than the first amplitude to the battery, estimating the parameters related to the overvoltage model based on the output of the battery corresponding to the first current, and estimating the parameters related to the hysteresis model based on the output of the battery corresponding to the second current.

[0004] The hysteresis voltage estimation device disclosed in Japanese Patent Application Publication No. 2019-185899 includes: a current measuring unit that measures the current flowing through a battery forming a multi-stage structure during charging and discharging; a voltage measuring unit that measures the inter-terminal voltage of the battery; a state of charge (SOC) estimation unit that estimates the SOC of the battery based on the current value measured by the current measuring unit or the voltage value measured by the voltage measuring unit; a mole fraction storage unit that stores the mole fraction of electrode material at each stage of charging (i.e., a first variable group) and the mole fraction of electrode material at each stage of discharging (i.e., a second variable group); and a transfer probability calculation unit that calculates a proportional coefficient representing the transfer probability between the first variable group and the second variable group during charging, based on the current value, the first variable group, and the second variable group. The system comprises: a first coefficient group; a second coefficient group that calculates proportional coefficients representing the transition probabilities of each discharge between the first variable group and the second variable group during discharge; a mole fraction calculation unit that, in addition to the first variable group and the second variable group, calculates a new first variable group and a new second variable group at the current moment based on the first coefficient group or the second coefficient group, and stores the new first variable group and the new second variable group in the mole fraction storage unit; and a hysteresis voltage calculation unit that calculates the hysteresis voltage based on the ratio of the sum of the new first variable group to the sum of the new second variable group, and the voltage determined by the SoC. Summary of the Invention

[0005] However, the technology described in Japanese Patent Application Publication No. 2017-198542 and Japanese Patent Application Publication No. 2019-185899 has the following problem: in order to improve the estimation accuracy of the hysteresis characteristics, it is necessary to use processing that requires high computing power, such as processing using Kalman filtering or processing using mole fraction.

[0006] The present invention was made in view of the above circumstances, and its object is to enable high-precision estimation of hysteresis characteristics by performing parameter fitting processing of the hysteresis voltage model with a small amount of computation.

[0007] In one aspect of the computer-readable medium storing a parameter fitting program, the parameter fitting program performs fitting processing on parameters for constructing a hysteresis voltage model, the hysteresis voltage model being connected in parallel with a battery model estimating the internal state of a secondary battery, and estimating the hysteresis of the output voltage; the parameter fitting program causes a computer to perform: a model building process that defines a hysteresis voltage model comprising an equivalent circuit through resistance parameters and capacitance parameters; an impulse response measurement process that obtains at least one voltage parameter as a fitting target from measured values ​​of the impulse voltage of the secondary battery obtained by applying a impulse current to the secondary battery; a resistance parameter setting process that calculates the value of the resistance parameter based on the voltage parameter; and a capacitance parameter optimization process that applies a current signal identical to the impulse current to the hysteresis voltage model to obtain a voltage response value and optimizes the value of the capacitance parameter in the hysteresis voltage model such that the voltage response value matches the value of the voltage parameter obtained in the impulse response measurement process, the hysteresis voltage model including the resistance parameter whose value was determined in the resistance parameter setting process.

[0008] The parameter fitting method involved in this invention is a parameter fitting method for constructing a hysteresis voltage model, which is connected in parallel with a battery model that estimates the internal state of a secondary battery and estimates the hysteresis of the output voltage. This parameter fitting method is executed through computer-based automatic processing: a model building process that defines a hysteresis voltage model that forms an equivalent circuit using resistance and capacitance parameters; a pulse response measurement process that obtains at least one voltage parameter as a fitting target from the measured value of the pulse voltage of the secondary battery obtained by applying a pulse current to the secondary battery; a resistance parameter setting process that calculates the value of the resistance parameter based on the voltage parameter; and a capacitance parameter optimization process that applies a current signal identical to the pulse current to the hysteresis voltage model to obtain a voltage response value and optimizes the value of the capacitance parameter in the hysteresis voltage model such that the voltage response value matches the value of the voltage parameter obtained in the pulse response measurement process. The hysteresis voltage model includes the resistance parameter whose value was determined in the resistance parameter setting process.

[0009] The parameter fitting device of this invention, within a control device for controlling a secondary battery, performs fitting processing on parameters for constructing a hysteresis voltage model. This hysteresis voltage model is connected in parallel with a battery model that estimates the internal state of the secondary battery, and estimates the hysteresis of the output voltage. The parameter fitting device includes a memory and an arithmetic unit that performs processing using the memory. The arithmetic unit performs: model construction processing, which defines a hysteresis voltage model that forms an equivalent circuit using resistance and capacitance parameters; and impulse response measurement processing, which measures the hysteresis voltage obtained by applying a pulse current to the secondary battery. At least one voltage parameter is obtained as a fitting target from the measured value of the pulse voltage of the secondary battery; a resistance parameter setting process is performed, which calculates the value of the resistance parameter based on the voltage parameter; and a capacitance parameter optimization process is performed, which applies the same current signal as the pulse current to the hysteresis voltage model to obtain a voltage response value and optimizes the value of the capacitance parameter in the hysteresis voltage model so that the voltage response value matches the value of the voltage parameter obtained in the pulse response measurement process, wherein the hysteresis voltage model includes the resistance parameter whose value was determined in the resistance parameter setting process.

[0010] According to the parameter fitting program, parameter fitting method, and parameter fitting device of the present invention, the parameters of a hysteresis voltage model can be fitted, so that the hysteresis characteristics can be estimated with high accuracy through a preset current pulse and simple calculation.

[0011] The above and other objects, features and advantages of this disclosure will be more fully understood from the following detailed description and accompanying drawings. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating the parameter fitting process involved in Implementation Method 1.

[0013] Figure 2 This is a diagram illustrating an example of the hysteresis voltage model involved in Implementation Method 1.

[0014] Figure 3 This is a diagram of an example of the measured value of the impulse response waveform involved in Implementation Method 1.

[0015] Figure 4 This is a diagram of an example of the impulse response waveform calculated using the hysteresis voltage model according to Implementation Method 1. Detailed Implementation

[0016] For clarity, the following descriptions and figures have been omitted and simplified as appropriate. Furthermore, the elements described in the figures as functional blocks performing various processes are either hardware components, which may be constructed by a central processing unit (CPU), memory, or other circuitry, or software components, which may be implemented through programs loaded into memory. Therefore, those skilled in the art will understand that these functional blocks can be implemented solely by hardware, solely by software, or in various combinations thereof, and are not limited to any one particular method. Additionally, in the figures, the same elements are labeled with the same reference numerals, and repeated descriptions have been omitted as necessary.

[0017] Additionally, the aforementioned program includes a set of instructions (or software code) that, when read into a computer, causes the computer to perform one or more functions described in the implementation scheme. The program may also be stored on a non-transitory computer-readable medium or tangible storage medium. By way of example, and not limitation, computer-readable media or tangible storage media include: random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, CD-ROM, digital versatile disc (DVD), Blu-ray disc or other optical disc storage, magnetic tape cassette, magnetic tape, disk storage, or other magnetic storage devices. The program may also be transmitted on a transient computer-readable medium or communication medium. By way of example, and not limitation, transient computer-readable media or communication media include: electrical, optical, acoustic, or other forms of propagation signals.

[0018] Implementation Method 1 The parameter fitting processing method described in Embodiment 1 is performed by executing a parameter fitting program in a control device (e.g., an Electronic Control Unit, ECU) that controls the secondary battery. This control device is, for example, a computer, which has a memory and an arithmetic unit that uses the memory to perform various main computational processes according to the program. In other words, the control device includes a parameter fitting device that performs the parameter fitting processing described below. Furthermore, the parameter fitting processing described in Embodiment 1 is performed before the operation of the secondary battery begins. Additionally, the parameter fitting processing described in Embodiment 1 can be appropriately implemented during the operation of the secondary battery.

[0019] First, the parameter fitting process involved in Implementation 1 fits the parameters of the component hysteresis voltage model, which estimates the hysteresis of the secondary battery's output voltage. Figure 1 A flowchart illustrating the parameter fitting process involved in Embodiment 1 is shown.

[0020] like Figure 1 As shown, in the parameter fitting process involved in Implementation Method 1, a model building process is first performed, which defines a hysteresis voltage model that constitutes an equivalent circuit using resistance and capacitance parameters (step S1). Here, Figure 2 A diagram illustrating an example of the hysteresis voltage model involved in Embodiment 1 is shown.

[0021] like Figure 2 As shown, the hysteresis voltage model described below is one of the numerical computational models incorporated into the state estimation model 10. The state estimation model 10 includes a battery model 11 and a hysteresis voltage model 12. The battery model 11 is a numerical computational model that estimates the internal states of the secondary battery of the monitored object, such as the electrochemical changes of the electrodes, through computation.

[0022] Furthermore, secondary batteries exhibit hysteresis, meaning that a difference in output voltage occurs before and after charging and discharging. This voltage difference is sometimes referred to as hysteresis voltage. Hysteresis voltage model 12 is a numerical calculation model for the hysteresis voltage generated by the hysteresis characteristic. For example... Figure 2 In the example shown, the hysteresis voltage model 12 is connected in parallel with the battery model 11. Furthermore, in Figure 2 In the example shown, hysteresis voltage model 12 illustrates an equivalent circuit formed by a resistor parameter R and a capacitor parameter C connected in parallel. In hysteresis voltage model 12, the value of the hysteresis voltage Vhy is calculated for a given current signal Ihy. The current signal Ihys and the hysteresis voltage Vhys will be described in detail later.

[0023] See Figure 1 In the parameter fitting process described in Embodiment 1, a pulse response measurement process is performed. This pulse response measurement process obtains at least one voltage parameter as a fitting target from the measured value of the pulse voltage of the secondary battery obtained by applying a pulse current to the secondary battery (step S2). Here, Figure 3 A schematic diagram showing an example of the measured value of the impulse response waveform involved in Embodiment 1.

[0024] Figure 3 The example shown measures the output voltage of a secondary battery while varying the pulse width of the charge / discharge pulse current by increasing the pulse width over time. In the following explanation, the period in the current pulse signal where the current value is zero is referred to as the rest period. Furthermore, in Figure 3 The example shown illustrates voltage parameters including: the initial voltage Vini during the period when the secondary battery is not being charged or discharged; the output voltage Vchmax after charging, which is the output voltage of the secondary battery at the end of the current zero period after the pulse current becomes a state simulating charging (hereinafter referred to as the post-charge rest period Tres_ch); and the output voltage Vdchmax after discharging, which is the output voltage of the secondary battery at the end of the current zero period after the pulse current becomes a state simulating discharging (hereinafter referred to as the post-discharge rest period Tres_dch).

[0025] like Figure 3 As shown, in a secondary battery, when the pulse width Tch of the charging pulse current and the pulse width Tdch of the discharging pulse current reach a certain width or more, the post-charging output voltage Vchmax and the post-discharging output voltage Vdchmax of the secondary battery after charging will converge to an almost constant voltage. Here, the post-charging output voltage Vchmax and the post-discharging output voltage Vdchmax are the voltages at the end of the post-charging pause Tres_ch and the post-discharging pause Tres_dch, respectively, after the application of the charging or discharging pulse current. Furthermore, in... Figure 3 In the diagram, the open circuit voltage (OCV) of the secondary battery after a sufficient time has elapsed following the completion of the charging and discharging operation is shown as the initial voltage Vini.

[0026] In the parameter fitting process involved in Implementation 1, the values ​​of the resistance parameter R and the capacitance parameter C of the hysteresis voltage model 12 are determined so that numerical calculations can be performed. Figure 3 The values ​​of the output voltage Vchmax after charging and the output voltage Vdchmax after discharging. In the parameter fitting process involved in Implementation Method 1, in Figure 1 In step S3, the value of the resistance parameter R is determined. Figure 1 In step S4, the value of the capacitor parameter C is determined.

[0027] In step S3, a resistance parameter setting process is performed. This process calculates the resistance parameter value based on the voltage parameters obtained in step S2 (e.g., the output voltage Vchmax after charging and the output voltage Vdchmax after discharging). Figure 4 A schematic diagram illustrating an example of the impulse response waveform calculated using the hysteresis voltage model according to Embodiment 1 is shown. Figure 4 In the figure, the current signal Ihy and the hysteresis voltage Vhy of the hysteresis voltage model 12 are shown. Figure 3 The waveform of one cycle of the charge / discharge cycle of the pulse current is shown. Figure 4As shown, when a current signal Ihys is provided to the hysteresis voltage model 12, it can be seen that the voltage changes from the resistance value of the resistance parameter R and the capacitance value of the capacitance parameter C to the voltage determined by the resistance parameter R and the current signal Ihys under the time constant.

[0028] Furthermore, in the hysteresis voltage model 12, during the period when the current signal Ihys is below a specific current value (e.g., 1A), the capacitance parameter C is set to a capacitance value much larger than the estimated value of the capacitance parameter C determined in step S4. For example, the capacitance parameter C is set to a capacitance value more than 100 times the estimated value of the capacitance parameter C determined in step S4. For example, during the period when the current signal Ihyfs is below a specific current value, the capacitance parameter C can be set to approximately 1.0e. 12 The magnitude of F. Therefore, in Figure 4 In the example shown, during periods when the current signal Ihyfs is below a specific current value (e.g., when the current signal Ihys is zero), the hysteresis voltage Vhys exhibits a nearly constant waveform. Furthermore, according to... Figure 4 In the example shown, the resistance parameter calculation and capacitance parameter optimization processes use hysteresis voltage model 12 to calculate the estimated post-charge output voltage Vchmax_md corresponding to the post-charge output voltage Vchmax, and the estimated post-discharge output voltage Vdchmax_md corresponding to the post-discharge output voltage Vdchmax. The estimated post-charge output voltage Vchmax_md is the output voltage value of hysteresis voltage model 12 at the end of the rest period Tres_ch following the charging pulse Tch of the current signal Ihys. The estimated post-discharge output voltage Vdchmax_md is the output voltage value of hysteresis voltage model 12 at the end of the rest period Tres_dch following the discharging pulse Tdch of the current signal Ihys.

[0029] In the resistance parameter setting process (step S3), the hysteresis voltage Vhyx output by the hysteresis voltage model 12 when a current signal Ihyx is applied to the resistance parameter R is made to be similar to the resistance parameter R. Figure 3 The resistance parameter R is determined by matching the output voltage Vchmax and the output voltage Vdcmax. For example, the resistance parameter is calculated using formula (1).

[0030] R={(Vchmax-Vini)+(Vini-Vdchmax)} / (2×I)…(1) In formula (1), I is the magnitude of the current signal Ihys. When the current signal Ihyx is set to 1.0A, in the resistance parameter setting process, the average value of the difference between the output voltage Vchmax after charging and the initial voltage Vini, and the difference between the initial voltage Vini and the output voltage Vdchmax after discharging is set as the value of the resistance parameter.

[0031] Next, the capacitor parameter optimization process in step S4 will be described in detail. In the capacitor parameter optimization process, a current signal Ihys, identical to the pulse current, is applied to the hysteresis voltage model 12, which contains the resistance parameter R whose value was determined in the resistance parameter setting process (step S3), to obtain a voltage response value. The value of the capacitor parameter C in the hysteresis voltage model 12 is then optimized so that the voltage response value (e.g., the hysteresis voltage Vhys) matches the values ​​of the voltage parameters obtained in the pulse response measurement process (e.g., the output voltage Vchmax after charging and the output voltage Vdchmax after discharging). More specifically, in the capacitor parameter optimization process (step S4), the hysteresis voltage model is used to calculate the estimated output voltage Vchmax_md corresponding to the output voltage Vchmax after charging, and the estimated output voltage Vdchmax_md corresponding to the output voltage Vdchmax after discharging. In addition, during the capacitor parameter optimization process (step S4), while changing the capacitor parameter C according to the prescribed rules, the capacitor parameter C that minimizes the sum of the difference between the output voltage Vchmax after charging and the estimated output voltage Vchmax_md after charging, and the difference between the output voltage Vdchmax after discharging and the estimated output voltage Vdchmax_md after discharging is explored.

[0032] As an example of this exploration method, one could consider a method that uses a pre-set capacitance value as the initial value of the capacitance parameter as the starting point, and then, based on the rule of the bisection method, changes and increases or decreases the initial capacitance value by a factor of 1 / 2, 3 / 4, or 3 / 5, or an exploration method that increases or decreases the initial capacitance value by a pre-set step size, etc.

[0033] In the hysteresis voltage model 12, by changing the capacitance value, Figure 4 The rise and fall times of the hysteresis voltage Vhys change. For example, when the capacitance value of the capacitor parameter C is large, during the charging pulse Tdc or the discharging pulse Tch, the hysteresis voltage Vhys may fail to reach the estimated output voltage Vchmax_md after charging (or the estimated output voltage Vdchmax_md after discharging). Therefore, in the capacitor parameter optimization process, it is preferable to set the initial capacitance value to be several times larger than the expected final value, and then gradually decrease the capacitance value while setting the capacitor parameter C to the maximum capacitance value among the capacitance values ​​that allow the hysteresis voltage Vhys to reach the estimated output voltage Vchmax_md after charging (or the estimated output voltage Vdcmax_md after discharging), regardless of the length of the charging pulse Tch and the discharging pulse Tdc.

[0034] It should be noted that, in Figure 4In this process, the charging pulse width Tch and the discharging pulse width Tdcch are set to the same type, but by making the pulse current and current signal Ihy used in the measurement in step S2 contain multiple pulse modes with different pulse widths and pulse periods, the capacitance value of the capacitance parameter C of the hysteresis voltage of the actual secondary battery can be set so that the hysteresis voltage model 12 can more accurately estimate the capacitance value of the capacitance parameter C of the actual secondary battery.

[0035] Furthermore, when the pulse current and current signal Ihy used in the measurement in step S2 contain multiple pulse modes, for each combination of pulse width and pulse period, the optimized output voltage after charging and the output voltage after discharging are set, and for each combination of output voltage after charging and the output voltage after discharging, the capacitor parameter C that minimizes the difference between the estimated output voltage after charging Vchmax_md and the estimated output voltage after discharging Vdchmax_md is calculated.

[0036] According to the above description, by using the parameter fitting process involved in Implementation 1, a hysteresis voltage model 12 that can estimate the hysteresis voltage with high estimation accuracy can be obtained by simple calculation without using processing that requires high computing power, such as processing that uses Kalman filtering or processing that uses mole fraction.

[0037] Furthermore, in the parameter fitting process involved in Implementation 1, the parameters in the hysteresis voltage model 12 are determined according to the hysteresis voltage characteristics known from the measured values ​​of the secondary battery. Therefore, a high-precision hysteresis voltage model 12 can also be constructed for secondary batteries with unknown internal parameters (e.g., secondary batteries with unknown properties).

[0038] As will be apparent from the described disclosure, there are various variations in the implementation of this disclosure. These variations should not be considered a departure from the spirit and scope of this disclosure, and all such modifications, which would be obvious to those skilled in the art, are intended to be included within the scope of the appended claims.

Claims

1. A computer-readable medium storing a parameter fitting program that fits parameters for constructing a hysteresis voltage model, the hysteresis voltage model being connected in parallel with a battery model for estimating the internal state of a secondary battery, and estimating the hysteresis of the output voltage, the parameter fitting program causing a computer to execute: The model building process defines a hysteresis voltage model that uses resistance and capacitance parameters to construct an equivalent circuit. The pulse response measurement process obtains at least one voltage parameter as a fitting target from the measured value of the pulse voltage of the secondary battery obtained by applying a pulse current to the secondary battery. The resistance parameter setting process calculates the value of the resistance parameter based on the voltage parameter. as well as The capacitor parameter optimization process applies the same current signal as the pulse current to the hysteresis voltage model to obtain a voltage response value, and optimizes the value of the capacitor parameter in the hysteresis voltage model so that the voltage response value matches the value of the voltage parameter obtained in the pulse response measurement process. The hysteresis voltage model includes the resistance parameter whose value was determined in the resistance parameter setting process.

2. The computer-readable medium storing a parameter fitting program according to claim 1, wherein, The pulsed current includes multiple pulse patterns with different pulse widths and pulse periods.

3. The computer-readable medium storing a parameter fitting program according to claim 1, wherein, The hysteresis voltage model is an equivalent circuit in which the resistance parameter and the capacitance parameter are connected in parallel.

4. The computer-readable medium storing a parameter fitting program according to claim 1, wherein, In the capacitor parameter optimization process, a predetermined value is set as the initial value of the capacitor parameter.

5. The computer-readable medium storing a parameter fitting program according to claim 4, wherein, When the current value shown by the current signal is lower than a certain value, the capacitance parameter has a fixed value that is more than 100 times the value of the capacitance parameter after optimization processing.

6. The computer-readable medium storing a parameter fitting program according to claim 1, wherein, The voltage parameters include: the initial voltage during the period when the secondary battery is not being charged or discharged; the output voltage after charging, which is the output voltage of the secondary battery at the end of the current zero period after the pulse current becomes a state simulating charging; and the output voltage after discharging, which is the output voltage of the secondary battery at the end of the current zero period after the pulse current becomes a state simulating discharging.

7. The computer-readable medium storing a parameter fitting program according to claim 6, wherein, In the resistance parameter setting process, the value obtained by dividing the average of the difference between the output voltage after charging and the initial voltage and the difference between the initial voltage and the output voltage after discharging by the magnitude of the pulse current is set as the value of the resistance parameter.

8. The computer-readable medium storing a parameter fitting program according to claim 6, wherein, In the capacitor parameter optimization process, The hysteresis voltage model is used to calculate the estimated output voltage after charging corresponding to the output voltage after charging, and the estimated output voltage after discharging corresponding to the output voltage after discharging. While changing the capacitor parameters according to the prescribed rules, the capacitor parameters that minimize the sum of the difference between the output voltage after charging and the estimated output voltage after charging, and the difference between the output voltage after discharging and the estimated output voltage after discharging.

9. A parameter fitting method, wherein the parameter fitting method is a parameter fitting method for fitting parameters of a constructed hysteresis voltage model, the hysteresis voltage model being connected in parallel with a battery model for estimating the internal state of a secondary battery, and for estimating the hysteresis of the output voltage, the parameter fitting method being executed by computer-based automatic processing: The model building process defines a hysteresis voltage model that uses resistance and capacitance parameters to construct an equivalent circuit. The pulse response measurement process obtains at least one voltage parameter as a fitting target from the measured value of the pulse voltage of the secondary battery obtained by applying a pulse current to the secondary battery. The resistance parameter setting process calculates the value of the resistance parameter based on the voltage parameter. as well as The capacitor parameter optimization process applies the same current signal as the pulse current to the hysteresis voltage model to obtain a voltage response value, and optimizes the value of the capacitor parameter in the hysteresis voltage model so that the voltage response value matches the value of the voltage parameter obtained in the pulse response measurement process. The hysteresis voltage model includes the resistance parameter whose value was determined in the resistance parameter setting process.

10. A parameter fitting device, within a control device for controlling a secondary battery, wherein the parameter fitting device performs fitting processing on parameters for constructing a hysteresis voltage model, the hysteresis voltage model being connected in parallel with a battery model for estimating the internal state of the secondary battery, and estimates the hysteresis of the output voltage; the parameter fitting device comprises: Memory, and The arithmetic unit that performs processing using the memory; The arithmetic unit performs: The model building process defines a hysteresis voltage model that uses resistance and capacitance parameters to construct an equivalent circuit. The pulse response measurement process obtains at least one voltage parameter as a fitting target from the measured value of the pulse voltage of the secondary battery obtained by applying a pulse current to the secondary battery. The resistance parameter setting process calculates the value of the resistance parameter based on the voltage parameter. as well as The capacitor parameter optimization process applies the same current signal as the pulse current to the hysteresis voltage model to obtain a voltage response value, and optimizes the value of the capacitor parameter in the hysteresis voltage model so that the voltage response value matches the value of the voltage parameter obtained in the pulse response measurement process. The hysteresis voltage model includes the resistance parameter whose value was determined in the resistance parameter setting process.

Images