JP6410986B1 - Battery impedance evaluation apparatus and battery impedance evaluation method - Google Patents

Abstract

A battery impedance evaluation apparatus capable of specifying the electrochemical impedance of a battery to be measured with high accuracy is provided.
A battery impedance evaluation apparatus includes a storage unit that holds impedance data Zm (ω), an input unit that acquires time domain current data I (t) and time domain voltage data V (t), The frequency domain voltage is calculated from the frequency domain converter 12 that converts the time domain current data I (t) into the frequency domain current data I (ω) by Fourier transform, and the frequency domain current data I (ω) and the impedance data Zm (ω). A voltage data calculator 15 for calculating data Vm (ω), a time domain converter 16 for converting frequency domain voltage data Vm (ω) to time domain voltage data Vm (t) by inverse Fourier transform, and time domain voltage data The difference calculation unit 17 that calculates the difference ε between Vm (t) and the time domain voltage data V (t) and the impedance data that decreases the difference ε are searched for. And a search unit 18.
[Selection] Figure 1

Description

  The present invention relates to a battery impedance evaluation apparatus and method, and more particularly to a technique for specifying the electrochemical impedance of a battery to be measured with high accuracy.

  Conventionally, various methods for evaluating (hereinafter, also referred to as “analysis”) the electrochemical impedance (hereinafter also simply referred to as “impedance”) of a battery have been proposed (see, for example, Patent Document 1).

  In Patent Document 1, in a battery characteristic evaluation apparatus configured to identify a circuit constant for an equivalent circuit model based on a current-voltage characteristic of a battery, a plurality of arbitrary battery measured current waveform data including an AC frequency component are stored. The current waveform dividing unit that divides and outputs the step function in a minute time interval, the step function, the measured voltage value, and the equivalent circuit model data are input, and the circuit constant of the optimized equivalent circuit model is calculated and output. A technique characterized by including a circuit constant optimization unit is disclosed. Thus, by superimposing the step responses for each step function obtained by dividing the battery actual measurement current waveform data, the voltage response of the battery can be calculated even if the input is an arbitrary current waveform. The equivalent circuit including the (Warburg) impedance can be identified.

Japanese Patent No. 4835757

  However, in the technique of Patent Document 1, the transient response voltage waveform V (t) when an arbitrary current waveform is passed through the impedance Z (s) is obtained by converting the battery measured current waveform data into a step function and using Laplace transform. It has gained. Therefore, there is a problem that it is impossible to identify an equivalent circuit including Warburg impedance expressed in the frequency domain and CPE (Constant Phase Element) as circuit elements.

  Specifically, with the technique of Patent Document 1, only the equivalent circuit in which the parameter Phi indicating the frequency dependency is fixed to 0.5 can be identified for the circuit element of Warburg impedance ([0042] of Patent Document 1). Further, with the technique of Patent Document 1, only an equivalent circuit including pure C (capacitance element) can be identified with respect to the capacitive element, and an equivalent circuit including CPE cannot be identified (FIGS. 4 and 5 of Patent Document 1). That is, the technique of Patent Document 1 has a problem that the electrochemical impedance of the battery to be measured cannot be specified with high accuracy.

  Then, an object of this invention is to provide the battery impedance evaluation apparatus which can specify the electrochemical impedance of a to-be-measured battery with higher precision than before, and its method.

  In order to achieve the above object, a battery impedance evaluation apparatus according to an aspect of the present invention is a battery impedance evaluation apparatus that evaluates the electrochemical impedance of a battery to be measured, and is expressed in the frequency domain for at least one battery. A storage unit that holds impedance data indicating electrochemical impedance, and time domain current data and time domain voltage data, which are time series data of current and voltage, obtained by applying current to the measured battery. The frequency domain conversion unit that converts the time domain current data acquired by the input unit into frequency domain current data that is current data expressed in the frequency domain by Fourier transform, and the frequency domain conversion. And the frequency domain current data converted by the storage unit and the image stored in the storage unit. A voltage data calculation unit that calculates frequency domain voltage data that is voltage data expressed in the frequency domain by calculating the impedance data, and the frequency domain voltage data calculated by the voltage data calculation unit is inverse Fourier transformed. By the conversion, a time domain conversion unit that converts time domain voltage data that is voltage data expressed in the time domain, the time domain voltage data converted by the time domain conversion unit, and the input unit that is acquired by the input unit A difference calculation unit that calculates a difference from the time domain voltage data, and a search unit that searches for impedance data that reduces the difference calculated by the difference calculation unit and outputs the searched impedance data.

  In order to achieve the above object, a battery impedance evaluation method according to an aspect of the present invention is a battery impedance evaluation method for evaluating the electrochemical impedance of a battery to be measured, which is obtained by applying a current to the battery to be measured. An input step for acquiring time domain current data and time domain voltage data, which are time series data of current and voltage, and the time domain current data acquired in the input step are expressed in the frequency domain by Fourier transform. A frequency domain conversion step for converting to frequency domain current data, which is current data, the frequency domain current data converted in the frequency domain conversion step, and at least one battery held in the storage unit in the frequency domain. The impedance data indicating the measured electrochemical impedance Thus, the voltage data calculation step for calculating the frequency domain voltage data, which is the voltage data expressed in the frequency domain, and the frequency domain voltage data calculated in the voltage data calculation step are converted in the time domain by inverse Fourier transform. A time domain conversion step for converting into time domain voltage data which is expressed voltage data, the time domain voltage data converted in the time domain conversion step, and the time domain voltage data acquired in the input step. A difference calculating step for calculating a difference; and a search step for searching for impedance data that minimizes the difference calculated in the difference calculating step and outputting the searched impedance data.

  According to the present invention, there is provided a battery impedance evaluation apparatus and method that can specify the electrochemical impedance of a battery to be measured with higher accuracy than before.

FIG. 1 is a block diagram illustrating a configuration of a battery impedance evaluation apparatus according to an embodiment. FIG. 2 is a diagram illustrating a configuration example of an electrochemical measurement system. FIG. 3 is a flowchart showing an operation example (when an equivalent circuit model is used as reference data) of the battery impedance evaluation apparatus according to the embodiment. FIG. 4 is a flowchart showing an operation example of the battery impedance evaluation apparatus according to the embodiment (when AC impedance data is used as reference data). FIG. 5A is a diagram illustrating an example of a fitting range in a charged state of about several seconds by the battery impedance evaluation apparatus according to the embodiment. FIG. 5B is a diagram illustrating an example of a fitting range in a resting state of about several tens of seconds or more from a charged state by the battery impedance evaluation apparatus according to the embodiment. FIG. 5C is a diagram illustrating an example of a fitting range in a charged state of about several tens of seconds or more by the battery impedance evaluation apparatus according to the embodiment. FIG. 5D is a diagram illustrating an example of a fitting range in a discharge state of about several seconds by the battery impedance evaluation apparatus according to the embodiment. FIG. 5E is a diagram illustrating an example of a fitting range in a rest state of about several tens of seconds or more from a discharge state by the battery impedance evaluation apparatus according to the embodiment. FIG. 5F is a diagram illustrating an example of a fitting range in a discharge state of about several tens of seconds or more by the battery impedance evaluation apparatus according to the embodiment. FIG. 6 is a diagram illustrating an example of an equivalent circuit model for explaining details of fitting by the battery impedance evaluation apparatus according to the embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that each of the embodiments described below shows a specific example of the present invention. The numerical values, shapes, materials, constituent elements, arrangement positions and connection forms of the constituent elements, method steps, order of steps, applied current waveforms, equivalent circuit models, etc. shown in the following embodiments are examples, and It is not intended to limit the invention. In addition, among the constituent elements in the following embodiments, constituent elements that are not described in the independent claims indicating the highest concept of the present invention are described as optional constituent elements. Also, the drawings are not necessarily shown strictly. In each figure, substantially the same configuration is denoted by the same reference numeral, and redundant description may be omitted or simplified.

1. Configuration of Battery Impedance Evaluation Device FIG. 1 is a block diagram showing a configuration of a battery impedance evaluation device 10 according to an embodiment. The battery impedance evaluation apparatus 10 is an apparatus for evaluating the electrochemical impedance of a battery to be measured, and includes an input unit 11, a frequency domain conversion unit 12, a storage unit 13, a voltage data calculation unit 15, a time domain conversion unit 16, and a difference calculation unit. 17 and the search part 18 are provided.

  The input unit 11 includes time domain current data I (t), which is time series data of current flowing through the measured battery, obtained by applying current to the measured battery, and a voltage generated at both ends of the measured battery. Time domain voltage data V (t) which is series data is acquired.

  The time domain current data I (t) and the time domain voltage data V (t) are obtained, for example, by measurement by the electrochemical measurement system 20 shown in the configuration example of FIG. The electrochemical measurement system 20 includes an electrochemical measurement device 21 such as a potentio galvanostat or a charge / discharge control amplifier and a data logger 22. The electrochemical measurement device 21 applies a current having a predetermined waveform to the measured battery 23 and measures the voltage generated at both ends of the measured battery 23 at that time. The applied current and the measured voltage are A / D converted by the data logger 22 and recorded as time domain current data I (t) and time domain voltage data V (t), respectively. The time domain current data I (t) and time domain voltage data V (t) recorded in the data logger 22 are read out by the input unit 11 of the battery impedance evaluation apparatus 10 via an interface such as GPIB.

  The input unit 11 may acquire time domain current data I (t) and time domain voltage data V (t) from an external electrochemical measurement system 20 as shown in FIG. When the time domain current data I (t) and the time domain voltage data V (t) are held in the storage unit 13 included in the device 10, the time domain current data I (t) and the time domain voltage are stored from the storage unit 13. Data V (t) may be acquired. Further, the input unit 11 may include the function of the electrochemical measurement system 20. That is, the input unit 11 generates and acquires time domain current data I (t) and time domain voltage data V (t) by sensing the current and voltage in the measured battery and performing A / D conversion. Also good. The frequency domain converter 12 converts the time domain current data I (t) acquired by the input unit 11 into frequency domain current data I (ω) that is current data expressed in the frequency domain by Fourier transform.

  The storage unit 13 is a memory that holds impedance data Zm (ω) indicating electrochemical impedance expressed in the frequency domain for at least one battery, and is, for example, a hard disk device. The impedance data Zm (ω) held in the storage unit 13 may be an equivalent circuit model composed of at least one circuit element, or obtained by actual measurement such as an electrochemical impedance spectroscopy (EIS) method. AC impedance data that is electrochemical impedance expressed in the frequency domain, or a mixture of them may be used.

The equivalent circuit model may include not only the resistance R and capacitance C of the electric circuit element but also CPE which is an electrochemical impedance equivalent circuit model component, and the Warburg impedance including an arbitrary parameter Phi indicating frequency dependence. May be included. The impedance Z _CPE of _CPE is represented by Z _CPE = 1 / (T · (jω) ^p ). T is the CPE constant and p is the CPE index. Generally, CPE is used when the electric double layer capacity shows frequency dependence. Further, the impedance Z _{W of the} Warburg impedance is represented by Z _W = σ · ω ^−p · (1−j). σ is a constant related to the diffusion condition, and p is a parameter Phi indicating frequency dependence.

  The voltage data calculation unit 15 expresses the frequency domain current data I (ω) converted by the frequency domain conversion unit 12 and the impedance data Zm (ω) held in the storage unit 13 in the frequency domain. The frequency domain voltage data Vm (ω), which is the voltage data thus obtained, is calculated.

  The time domain converter 16 converts the frequency domain voltage data Vm (ω) calculated by the voltage data calculator 15 into time domain voltage data Vm (t), which is voltage data expressed in the time domain, by inverse Fourier transform. Convert.

  The difference calculation unit 17 calculates a difference ε between the time domain voltage data Vm (t) converted by the time domain conversion unit 16 and the time domain voltage data V (t) acquired by the input unit 11. More specifically, the difference calculation unit 17 changes the time domain voltage data Vm (t) converted by the time domain conversion unit 16 in a predetermined time and the time domain voltage data V (t) acquired by the input unit 11. Is calculated as a difference ε. The reason for using the change amount ratio is that the time domain voltage data Vm (t) does not contain an appropriate DC offset component due to limitations of this analysis method (battery impedance evaluation method).

  The search unit 18 searches for impedance data Zm ′ (ω) that decreases the difference ε calculated by the difference calculation unit 17 and outputs the searched impedance data Zm ′ (ω). More specifically, when searching for an equivalent circuit model Zm (ω) for a plurality of types of batteries held in the storage unit 13, the search unit 18 selects the plurality of equivalent circuit models Zm (ω). Among them, the equivalent circuit model Zm (ω) of the battery closest to the battery operating condition including at least one of the state of charge (SOC), the number of charge / discharge cycles, the battery temperature, the environmental temperature, and the battery capacity Is selected as the initial value, and the voltage data calculation unit 15 is caused to calculate the frequency domain voltage data Vm (ω) using the selected equivalent circuit model Zm (ω). Then, the search unit 18 identifies or changes the constants of the circuit elements constituting the equivalent circuit model Zm (ω) that is the difference calculation target so as to reduce the difference ε calculated by the difference calculation unit 17. The changed equivalent circuit model is output as the searched impedance data Zm ′ (ω).

  Further, when searching for the AC impedance data Zm (ω) for a plurality of types of batteries held in the storage unit 13, the search unit 18 includes the plurality of AC impedance data Zm (ω). , At least two AC impedance data including AC impedance data Zm (ω) of the battery closest to the battery operating condition including at least one of the state of charge of the measured battery, the number of charge / discharge cycles, the battery temperature, the environmental temperature, and the battery capacity Zm (ω) is selected, and the voltage data calculation unit 15, the time domain conversion unit 16, and the difference calculation unit 17 perform processing on each of the selected at least two AC impedance data Zm (ω). The AC impedance data Zm (ω) having the smallest difference ε calculated by the difference calculating unit 17 is used as the search result impedance. And outputs it as Nsu data Zm' (ω).

  The battery impedance evaluation apparatus 10 is realized by, for example, a personal computer (PC) that stores a battery impedance evaluation method program. The PC is a non-volatile memory such as a hard disk or ROM that holds a program, a volatile memory such as a RAM, a processor that executes the program, various input / output circuits including a communication interface that communicates with an external device, a peripheral device such as a keyboard and a display. Etc. The input unit 11, the frequency domain conversion unit 12, the voltage data calculation unit 15, the time domain conversion unit 16, the difference calculation unit 17, and the search unit 18 of the battery impedance evaluation apparatus 10 are realized by a processor included in the PC executing a program. Can be done. The storage unit 13 of the battery impedance evaluation device 10 can be realized by a nonvolatile memory included in the PC.

2. Operation of Battery Impedance Evaluation Device Next, an equivalent circuit model is used as reference impedance data for the operation of the battery impedance evaluation device 10 according to the present embodiment configured as described above (that is, the battery impedance evaluation method). A case where AC impedance data is used will be described.

2.1. When using an equivalent circuit model First, a method for calculating impedance data of a battery to be measured when an equivalent circuit model is used as reference data will be described.

  FIG. 3 is a flowchart showing an operation example (when an equivalent circuit model is used as reference data) of the battery impedance evaluation apparatus 10 according to the embodiment. In this case, an equivalent circuit model Zm (ω) for a plurality of types of batteries is held in the storage unit 13 as impedance data Zm (ω), and the impedance of the battery to be measured is calculated using the equivalent circuit model Zm (ω). Data is searched.

  First, the input unit 11 acquires time domain current data I (t) and time domain voltage data V (t) obtained by applying a current to the battery under measurement (S10). Here, the time domain current data I (t) and the time domain voltage data V (t) are simultaneous (same sampling clock) and at a constant time interval Δt (sec) (= 1 / Fs; Fs is a sampling frequency (Hz). ) Is time-series data sampled in The time domain current data I (t) and the time domain voltage data V (t) are preferably A / D converted with high level resolution. In particular, for the time domain voltage data V (t), the level resolution greatly affects the accuracy in battery impedance evaluation. Further, the time domain current data I (t) and the time domain voltage data V (t) preferably have a start point and an end point of 0 or substantially the same value (difference << end point level) in the analysis range time T (sec). . Further, the analysis range sample number N (= T / Δt) is preferably a power of two. Details of the analysis range will be described later in “3. Details of analysis range of time series data”.

  Next, the frequency domain transform unit 12 transforms the time domain current data I (t) acquired by the input unit 11 into frequency domain current data I (ω) by Fourier transform according to the following formula 1 (S11). .

  I (ω) = FFT {I (t)} Equation 1

  Here, FFT {} indicates fast Fourier transform algorithm processing of DFT (Discrete Fourier Transform). One feature of this battery impedance evaluation method is that it includes time domain current data I (t). In an actual measurement environment, the slew rate (ΔI / Δt), which is a transient characteristic of the output amplifier at the start of charge / discharge control, is finite. In particular, in a charger / discharger for a large capacity battery, this slew rate may be limited. In this battery impedance evaluation method, the result including the output amplifier characteristics and the like is obtained by using time domain current data I (t) obtained by actual measurement instead of the ideal waveform of current.

  And the search part 18 is charge state of a to-be-measured battery, the number of charge / discharge cycles, battery temperature, environmental temperature, and battery among the equivalent circuit models Zm ((omega)) about several types of battery hold | maintained at the memory | storage part 13. The battery equivalent circuit model Zm (ω) closest to the battery operating condition including at least one of the capacities is selected as an initial value (S12). Here, the frequency resolution Δf of the equivalent circuit model Zm (ω) is Δf = 1 / (Δt · N), which is a constant value. The equivalent circuit model Zm (ω) corresponds not only to the resistance R and capacitance C of the electric circuit elements, but also to the electrochemical impedance equivalent circuit model component CPE, the Warburg impedance including an arbitrary parameter Phi indicating frequency dependence, and the like. . In the calculation, f = 0 is included for convenience.

  When the number of equivalent circuit models Zm (ω) held in the storage unit 13 is large, the search unit 18 stores all equivalent circuit models Zm (ω) held in the storage unit 13 in advance in 10 stages. Ranks may be classified into the following levels, and the typical value of the closest approximation rank may be set as the initial value. Deep learning may be used for ranking. In this case, the input, the number of intermediate layers, and the output in deep learning are as follows, for example.

Input: Voltage time-series waveform information obtained by normalizing time value and level value (for example, black and white image of horizontal 32 × vertical 32 = 1024 pixels)
Number of intermediate layer layers: 1-2
Output: Rank below 10 levels

  The relationship between the frequency range of the selected equivalent circuit model Zm (ω) and the sampling time series data (I (t) and V (t)) is, for example, as follows.

  That is, the maximum frequency Fmax of the equivalent circuit model Zm (ω) is set to Fs / 2. This maximum frequency Fmax needs to be larger than the frequency Fesr of the equivalent series resistance value (the real value of the impedance where the imaginary value of the impedance is 0) as shown in the following Expression 2.

  Fmax = Fs / 2> = Fesr Equation 2

  Further, the minimum frequency Fmin of the equivalent circuit model Zm (ω) needs to satisfy the following conditions. That is, in the case of an R-CPE parallel circuit component that does not include Warburg impedance, the minimum frequency Fmin needs to be sufficiently smaller than the reciprocal of the relaxation time τ of the R-CPE parallel circuit. Further, in the case of the R-CPE parallel circuit component including the Warburg impedance, the minimum frequency Fmin needs to satisfy that the frequency region of the diffusion component is sufficiently included.

  Next, the voltage data calculation unit 15 reads out the equivalent circuit model Zm (ω) selected by the search unit 18 from the storage unit 13, and reads out the equivalent circuit model Zm (ω) read out and the frequency domain according to the following Equation 3. The frequency domain voltage data Vm (ω) is calculated by computing (specifically, multiplying) the frequency domain current data I (ω) converted by the converter 12 (S13).

  Vm (ω) = Zm (ω) · I (ω) Equation 3

  Then, the time domain conversion unit 16 converts the frequency domain voltage data Vm (ω) calculated by the voltage data calculation unit 15 into the time domain voltage data Vm (t) by inverse Fourier transform according to the following Expression 4. (S14).

  Vm (t) = IFFT {Vm (ω)} Equation 4

  Here, IFFT {} indicates an inverse Fourier transform process.

Next, the difference calculation unit 17 calculates a difference ε between the time domain voltage data Vm (t) converted by the time domain conversion unit 16 and the time domain voltage data V (t) acquired by the input unit 11. (S15). More specifically, the difference calculation unit 17 changes the time domain voltage data Vm (t) converted by the time domain conversion unit 16 in a predetermined time (Vm _{(k + 1)} −Vm _(k) ) according to the following formula 5. And the ratio of the amount of change (V _{(k + 1)} −V _(k) ) in the predetermined time of the time domain voltage data V (t) acquired by the input unit 11 as the difference ε.

ε = Σabs {(Vm _{(k + 1)} −Vm _(k) ) / (V _{(k + 1)} −V _(k) )} Equation 5

  Here, abs {} indicates an absolute value.

  The reciprocal of the difference ε indicates the degree of approximation between the battery data (V (t)) and the model characteristics (Vm (t)). Note that V (t) applied to the calculation of Equation 5 is subjected to filtering processing such as suitable smoothing or moving average except in the transient response region (that is, near the control current change). The purpose of applying the filtering process is noise removal, and care must be taken so that the battery characteristics are not reduced.

  And the difference calculation part 17 determines whether the calculated difference (epsilon) is below a predetermined value (S16). The predetermined value is arbitrarily set in consideration of the accuracy required for the battery impedance evaluation method and the allowable search time.

  As a result, when it is determined that the calculated difference ε is not less than or equal to the predetermined value (No in S16), the search unit 18 calculates the difference so as to reduce the difference ε calculated by the difference calculation unit 17. The constants of the circuit elements constituting the target equivalent circuit model Zm (ω) are changed, that is, the equivalent circuit model Zm (ω) is identified (hereinafter, this identification is also referred to as “fitting”) (S17). Then, the processes in steps S13 to S16 are repeated again. Steps S13 to S17 are repeated until it is determined in step S16 that the difference ε is equal to or less than a predetermined value. Details of the fitting will be described later in “4. Details of fitting”.

  On the other hand, when it is determined that the difference ε calculated in step S16 is equal to or smaller than the predetermined value (Yes in S16), the search unit 18 uses the equivalent circuit model Zm (ω) that is the last difference calculation target. That is, the equivalent circuit model Zm (ω) having the highest degree of approximation is output as the searched impedance data Zm ′ (ω) (S18).

  In this way, calculation of the impedance data of the battery to be measured when the equivalent circuit model is used as the reference data is completed.

2.2. When Using AC Impedance Data Next, a method for calculating impedance data of a battery to be measured when using AC impedance data as reference data will be described.

  FIG. 4 is a flowchart illustrating an operation example (when AC impedance data is used as reference data) of the battery impedance evaluation apparatus 10 according to the embodiment. In this case, the storage unit 13 holds a plurality of AC impedance data (that is, electrochemical impedance expressed in the frequency domain obtained by actual measurement) Zm (ω) as impedance data Zm (ω). The impedance data of the battery to be measured is searched using the AC impedance data Zm (ω). In this figure, the same steps as those in FIG. 3 are denoted by the same reference numerals as those in FIG.

  The time domain current data I (t) and time domain voltage data V (t) of the measured battery are acquired by the input unit 11 (S10), and the time domain current data I (t) is converted by the frequency domain conversion unit 12 into the frequency domain current. When converted to data I (ω) (S11), the search unit 18 among the plurality of AC impedance data Zm (ω) held in the storage unit 13, the state of charge of the battery to be measured, the number of charge / discharge cycles, At least two AC impedance data Zm (ω) including AC impedance data Zm (ω) of the closest battery for battery operating conditions including at least one of battery temperature, environmental temperature and battery capacity are selected, and at least two selected For each of the AC impedance data Zm (ω), the process by the voltage data calculation unit 15 (S13) and the time domain conversion unit 16 Processing (S14) and processing by the difference calculation unit 17 (S15) are performed (loop A; S20 to S21).

  Then, the search unit 18 uses the AC impedance data Zm (ω) used for calculating the smallest difference ε among at least two differences ε calculated by the difference calculation unit 17 as impedance data Zm ′ (ω ) Is output (S22).

  In this way, calculation of impedance data of the battery to be measured in the case where AC impedance data is used as reference data is completed.

3. Details of Analysis Range of Time Series Data Next, details of the analysis range of the time series data (I (t) and V (t)) will be described.

  When acquiring the time series data (I (t) and V (t)), the input unit 11 extracts an analysis applicable region from the acquired time series data (I (t) and V (t)), To analyze.

  The input unit 11 has (1) a state where current application is suspended, (2) a state where constant current is applied, and (3) current application is suspended for the battery under measurement. The time series data of the current obtained when the current application pattern is changed to the state in this order is acquired as the time domain current data I (t). In other words, current application patterns (that is, current control patterns) to the battery under measurement for obtaining I (t) include (1) pause (current value: OCV), (2) constant current (constant during charging). The pattern changes in the order of current value: + CC, constant current value during discharge: -CC), and (3) rest (current value: OCV). The analysis range in “(2) constant current” is after the transition from the resting state to the constant current control, and the analysis range in “(3) resting” is after the transition from the constant current control state to the resting state. The time domain voltage data Vm (t) obtained from this battery impedance evaluation method is the same in each state in which the behavior of the measured battery is different, such as “(2) constant current” and “(3) rest”. The voltage response waveform obtained from the model characteristics (that is, impedance data).

  At this time, the following (a) to (c) are mentioned as points to be noted.

  (A) It is preferable to set a resting state longer than the analysis time before analysis. It can be expected that a more accurate result is obtained in a battery state that is relatively stable in analysis time, that is, in a state where there is no sudden characteristic change.

  (B) It is preferable that the constant current output level to the battery to be measured does not become a heavy load on the battery and does not give a sudden characteristic change (that is, within a range in which IV linearity is maintained).

  (C) The constant current output time is preferably not set to an extremely short time that promotes the deterioration of the battery under measurement. In addition, when the model characteristics (that is, impedance data) greatly change due to charging and discharging within the analysis range, this battery impedance evaluation method is not applicable. Therefore, it is preferable that the time length of charging and discharging is sufficiently examined in advance for each type of battery to be measured.

  By the way, in an actual battery, it is assumed that the time change of a characteristic generate | occur | produces by hold | maintaining each said state. This is because the SOC of the battery changes due to constant current control (charging or discharging).

  Therefore, in the fitting process in this battery impedance evaluation method, Vm (t) only in the “(2) constant current” state is set as the fitting range, so that the model characteristics (that is, impedance) at the time of transition to the charge or discharge state are obtained. Data). Similarly, by setting Vm (t) in only the “(3) pause” state as the fitting range, it is possible to obtain model characteristics (that is, impedance data) when shifting from charge or discharge to the pause state. Furthermore, if the time of “(2) constant current” is long, it is assumed that the battery characteristics change during that time. By further limiting the analysis range during the time of “(2) constant current” (immediately after the start of charging, after a specified time after the start of charging, etc.), each characteristic that changes with time can be obtained.

  If the model element is only the R-CPE parallel circuit element (when the Warburg impedance is not included), Vm (t in “(2) constant current” and “(3) pause” regardless of charging and discharging. ) Rise waveform and fall waveform are symmetrical.

3.1. Current Control Waveform and Fitting Range Pattern Next, an example of the current control waveform and the fitting range in the analysis range of the present battery impedance evaluation method is shown.

3.1.1. FIG. 5A is a diagram illustrating an example of a fitting range in a charged state of about several seconds by the battery impedance evaluation apparatus 10 according to the present embodiment.

  In this case, the contribution of R-CPE is high below the time constant region of R-CPE. In the subsequent time, the contribution of the diffusion component is high. Further, for example, when the region where the diffusion behavior is rate-limiting is 0.5 Hz or less, diffusion occurs in Vm (t) within a charging time of about 4 seconds (twice the reciprocal (s) of the frequency 0.5 Hz) or more. Includes the effects of ingredients.

3.1.2. FIG. 5B shows a fitting range in a dormant state of about several tens of seconds or more from the charged state by the battery impedance evaluation apparatus 10 according to the present embodiment. It is a figure which shows an example.

  In this case, since the diffusion region can be fitted, the Warburg impedance can be derived with relatively high accuracy. However, in the equivalent circuit model acquired by the AC impedance method, even when the contribution of the Warburg impedance is large, it is considered that the following V (t) may be obtained.

  -As the resting state continues, the voltage response waveform gradually approaches a constant state within the analysis range.

  -Due to the influence of the battery state before analysis and the state of "(2) constant current", in particular, the behavior of the diffusion region remains (the voltage change direction (up / down) is opposite to the calculation result).

  Therefore, when this fitting range is applied, the time value of the range in the second half of the analysis range of V (t) is also important information.

3.1.3. FIG. 5C is a diagram illustrating an example of a fitting range in a charged state of about several tens of seconds or more by the battery impedance evaluation apparatus 10 according to the present embodiment. .

  In this case, each characteristic can be obtained by dividing the charging range.

3.1.4. FIG. 5D is a diagram showing an example of a fitting range in a discharge state of about several seconds by the battery impedance evaluation apparatus 10 according to the present embodiment.

  In this case, the above 3.1.1. The same can be said.

3.1.5. FIG. 5E shows a fitting range in a rest state of about several tens of seconds or more from the discharge state by the battery impedance evaluation apparatus 10 according to the present embodiment. It is a figure which shows an example.

  In this case, the above 3.1.2. The same can be said.

3.1.6. FIG. 5F is a diagram showing an example of a fitting range in a discharge state of about several tens of seconds or more by the battery impedance evaluation apparatus 10 according to the present embodiment. .

  In this case, each characteristic can be obtained by dividing the range during discharge.

4). Details of Fitting FIG. 6 is a diagram illustrating an example of an equivalent circuit model for explaining details of fitting by the battery impedance evaluation apparatus 10 according to the present embodiment. The equivalent circuit model shown in this figure is a circuit in which a resistor Rs, an R1-CPE1 parallel circuit, and an R2-CPE2 parallel circuit including a Warburg impedance Ws1 are connected in series in this order as circuit elements.

  Fitting using the equivalent circuit model shown in this figure is performed according to the following policy. That is, starting from the resistor Rs, the R-CPE parallel circuit that is a higher frequency region is obtained in order. At the time of deriving the subsequent R-CPE parallel circuit, the already derived component value is fixed or the change range is narrowed. If the relaxation time τ of the R-CPE parallel circuit is obtained in advance, circuit elements are derived within that time range.

  More specifically, when the approximate relaxation time according to each condition (SOC, temperature, number of cycles, etc.) is known, fitting can be applied to the relaxation time range.

Specifically, fitting is performed according to the following procedure.
(1) The resistance Rs is obtained from Rs = ΔV / ΔI, where ΔI is the change due to the rise or fall of I (t) and ΔV is the change in V (t) corresponding to ΔI.
(2) The resistance Rs obtained in the above (1) is set as a fixed value, and the fitting process of Vm (t) to V (t) is performed using the voltage change amount up to the relaxation time τ1 of the R1-CPE1 parallel circuit. Thus, R1 and CPE1 are identified.
(3) Using Rs, R1, and CPE1 obtained in (1) and (2) above as fixed values and using the voltage change amount up to the relaxation time τ2 of the R2-CPE2 parallel circuit, V (t) of Vm (t) ) To identify R2 and CPE2.
(4) Rs, R1, CPE1, R2, and CPE2 obtained in the above (1) to (3) are fixed values, and the voltage change amount of the entire analysis range is used to change Vm (t) to V (t). The Warburg impedance Ws1 is identified by performing the fitting process.

  As described above, the battery impedance evaluation apparatus 10 according to the present embodiment is an apparatus for evaluating the electrochemical impedance of a battery to be measured, and shows the electrochemical impedance expressed in the frequency domain for at least one battery. Time domain current data I (t) and time domain voltage data, which are time series data of current and voltage, obtained by applying a current to the battery to be measured and the storage unit 13 holding the impedance data Zm (ω). The input unit 11 that acquires V (t), and the time domain current data I (t) acquired by the input unit 11 is subjected to Fourier transform, and frequency domain current data I (ω that is current data expressed in the frequency domain is obtained. ), The frequency domain current data I (ω) converted by the frequency domain converter 12, and the storage unit 13. A voltage data calculation unit 15 that calculates frequency domain voltage data Vm (ω) that is voltage data expressed in the frequency domain by calculating the held impedance data Zm (ω), and a voltage data calculation unit 15 A time domain converter 16 that converts the calculated frequency domain voltage data Vm (ω) into time domain voltage data Vm (t) that is voltage data expressed in the time domain by inverse Fourier transform; and a time domain converter 16 is calculated by the difference calculation unit 17 that calculates the difference ε between the time domain voltage data Vm (t) converted at 16 and the time domain voltage data V (t) acquired by the input unit 11, and the difference calculation unit 17. A search unit 18 that searches for impedance data that reduces the difference ε and outputs the searched impedance data.

  Here, when the impedance data Zm (ω) held in the storage unit 13 is an equivalent circuit model composed of circuit elements, the search unit 18 configures an equivalent circuit model so as to reduce the difference ε. The circuit element constant to be changed is changed, and the changed equivalent circuit model is output as the searched impedance data Zm ′ (ω). The equivalent circuit model may include CPE as a circuit element. The equivalent circuit model may include a Warburg impedance including an arbitrary parameter Phi (φ) as a circuit element.

  Thus, according to the battery impedance evaluation apparatus 10 according to the present embodiment, the time domain current data I (t) obtained at the time of applying the current to the battery to be measured is frequency domain current data I (ω) by Fourier transform. The frequency domain voltage data Vm (ω) is calculated by calculating the impedance data Zm (ω) stored in the storage unit 13 and the inverse frequency-transformed frequency domain voltage data Vm (ω). Impedance data for reducing the difference ε between the converted time domain voltage data Vm (t) and the time domain voltage data V (t) acquired at the input unit 11. Is searched. That is, calculation with the impedance data Zm (ω) in the frequency domain is used, and the impedance data of the battery to be measured is searched so as to minimize the voltage data difference in the time domain.

  Therefore, according to the battery impedance evaluation apparatus 10 according to the present embodiment, for example, the Warburg impedance including an arbitrary parameter Phi (φ) is different from the conventional technique using the Laplace transform by converting the battery actual measurement current waveform data into a step function. Can be identified as a circuit element, and an equivalent circuit including CPE as well as resistance R and capacitance C can be identified, and the electrochemical impedance of the battery to be measured is more accurate than before. Identified.

  Further, when the impedance data Zm (ω) for a plurality of types of batteries is held in the storage unit 13, the search unit 18 at least of the plurality of impedance data Zm (ω) held in the storage unit 13. The impedance data Zm (ω) having the smallest difference ε calculated by the difference calculating unit 17 is obtained by causing the voltage data calculating unit 15, the time domain converting unit 16, and the difference calculating unit 17 to perform processing for each of the two. Explore.

  Thereby, when the impedance data Zm (ω) for a plurality of types of batteries is held in the storage unit 13, the voltage difference in the time domain is determined from at least two of the plurality of impedance data Zm (ω). Since the minimum impedance data is searched, the impedance data closest to the electrochemical impedance of the battery to be measured is searched from a plurality of reference impedance data Zm (ω).

  Further, even if the plurality of impedance data Zm (ω) held in the storage unit 13 includes AC impedance data Zm (ω), which is electrochemical impedance expressed in the frequency domain, obtained by actual measurement. Good.

  Thereby, the electrochemical impedance of the battery to be measured is specified using the AC impedance data obtained by actual measurement for various batteries.

  At this time, the search unit 18 operates the battery including at least one of the charge state, the number of charge / discharge cycles, the battery temperature, the environmental temperature, and the battery capacity among the plurality of impedance data Zm (ω) held in the storage unit 13. For the condition, the impedance data Zm (ω) of the battery closest to the battery to be measured is selected as an initial value, and the voltage data calculation unit 15, the time domain conversion unit 16, and the difference calculation are selected for the selected impedance data Zm (ω). The processing in the unit 17 is performed, and the impedance data Zm (ω) having the smallest difference ε calculated by the difference calculating unit 17 is searched.

  As a result, impedance data is selected as an initial value for a battery having battery characteristics close to those of the battery to be measured, so that the search time for the impedance data of the battery to be measured is shortened.

  In addition, the input unit 11 applies a current application pattern in this order to the battery under measurement in a state where current application is suspended, a state where constant current is applied, and a state where current application is suspended. Is obtained as time domain current data I (t).

  Thereby, fitting for the R-CPE parallel circuit in the equivalent circuit model by using time domain current data I (t) or the like in the application state (charge state or discharge state) following the rest state of current application, By using the time domain current data I (t) in the rest state following the application state (charge state or discharge state), it is possible to fit the Warburg impedance in the equivalent circuit model.

  Further, the difference calculation unit 17 changes the amount of time domain voltage data Vm (t) converted by the time domain conversion unit 16 in a predetermined time and the predetermined time domain voltage data V (t) acquired by the input unit 11. The ratio with the amount of change in time is calculated as the difference ε.

  As a result, the difference ε is calculated from the change in the time domain voltage data Vm (t) and V (t), so that the time domain voltage data Vm (t) is appropriately included even though the DC offset component is not correctly included. The difference ε is calculated.

  The battery impedance evaluation method according to the present embodiment is a method for evaluating the electrochemical impedance of a battery to be measured, and is time-series data of current and voltage obtained by applying a current to the battery to be measured. Input step S10 for acquiring time domain current data I (t) and time domain voltage data V (t), and time domain current data I (t) acquired in input step S10 are expressed in the frequency domain by Fourier transform. Frequency domain conversion step S11 for converting the frequency data into the frequency domain current data I (ω), the frequency domain current data I (ω) converted in the frequency domain conversion step S11, and the storage unit 13 , Impedance data Zm (ω indicating the electrochemical impedance expressed in the frequency domain for at least one battery Are calculated, voltage data calculation step S13 for calculating frequency domain voltage data Vm (ω), which is voltage data expressed in the frequency domain, and frequency domain voltage data Vm ( ω) is converted into time domain voltage data Vm (t) which is voltage data expressed in the time domain by inverse Fourier transform, and the time domain voltage data converted in the time domain conversion step S14. Difference calculation step S15 for calculating the difference ε between Vm (t) and the time domain voltage data V (t) acquired in the input step S10, and impedance data for minimizing the difference ε calculated in the difference calculation step S15 And searching for output impedance data (S20 to S22, S16 to S1) 8). The program according to the present embodiment is a program that causes a computer to execute the battery impedance evaluation method.

  With these battery impedance evaluation methods and programs, the battery current measurement current waveform data is converted into a step function to differentiate it from Laplace transform. For example, the battery impedance is identified as an equivalent circuit including a Warburg impedance including an arbitrary parameter Phi as a circuit element. In addition, the equivalent circuit including CPE as well as the resistance R and the capacitance C can be identified, and the electrochemical impedance of the battery to be measured is specified with higher accuracy than before.

  As described above, the battery impedance evaluation apparatus and the battery impedance evaluation method according to the present invention have been described based on the embodiment. However, the present invention is not limited to this embodiment. Without departing from the gist of the present invention, various modifications conceived by those skilled in the art have been made in the present embodiment, and other forms constructed by combining some components in the embodiment are also within the scope of the present invention. Contained within.

  For example, in the above embodiment, as an equivalent circuit model held in the storage unit 13, a circuit in which a resistor Rs, an R1-CPE1 parallel circuit, and an R2-CPE2 parallel circuit including a Warburg impedance are connected in series in this order. Is used, but is not limited to this circuit. A circuit that does not include CPE, a circuit that does not include Warburg impedance, or a circuit having a different connection form may be used.

  In the above embodiment, the case of searching for the impedance of the battery to be measured using an equivalent circuit model (FIG. 3) and the case of searching for the impedance of the battery to be measured using AC impedance data obtained by actual measurement (FIG. 3) 4), these two methods may be selected by an instruction from the user, may be automatically selected according to a predetermined parameter such as a priority order, or both are executed. May be.

  The present invention can be used as a battery impedance evaluation apparatus, in particular, as a battery impedance evaluation apparatus that specifies the electrochemical impedance of a battery to be measured with high accuracy, for example, as a battery impedance evaluation apparatus realized using a personal computer.

DESCRIPTION OF SYMBOLS 10 Battery impedance evaluation apparatus 11 Input part 12 Frequency domain conversion part 13 Memory | storage part 15 Voltage data calculation part 16 Time domain conversion part 17 Difference calculation part 18 Search part 20 Electrochemical measurement system 21 Electrochemical measurement apparatus 22 Data logger 23 Battery to be measured

Claims (11)

A battery impedance evaluation apparatus for evaluating the electrochemical impedance of a measured battery,
For at least one battery, a storage unit holding impedance data indicating electrochemical impedance expressed in the frequency domain;
An input unit for obtaining time domain current data and time domain voltage data, which are time series data of current and voltage, obtained by applying current to the battery to be measured;
A frequency domain conversion unit that converts the time domain current data acquired by the input unit into frequency domain current data that is current data expressed in the frequency domain by Fourier transform;
By calculating the frequency domain current data converted by the frequency domain conversion unit and the impedance data held in the storage unit, frequency domain voltage data that is voltage data expressed in the frequency domain is calculated. A voltage data calculator;
A time domain converter that converts the frequency domain voltage data calculated by the voltage data calculator to time domain voltage data that is voltage data expressed in the time domain by inverse Fourier transform;
A difference calculation unit that calculates a difference between the time domain voltage data converted by the time domain conversion unit and the time domain voltage data acquired by the input unit;
A battery impedance evaluation apparatus comprising: a search unit that searches for impedance data that reduces the difference calculated by the difference calculation unit and outputs the searched impedance data. The impedance data held in the storage unit is an equivalent circuit model composed of circuit elements,
The said search part changes the constant of the said circuit element which comprises the said equivalent circuit model so that the said difference may be made small, and outputs the said equivalent circuit model after change as the searched said impedance data. Battery impedance evaluation device. The battery impedance evaluation apparatus according to claim 2, wherein the equivalent circuit model includes CPE (Constant Phase Element) as a circuit element. The battery impedance evaluation apparatus according to claim 2, wherein the equivalent circuit model includes a Warburg impedance including an arbitrary parameter Phi as a circuit element. The storage unit stores the impedance data for a plurality of types of batteries,
The search unit causes the voltage data calculation unit, the time domain conversion unit, and the difference calculation unit to perform processing for at least two of the plurality of impedance data held in the storage unit, and The battery impedance evaluation apparatus according to claim 1, wherein impedance data having the smallest difference calculated by the difference calculation unit is searched. The battery impedance evaluation apparatus according to claim 5, wherein the plurality of impedance data held in the storage unit include AC impedance data that is electrochemical impedance expressed in a frequency domain and obtained by actual measurement. The search unit is configured to measure the battery operation condition including at least one of a charge state, a charge / discharge cycle number, a battery temperature, an environmental temperature, and a battery capacity among the plurality of impedance data held in the storage unit. The impedance data of the battery closest to the battery is selected as an initial value, and the voltage data calculation unit, the time domain conversion unit, and the difference calculation unit are processed for the selected impedance data, and the difference calculation is performed. The battery impedance evaluation apparatus according to claim 6, wherein the impedance data having the smallest difference calculated by the unit is searched. The input unit applies a current application pattern in this order to the battery under test in a state where current application is suspended, a state where constant current is applied, and a state where current application is suspended. The battery impedance evaluation apparatus according to any one of claims 1 to 7, wherein time-series data of current obtained when changed is acquired as the time-domain current data. The difference calculation unit is a ratio between a change amount of the time domain voltage data converted by the time domain conversion unit in a predetermined time and a change amount of the time domain voltage data acquired by the input unit in the predetermined time. The battery impedance evaluation apparatus according to claim 1, wherein the battery impedance is calculated as the difference. A battery impedance evaluation method for evaluating the electrochemical impedance of a measured battery,
An input step of obtaining time domain current data and time domain voltage data, which are time series data of current and voltage, obtained by applying current to the battery to be measured;
A frequency domain transforming step for transforming the time domain current data acquired in the input step into frequency domain current data that is current data expressed in the frequency domain by Fourier transform;
By calculating the frequency domain current data converted in the frequency domain conversion step and the impedance data indicating the electrochemical impedance expressed in the frequency domain for at least one battery held in the storage unit, A voltage data calculation step for calculating frequency domain voltage data, which is voltage data represented by
A time domain conversion step of converting the frequency domain voltage data calculated in the voltage data calculation step into time domain voltage data that is voltage data expressed in the time domain by inverse Fourier transform;
A difference calculating step of calculating a difference between the time domain voltage data converted in the time domain conversion step and the time domain voltage data acquired in the input step;
A battery impedance evaluation method comprising: searching for impedance data that minimizes the difference calculated in the difference calculating step, and outputting the searched impedance data.   The program which makes a computer perform the battery impedance evaluation method of Claim 10.

Images