Battery state estimation device

Abstract

The present invention provides a battery state estimation device that does not require synchronization between voltage measurement and current measurement, and can reduce the time required to estimate the battery state. [Solution] The battery state estimation device comprises a current application unit that applies a falling-wavelength current to the battery, a voltage measurement unit that measures the voltage of the battery to which the current has been applied, and a monitoring unit that estimates the state of the battery based on the measured voltage. The monitoring unit converts the transient change of the voltage waveform of the battery measured by the voltage measurement unit when a current is applied by the current application unit into a complex frequency, converts the transient change of the falling-wavelength current into a complex frequency, and estimates the state of the battery by creating a Cole-Cole plot of the battery impedance from the transient change of the voltage waveform converted into a complex frequency and the transient change of the current waveform converted into a complex frequency.

Description

[Technical Field]

[0001] This invention relates to a battery state estimation device. [Background technology]

[0002] Devices have been proposed to estimate the state of a battery using electrochemical impedance spectroscopy (EIS). For example, Patent Document 1 discloses a technique for estimating the state of a battery by applying an AC current to the battery while sweeping the frequency, and obtaining a Cole-Cole plot of the battery's impedance from synchronized voltage and current measurements. Patent Document 2 also discloses a technique for estimating the state of a battery by applying a square wave current to the battery, and obtaining a Cole-Cole plot of the battery's impedance from the Fourier transform of synchronized voltage and current measurements. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2013-250223 [Patent Document 2] Japanese Patent Publication No. 2014-238948 [Overview of the project] [Problems that the invention aims to solve]

[0004] Incidentally, the above technology requires high synchronization between voltage and current measurement. Furthermore, applying low-frequency signals, such as 0.1 Hz, presents the problem of a time-consuming process, requiring 10 seconds per cycle.

[0005] The present invention has been made in view of the above problems, and its objective is to provide a battery state estimation device that can reduce the time required to estimate the state of a battery without requiring synchronization between voltage measurement and current measurement. [Means for solving the problem]

[0006] The battery state estimation device according to the present invention is a battery state estimation device that estimates the state of a battery in which the constants of an equivalent circuit are a solution resistance R1, a resistance component R2 of a negative electrode, and a capacitance component C2 of the negative electrode. The battery state estimation device includes a current application unit that applies a current having a falling waveform such that the difference between the maximum applied current value and the minimum applied current value of the battery is greater than a first predetermined value, a voltage measurement unit that measures the voltage of the battery to which the current is applied by the current application unit, and a monitoring unit that estimates the state of the battery based on the voltage measured by the voltage measurement unit. The current application unit applies a current having a waveform with a time T1 = 1 / ω1×2 from the maximum applied current value to the minimum applied current value. The third frequency ω1 at the time T1 = 1 / ω1×2 is greater than the first frequency ω2 = 1 / R2C2 and less than the second frequency ω0 at which the real axis becomes the value of the solution resistance R1 in the Cole-Cole plot of the impedance of the equivalent circuit. The monitoring unit converts the transient change in the waveform of the voltage of the battery measured by the voltage measurement unit when the current is applied by the current application unit into a complex frequency, creates a falling current waveform such that the difference between the maximum created current value and the minimum created current value is greater than a second predetermined value, and the time T2 from the maximum created current value to the minimum created current value is greater than the time T0 = 1 / ω0×2 and less than the time T1. The monitoring unit converts the transient change in the waveform of the current into a complex frequency, and creates a Cole-Cole plot of the impedance of the battery from the transient change in the waveform of the voltage converted into a complex frequency and the transient change in the waveform of the current converted into a complex frequency, thereby estimating the state of the battery.

Advantages of the Invention

[0007] According to the present invention, the time for estimating the state of the battery can be reduced without requiring synchronization in voltage measurement and current measurement. Further features related to the present invention will become apparent from the description in this specification and the accompanying drawings. Also, problems, configurations, and effects other than those described above will be clarified by the description of the following embodiments.

Brief Description of the Drawings

[0008] [Figure 1] Diagram showing the control system of an electric vehicle and the cloud in the first embodiment. [Figure 2] Functional block diagram showing the battery state estimation device in the first embodiment. [Figure 3] Equivalent circuit diagram of the battery in FIG. 2. [Figure 4] Theoretical Cole-Cole plot of the impedance of the battery in FIG. 3. [Figure 5] Actual Cole-Cole plot of the impedance of the battery in FIG. 3. [Figure 6] Flowchart including the process of estimating the remaining capacity of the battery from the real resistance component of the battery state estimation device in the first embodiment. [Figure 7] (a) is a graph showing the applied current waveform, (b) is a graph showing the measured voltage waveform, and (c) is a graph showing the created current waveform. [Figure 8] (a) is a graph showing the applied current waveform, (b) is a graph showing the measured voltage waveform, and (c) is an enlarged graph of the dashed part of (b). [Figure 9] Diagram showing a method of estimating the remaining capacity of the battery from the real resistance component of the Cole-Cole plot. [Figure 10] Flowchart including the process of estimating the remaining capacity of the battery from the integration of the values on the imaginary axis of the battery state estimation device in the first embodiment. [Figure 11] Diagram showing a method of estimating the remaining capacity of the battery from the integration of the values on the imaginary axis of the Cole-Cole plot. [Figure 12] Flowchart including the process of checking the phase difference between the voltage waveform and the current waveform of the battery state estimation device in the first embodiment. [Figure 13] Graph showing the phase difference between the voltage waveform and the current waveform. [Figure 14] Diagram showing the Cole-Cole plot created when the phase difference between the voltage waveform and the current waveform is within a predetermined range. [Figure 15] Flowchart including the process of adjusting the phase difference between the voltage waveform and the current waveform of the battery state estimation device in the first embodiment. [Figure 16]A functional block diagram showing the battery state estimation device of the second embodiment. [Figure 17] A graph showing the relationship between phase change with respect to frequency and battery temperature. [Figure 18] A flowchart illustrating the operation of the battery state estimation device according to the second embodiment. [Figure 19] A functional block diagram showing a battery state estimation device according to the third embodiment. [Figure 20] A flowchart including a process for converting the second distribution result of the battery state estimation device of the third embodiment into a third distribution result corresponding to the first distribution result. [Figure 21] This figure shows call-call plots for the results of the first, second, and third distributions. [Figure 22] A flowchart including a process to create a fourth distribution result by interpolating the first distribution result in the low frequency band with the third distribution result in the low frequency band of the battery state estimation device of the third embodiment. [Figure 23] A figure showing the call-call plot of the first distribution results. [Figure 24] This figure shows a call-call plot of the results for the fourth distribution. [Figure 25] A functional block diagram showing the battery state estimation device of the fourth embodiment. [Figure 26] A functional block diagram showing a conventional battery state estimation device. [Modes for carrying out the invention]

[0009] Embodiments will be described below with reference to the attached drawings. In the attached drawings, functionally identical elements are indicated by the same numbers. The attached drawings show embodiments in accordance with the principles of this disclosure, but they are for the purpose of understanding this disclosure and are not to be used in any way to restrict the interpretation of this disclosure. The descriptions herein are typical examples and do not limit the claims or applications of this disclosure in any way.

[0010] [First Embodiment] The first embodiment will be described below. As shown in Figure 1, the battery state estimation device 1A of the first embodiment is mounted on an electric vehicle 500. The electric vehicle 500 is, for example, a BEV (Battery Electric Vehicle), a PHEV (Plug-in Hybrid Electric Vehicle), and a HEV (Hybrid Electric Vehicle). The electric vehicle 500 is equipped with a central ECU (Electronic Control Unit) 400, domains 100, 200, 300, a BMU (Battery Management Unit) 5, a plurality of CMUs (Cell Management Units) 9, and a plurality of batteries 10.

[0011] The central ECU 400 controls the overall operation of the electric vehicle 500. Domains 100, 200, and 300 control the operation of individual parts of the electric vehicle 500. Domain 100 estimates the state of the battery 10, including its solution resistance and remaining capacity (SOH: State of Health). The BMU 5 is connected to multiple CMUs 9 and controls the overall operation of the battery 10. Each of the multiple CMUs 9 is connected to each of the multiple batteries 10 and measures the voltage of each battery 10. Specifically, the battery 10 is a cell, which is the smallest unit of a lithium-ion battery.

[0012] As shown in Figure 2, the battery state estimation device 1A of this embodiment includes a current application unit 2A, a voltage measurement unit 3, and a monitoring unit 4A. The current application unit 2A is provided in the BMU 5. The current application unit 2A applies a current with a falling waveform to the battery 10 such that the difference between the maximum applied current and the minimum applied current is greater than a first predetermined value. The current application unit 2A includes a signal generation unit 6, a MOSFET (Metal-Oxide-Semiconductor Field Effect Transistor) 7, and a load resistor 8. Based on a command signal from the monitoring unit 4A, the signal generation unit 6 inputs a signal to the gate of the MOSFET 7 for applying a current with a falling waveform to the battery 10 such that the difference between the maximum applied current and the minimum applied current is greater than a first predetermined value. The load resistor 8 is a variable resistor that adjusts the magnitude of the current applied to the battery 10. The current application unit 2A includes a current source (not shown) that applies current to the MOSFET 7 and the load resistor 8.

[0013] The voltage measurement unit 3 is composed of multiple CMUs 9. The voltage measurement unit 3 measures the voltage of the battery 10 to which current is applied by the current application unit 2A. In this embodiment, however, a unit that measures the current of the battery 10 in synchronization with the voltage of the battery 10 measured by the voltage measurement unit 3 is not essential.

[0014] The monitoring unit 4A consists of a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random-access memory), and HDD (hard disk drive). The CPU (Central Processing Unit) is a form of processor and may be a GPU (Graphics Processing Unit), other semiconductor devices capable of computational processing, or a combination thereof. In the monitoring unit 4A, the CPU (Central Processing Unit) executes computer programs stored in the ROM (Read Only Memory) and RAM (Random-access memory) to realize the functional blocks of the software configuration described below.

[0015] The monitoring unit 4A consists of a current waveform creation unit 11 and a call-call plot creation unit 12 inside the BMU5, and a domain 100 outside the BMU5. The monitoring unit 4A estimates the state of the battery 10 based on the voltage measured by the voltage measurement unit 3. The current waveform creation unit 11 of the monitoring unit 4A creates a calculated current waveform that falls so that the difference between the maximum current value and the minimum current value is greater than a second predetermined value. The current waveform creation unit 11 does not actually generate current and does not apply current to the battery 10. The current waveform creation unit 11 transmits information to the call-call plot creation unit 12, such as the maximum current value, minimum current value, second predetermined value, and the time it takes for the current to fall from the maximum current value to the minimum current value, regarding the current waveform that falls so that the difference between the maximum current value and the minimum current value is greater than a second predetermined value.

[0016] The Cole-Cole plot creation unit 12 of the monitoring unit 4A converts the transient change in the voltage waveform of the battery 10, measured by the voltage measurement unit 3 when current is applied by the current application unit 2A, into a complex frequency. The Cole-Cole plot creation unit 12 of the monitoring unit 4A converts the transient change in the current waveform created by the current waveform creation unit 11 into a complex frequency. The Cole-Cole plot creation unit 12 of the monitoring unit 4A creates a Cole-Cole plot (Nyquist diagram) of the impedance of the battery 10 from the transient change in the voltage waveform converted to a complex frequency and the transient change in the current waveform converted to a complex frequency. Domain 100 estimates the state of the battery 10, such as the solution resistance and remaining capacity, from the Cole-Cole plot of the impedance of the battery 10 created by the Cole-Cole plot creation unit 12.

[0017] Before describing the operation of the battery state estimation device 1A of this embodiment, the equivalent circuit of the battery 10 will be described below. As shown in Figure 3, the equivalent circuit of the battery 10 is composed of a solution resistance R1, a negative electrode resistance component R2 and a negative electrode capacitive component C2 connected in parallel with each other, a positive electrode resistance component R3 and a positive electrode capacitive component C3 connected in parallel with each other, and a diffusion resistance Z0, all connected in series.

[0018] The Cole-Cole plot of the impedance of battery 10 in Figure 3 is theoretically represented as shown in Figure 4. At a second frequency ω0 of about 1 kHz, the real axis represents the value of the solution resistance R1. At a first frequency ω2 = 1 / R2C2, determined by the time constant of the negative electrode of battery 10 at about 20-80 Hz, the imaginary axis represents the value of the capacitance component C2 of the negative electrode. The radius on the real axis of the arc containing the first frequency ω2 represents the value of the resistance component R2 of the negative electrode. At ω3 = 1 / R3C3, determined by the time constant of the positive electrode of battery 10, the imaginary axis represents the value of the capacitance component C3 of the positive electrode. The radius on the real axis of the arc containing ω3 represents the value of the resistance component R3 of the positive electrode. The region with a frequency lower than the arc containing ω3 represents the diffusion resistance Z0. As described above, the battery state estimation device 1A of this embodiment estimates the state of battery 10 where the constants of the equivalent circuit of battery 10 are the solution resistance R1, the resistance component R2 of the negative electrode, and the capacitance component C2 of the negative electrode.

[0019] As shown in Figure 5, in the Cole-Cole plot of the battery 10, which is created by the voltage actually measured by the voltage measurement unit 3 and the current waveform creation unit 11, the arc of the negative electrode and the arc of the positive electrode overlap and appear as a single arc. Hereinafter, in the Cole-Cole plot created by the Cole-Cole plot creation unit 12 of the monitoring unit 4A, the value on the real axis where the imaginary axis is 0 is defined as the current solution resistance R'1. The frequency at which the current solution resistance R'1 is defined is defined as the second frequency ω0. The frequency determined by the time constant of the negative electrode is defined as the first frequency ω2 = 1 / R2C2. The frequency determined by the time constant of the positive electrode is defined as ω3 = 1 / R3C3. In the Cole-Cole plot created by the Cole-Cole plot creation unit 12, the difference between the value on the real axis at the fourth frequency ω4, which is the inflection point of the arc containing the first frequency ω2, and the current solution resistance R'1 is defined as the current resistance component R'2. The current resistance component R'2 represents the sum of the current resistance components of the negative electrode and the positive electrode. The largest value of the imaginary axis on the arc containing the first frequency ω² is R', which is the sum of the current capacitance components of the negative and positive electrodes. j It represents.

[0020] Next, the process of estimating the remaining capacity of the battery 10 from the current resistance component R´2, etc. of the battery state estimation device 1A of the present embodiment will be described. As shown in FIG. 6, the current application unit 2A applies a current having a waveform that falls so that the difference between the maximum applied current value and the minimum applied current value to the battery 10 is greater than the first predetermined value (S001).

[0021] As shown in FIG. 4, let the first frequency ω2 = 1 / R2C2. Let the second frequency ω0 be the value of the solution resistance R1 on the real axis in the Cole-Cole plot of the impedance of the equivalent circuit of the battery 10. As shown in FIG. 7(a), the current application unit 2A applies a current having a waveform such that the time T1 from the maximum applied current value to the fall until the minimum applied current value is T1 = 1 / ω1×2. The third frequency ω1 at time T1 = 1 / ω1×2 is greater than the first frequency ω2 and less than the second frequency ω0 with respect to the first frequency ω2 and the second frequency ω0. That is, 1 / (ω0×2) < T1 < 1 / (ω2×2). The first predetermined value can be, for example, about 10% of the output current of the battery 10. In the present embodiment, the minimum applied current value of the current applied by the current application unit 2A is 0A.

[0022] When a current is applied to the battery 10 by the current application unit 2A, the voltage waveform of the battery 10 as shown in FIG. 7(b) is measured by the voltage measurement unit 3. As shown in FIG. 8(a), the minimum applied current value of the current applied by the current application unit 2A is 0A. That is, the MOSFET 7 is turned off. Therefore, as shown in FIGS. 8(b) and 8(c), the voltage waveform measured by the voltage measurement unit 3 is not affected by the noise due to switching and has a waveform with little noise. Therefore, the measurement accuracy is improved, and it is possible to estimate a small resistance value even on the Cole-Cole plot.

[0023] As shown in FIG. 6, the call call plot creation unit 12 converts the transient change in the waveform of the voltage of the battery 10 measured by the voltage measurement unit 3 when a current is applied by the current application unit 2A into a complex frequency. Specifically, the measured voltage sampled at the period of time T1 as described above is subjected to Fourier transform (S002). The Fourier transform can be performed, for example, by a fast Fourier transform (FFT: Fast Fourier Transform).

[0024] The current waveform creation unit 11 creates a current waveform (S003). As shown in FIG. 7(c), the current waveform created by the current waveform creation unit 11 falls such that the difference from the created current maximum value to the created current minimum value is greater than the second predetermined value, and the time T2 from the created current maximum value to the time when it falls to the created current minimum value is greater than the time T0 = 1 / ω0 × 2 and smaller than the time T1. That is, 1 / (ω0 × 2) < T2 < 1 / (ω1 × 2). The second predetermined value can be, for example, about the same as the first predetermined value. In the present embodiment, the created current minimum value of the current created by the current waveform creation unit 11 is 0 A. As described above, the current waveform creation unit 11 does not actually generate a current, but transmits information regarding the waveform of the current to the call call plot creation unit 12.

[0025] As shown in FIG. 6, the call call plot creation unit 12 converts the transient change in the waveform of the current created by the current waveform creation unit 11 into a complex frequency. Specifically, the created current sampled at the period of time T2 as described above is subjected to Fourier transform (S004). The Fourier transform can be performed, for example, by a fast Fourier transform as in S004.

[0026] The Cole-Cole plot creation unit 12 creates a Cole-Cole plot of the impedance of the battery 10 from the transient changes of the voltage waveform converted to complex frequency and the transient changes of the current waveform converted to complex frequency (S005). The Cole-Cole plot creation unit 12 creates a Cole-Cole plot as shown in Figure 9. As shown in Figures 6 and 9, domain 100 outputs the current solution resistance R'1 (S006). Domain 100 also estimates and outputs the remaining capacity of the battery 10 from the relationship between the current resistance component R'2 and the remaining capacity of the battery 10 or the relationship between the current resistance component R'3 and the remaining capacity of the battery 10 (S006).

[0027] Domain 100 calculates the value on the real axis where the imaginary axis is zero in the Cole-Cole plot created by the Cole-Cole plot creation unit 12 as the current solution resistance R'1. Domain 100 also calculates the difference between the value on the real axis at the fourth frequency ω4, which is the inflection point of the arc containing the first frequency ω2, and the solution resistance R'1 in the Cole-Cole plot created by the Cole-Cole plot creation unit 12 as the current resistance component R'2. Domain 100 also estimates the remaining capacity of the battery from the current resistance component R'2.

[0028] Alternatively, Domain 100 calculates the difference between the real axis value at the first frequency ω2 and the current solution resistance R'1 in the call-call plot created by the call-call plot creation unit 12 as the current resistance component R'3. Domain 100 also estimates the remaining capacity of the battery from the current resistance component R'3. When determining the remaining capacity with a certain degree of accuracy, the remaining capacity is correlated with the current resistance components R'2 and R'3, so Domain 100 can estimate the remaining capacity from only the resistance component on the real axis with a simple calculation.

[0029] The following describes the process by which Domain 100 estimates the remaining battery capacity from the integral of the imaginary axis value in the call-call plot. As shown in S101 to S105 of Figure 10, the call-call plot creation unit 12 performs the same process as described above in S001 to S005 of Figure 6 until the call-call plot is created.

[0030] As shown in Figures 10 and 11, Domain 100 calculates the current solution resistance R'1 (S106). Domain 100 calculates the current resistance component R'2 as the difference between the real axis value at the fourth frequency ω4, which is the inflection point of the arc containing the first frequency ω2 in the Cole-Cole plot created by the Cole-Cole plot creation unit 12, and the current solution resistance R'1. Domain 100 calculates the area S by integrating the imaginary axis value in the Cole-Cole plot created by the Cole-Cole plot creation unit 12 over the interval from the current solution resistance R'1 on the real axis to the sum of the current solution resistance R'1 and the current resistance component R'2 (S106). Domain 100 outputs the current solution resistance R'1 (S107). Domain 100 also estimates and outputs the remaining capacity of the battery 10 from the relationship between the area S and the remaining capacity (S107).

[0031] Domain 100 calculates the difference between the real axis value at the first frequency ω2 and the current solution resistance R'1 in the Coal-Call plot created by the Coal-Call plot creation unit 12 as the current resistance component R'3. Domain 100 calculates the area S by integrating the imaginary axis value in the Coal-Call plot created by the Coal-Call plot creation unit 12 over the interval from the current solution resistance R'1 on the real axis to the sum of the current solution resistance R'1 and the current resistance component R'3 (S106). Domain 100 also estimates and outputs the remaining capacity of the battery 10 from the relationship between the area S and the remaining capacity (S107).

[0032] The remaining capacity is highly correlated with the area S. In the process described above, domain 100 estimates the remaining capacity of the battery from the integral of the imaginary axis values ​​in the Cole-Cole plot. Since the remaining capacity includes the capacity distribution, i.e., the imaginary axis side, the accuracy of estimating the remaining capacity is improved compared to the process that estimates it only from the current resistance component R'2 or R'3.

[0033] The following describes the process of checking the phase difference between the voltage waveform and the current waveform of the battery state estimation device 1A. As shown in S201 to S204 of Figure 12, the process is performed up to the point where the generated current, sampled with a period of time T2, is Fourier transformed, similar to S101 to S104 of Figure 10. The Cole-Cole plot creation unit 12 checks whether the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency is within a third predetermined value in any frequency band (S205). The third predetermined value can be set, for example, in any range from 10° to 30°. After the check in S205 is completed, as shown in S206 to S208 of Figure 12, the process is performed up to the point where the current solution resistance R'1 is output, and the remaining capacity is estimated and output, similar to S105 to S107 of Figure 10.

[0034] As shown in Figure 13, any frequency band can be, for example, within the range of a fourth frequency ω4 to 2 × a first frequency ω2. With the current application waveform, measured voltage waveform, and generated current waveform shown in Figures 7(a) to 7(c), the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency becomes approximately constant. As a result, as shown in Figure 14, the diameter of the arc of the Cole-Cole plot created by the Cole-Cole plot creation unit 12 is reduced, but results similar to those of the electrochemical impedance method can be obtained.

[0035] The process for adjusting the phase difference between the voltage waveform and the current waveform of the battery state estimation device 1A will be described. As shown in S301 to S305 of Figure 15, the process is performed to confirm the phase difference between the voltage waveform and the current waveform, similar to S201 to S205 in Figure 12. The call-call plot creation unit 12 determines whether the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency is within a third predetermined value in an arbitrary frequency band, for example, in the range of fourth frequency ω4 to 2 × first frequency ω2 (S306).

[0036] If the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency is not within a third predetermined value in any frequency band (S306), the current waveform creation unit 11 adjusts the time T2 of the current waveform so that the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency is within a third predetermined value in any frequency band (S307).

[0037] From this point onward, the processes described in S303 to S307 are executed until the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency falls within a third predetermined value in any frequency band (S306). When the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency falls within a third predetermined value in any frequency band (S306), the process is carried out to output the current solution resistance R'1, as shown in S308 to S310 in Figure 15, and to estimate and output the remaining capacitance, similar to the process in S206 to S208 in Figure 12.

[0038] As shown in Figure 26, in the conventional battery state estimation device 1000, the current application unit 2B applies a sinusoidal alternating current to the battery 10 while sweeping the frequency in the range of 1 kHz to 0.01 Hz, for example. The voltage measurement unit 3 measures the voltage of the battery 10. The current measurement unit 14 measures the current flowing through the shunt resistor 13. The EIS analysis unit 15 creates a Cole-Cole plot of the impedance of the battery 10 from the voltage measured by the voltage measurement unit 3 and the current measured by the current measurement unit 14. In domain 100, the current solution resistance R'1 and remaining capacity of the battery 10 are estimated based on the Cole-Cole plot analyzed by the EIS analysis unit 15.

[0039] Conventional battery state estimation devices 1000 require high synchronization between the voltage measurement unit 3 and the current measurement unit 14. Furthermore, conventional battery state estimation devices 1000 have a problem where, for example, when applying a low-frequency current of 1 Hz or less, sampling takes time and power consumption increases. In addition, measurement errors due to synchronization mismatches in the waveform detection of voltage and current, and measurement errors due to noise in the detection circuit, worsen the accuracy of estimating the state of a battery 10 with an impedance of about 0.1 mΩ.

[0040] On the other hand, according to this embodiment, the current application unit 2A of the battery state estimation device 1A applies a current with a falling waveform to the battery 10 such that the difference between the maximum applied current and the minimum applied current is greater than a first predetermined value. The voltage measurement unit 3 measures the voltage of the battery 10 to which current has been applied by the current application unit 2A. The monitoring unit 4A estimates the state of the battery based on the voltage measured by the voltage measurement unit 3.

[0041] The third frequency ω1 in the time T1 = 1 / ω1 × 2, during which the current applied by the current application unit 2A falls from its maximum value to its minimum value, is greater than the first frequency ω2 and less than the second frequency ω0. In other words, in this embodiment, voltage waveform sampling is performed at a relatively high frequency, and no sweeping is performed at low frequencies.

[0042] The monitoring unit 4A converts the transient change in the voltage waveform of the battery 10, measured by the voltage measurement unit 3 when current is applied by the current application unit 2A, into a complex frequency. The monitoring unit 4A also converts the transient change in the current waveform into a complex frequency when the difference between the maximum and minimum generated current is greater than a second predetermined value, and the time T2 for the current to fall from the maximum to the minimum generated current is greater than time T0 = 1 / ω0 × 2 and less than time T1. In other words, in this embodiment, sampling of the current waveform is performed at a relatively high frequency, and sweeping at low frequencies is not performed.

[0043] The monitoring unit 4A estimates the state of the battery 10 by creating a Cole-Cole plot of the battery 10's impedance from the transient changes in the voltage waveform converted to complex frequency and the transient changes in the current waveform converted to complex frequency. Although this Cole-Cole plot is not based on the actually measured current waveform, it is similar to the Cole-Cole plot obtained by the conventional electrochemical impedance method, but with a smaller arc diameter. Therefore, it is possible to obtain results similar to those of the conventional electrochemical impedance method.

[0044] In other words, in this embodiment, the current of the battery 10 is not actually measured, and a configuration for measuring the current of the battery 10 is not essential, so synchronization between voltage measurement and current measurement is not required. Furthermore, in this embodiment, sampling of the voltage waveform and current waveform is performed at a relatively high frequency, which reduces the time required to estimate the state of the battery 10. Moreover, in this embodiment, measurements are not performed at low frequencies, so even with a battery 10 having an impedance of about 0.1 mΩ, the state of the battery 10 can be estimated with high accuracy.

[0045] Furthermore, according to this embodiment, the monitoring unit 4A performs the conversion of transient changes in the voltage waveform to complex frequencies and the conversion of transient changes in the current waveform to complex frequencies using Fourier transforms. Therefore, the state of the battery 10 can be estimated using existing calculations.

[0046] Furthermore, according to this embodiment, since the first frequency ω2 is 20 to 80 Hz and the second frequency ω0 is 1 kHz, sampling of the voltage waveform and current waveform is performed at a relatively high frequency, which reduces the time required to estimate the state of the battery 10 and allows for the estimation of the state of the battery 10 with high accuracy.

[0047] Furthermore, according to this embodiment, the minimum current applied by the current application unit 2A is 0A. In other words, in this embodiment, after the current changes, the MOSFET 7 turns off and the current becomes 0A, so the effect of switching noise is eliminated and the voltage measurement unit 3 can measure the voltage with high accuracy.

[0048] Furthermore, according to this embodiment, the monitoring unit 4A adjusts the time T2 of the current waveform so that the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency falls within a third predetermined value in any frequency band. By doing so, the monitoring unit 4A can easily create a call-call plot by keeping the phase difference in the frequency band where the distribution of resistance and capacitance is to be monitored within a third predetermined value.

[0049] Furthermore, according to this embodiment, the monitoring unit 4A calculates the value on the real axis where the imaginary axis is 0 in the created Cole-Cole plot as the current solution resistance R'1. The monitoring unit 4A calculates the difference between the value on the real axis at the fourth frequency ω4, which is the inflection point of the arc containing the first frequency ω2, and the solution resistance R'1 in the created Cole-Cole plot as the current resistance component R'2. The monitoring unit 4A also calculates the difference between the value on the real axis at the first frequency ω2 and the current solution resistance R'1 in the created Cole-Cole plot as the current resistance component R'3. The monitoring unit 4A estimates the remaining capacity of the battery 10 from the current resistance component R'2 or the current resistance component R'3. When determining the remaining capacity with a certain degree of accuracy, the remaining capacity is correlated with the current resistance components R'2 and R'3, so the monitoring unit 4A can estimate the remaining capacity from the resistance component on the real axis with a simple calculation.

[0050] Furthermore, according to this embodiment, the monitoring unit 4A estimates the remaining capacity of the battery 10 from the area S calculated by integrating the imaginary axis value in the created Cole-Cole plot over the interval from the current solution resistance R'1 on the real axis to the sum of the current solution resistance R'1 and the current resistance component R'2. Alternatively, the monitoring unit 4A estimates the remaining capacity of the battery 10 from the area S calculated by integrating the imaginary axis value in the created Cole-Cole plot over the interval from the current solution resistance R'1 on the real axis to the sum of the current solution resistance R'1 and the current resistance component R'3. The remaining capacity has a high correlation with the area S. In the process described above, where domain 100 estimates the remaining capacity of the battery from the integral of the imaginary axis value in the Cole-Cole plot, the remaining capacity includes the capacity distribution, i.e., the imaginary axis side as well, so the accuracy of estimating the remaining capacity is improved compared to the process of estimating from only the current resistance component R'2 or the current resistance component R'3.

[0051] [Second Embodiment] The second embodiment will now be described. As shown in Figure 16, the battery state estimation device 1B of this embodiment includes a monitoring unit 4B instead of the monitoring unit 4A of the first embodiment. The monitoring unit 4B includes a temperature estimation unit 16, a temperature estimation database 17, a room temperature conversion unit 18, and a temperature characteristic database 19. The temperature estimation database 17 stores the relationship between the phase change with respect to frequency during transient changes in the waveform of the voltage converted to a complex frequency and the temperature of the battery 10. The temperature estimation unit 16 estimates the temperature of the battery 10 from the relationship between the phase change with respect to frequency during transient changes in the waveform of the voltage converted to a complex frequency and the temperature of the battery 10.

[0052] The phase relative to frequency in the transient change of the voltage waveform converted to a complex frequency shows little change due to the degradation of the battery 10. However, as shown in Figure 17, the phase relative to frequency in the transient change of the voltage waveform converted to a complex frequency shows a large change with respect to the temperature of the battery 10. As shown in Figure 17, at the same temperatures of 5°C and 25°C, there is almost no phase difference with respect to the same frequency between a normal battery 10 and a degraded battery 10. However, at different temperatures of 5°C and 25°C, a large phase difference with respect to the same frequency appears for both a normal battery 10 and a degraded battery 10 in the range of fourth frequency ω4 ~ 2 × first frequency ω2. In this embodiment, this relationship is utilized to employ a method for estimating the temperature of the battery 10 that does not require temperature sensors such as thermometers and thermocouples.

[0053] The temperature characteristics database 19 stores the temperature characteristics of the relationship between the temperature of the battery 10 and the current solution resistance R'1. The temperature characteristics database 19 also stores the temperature characteristics of the relationship between the temperature of the battery 10 and the current resistance component R'2. Furthermore, the temperature characteristics database 19 stores the temperature characteristics of the relationship between the temperature of the battery 10 and the current resistance component R'3. Additionally, the temperature characteristics database 19 stores the temperature characteristics of the relationship between the temperature of the battery 10 and the current capacity component R' j The temperature characteristics related to this are stored in memory.

[0054] The room temperature conversion unit 18 converts the current solution resistance R'1 to the room temperature solution resistance R' at room temperature based on the temperature of the battery 10 estimated by the temperature estimation unit 16 and the temperature characteristics of the current solution resistance R'1 stored in the temperature characteristics database 19. nt1 The room temperature conversion unit 18 converts the current resistance component R'2 to the room temperature resistance component R'2 at room temperature, based on the temperature of the battery 10 estimated by the temperature estimation unit 16 and the temperature characteristics of the current resistance component R'2 stored in the temperature characteristics database 19. nt2 Convert to.

[0055] The room temperature conversion unit 18 converts the current resistance component R'3 to the room temperature resistance component R' at room temperature based on the temperature of the battery 10 estimated by the temperature estimation unit 16 and the temperature characteristics of the current resistance component R'3 stored in the temperature characteristics database 19. nt3is converted. The normal temperature conversion unit 18 uses the temperature of the battery 10 estimated by the temperature estimation unit 16 and the current capacity component R' stored in the temperature characteristic database 19 j to convert the value on the imaginary axis of the Cole-Cole plot created by the Cole-Cole plot creation unit 12 based on the temperature characteristics.

[0056] Next, the operation of the battery state estimation device 1B of this embodiment will be described. As shown in S401 to S404 of FIG. 18, processing until the created current sampled at the cycle of time T2 similar to S101 to 104 of FIG. 10 is Fourier-transformed is performed. The temperature estimation unit 16 calculates, for example, the change in phase with respect to the frequency in the range of the fourth frequency ω4 to twice the first frequency ω2 in the transient change of the waveform of the voltage converted into a complex frequency (S405). The temperature estimation unit 16 calculates the temperature of the battery 10 from the relationship between the calculated change in phase with respect to the frequency, the change in phase with respect to the frequency stored in the temperature estimation database 17, and the temperature of the battery 10 (S406).

[0057] The Cole-Cole plot creation unit 12 creates a Cole-Cole plot in the same manner as in the first embodiment, and calculates the current solution resistance R'1, the current resistance component R'2, and the current resistance component R'3 (S407). The normal temperature conversion unit 18 uses the temperature of the battery 10 estimated by the temperature estimation unit 16 and the temperature characteristics of the current solution resistance R'1, the current resistance component R'2, and the current resistance component R'3 stored in the temperature characteristic database 19 to convert the current solution resistance R'1, the current resistance component R'2, and the current resistance component R'3 into the normal temperature solution resistance R' tn1 , the normal temperature resistance component R' tn2 and the normal temperature resistance component R' tn3 (S408). The normal temperature conversion unit 18 uses the temperature of the battery 10 estimated by the temperature estimation unit 16 and the temperature characteristics of the current capacity component R' j to convert the value on the imaginary axis of the Cole-Cole plot created by the Cole-Cole plot creation unit 12 (S408).

[0058] The domain 100 uses the value on the imaginary axis in the converted Cole-Cole plot as the real-axis normal temperature solution resistance R' nt1From room temperature solution resistance R' nt1 and room temperature resistance component R' nt2 The area S is calculated by integrating over the interval up to the sum of (S409). In this case, similar to the first embodiment described above, domain 100 uses the value of the imaginary axis in the transformed Cole-Cole plot as the real axis of room temperature solution resistance R'. nt1 From room temperature solution resistance R' nt1 and room temperature resistance component R' nt3 Alternatively, the area S can be calculated by integrating over the interval up to the sum of (S409).

[0059] Domain 100 is the room temperature solution resistance R'. nt1 Domain 100 outputs (S410). In addition, Domain 100 estimates and outputs the remaining capacity of the battery 10 from the relationship between the area S and the remaining capacity (S410). In this case, Domain 100 outputs the room temperature resistance component R'. tn2 The relationship between this and the remaining capacity or the room temperature resistance component R' tn3 Based on the relationship with the remaining capacity, the remaining capacity of battery 10 may be estimated and output (S410).

[0060] According to this embodiment, the monitoring unit 4B estimates the temperature of the battery 10 from the relationship between the phase change with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the temperature of the battery 10. Therefore, the internal temperature of the battery 10 and the degradation of the battery 10 can be calculated independently without the need for temperature sensors such as thermometers and thermocouples.

[0061] Furthermore, according to this embodiment, the monitoring unit 4B calculates the current solution resistance R'1 from the estimated temperature of the battery 10 and the temperature characteristics of the current solution resistance R'1, and then calculates the room temperature solution resistance R'1 from the room temperature solution resistance R' nt1 It converts the current resistance component R'2 to the room temperature resistance component R'2 at room temperature, based on the estimated temperature of the battery 10 and the temperature characteristics of the current resistance component R'2. nt2 It converts to [this]. Furthermore, the monitoring unit 4B detects the room temperature solution resistance R'. nt1 The state of the solution in battery 10 is estimated from this, and the room temperature resistance component R' is determined. nt2The remaining capacity of the battery 10 is estimated from this. For this purpose, the monitoring unit 4B corrects the estimated state of the battery 10 to the state at room temperature and can estimate the remaining capacity from only the resistance component of the real axis with a simple calculation.

[0062] Furthermore, according to this embodiment, the monitoring unit 4B determines the estimated temperature of the battery 10 and the current capacity component R'. j The imaginary axis values ​​of the created Cole-Cole plot are converted based on the temperature characteristics. The monitoring unit 4B also converts the imaginary axis values ​​in the converted Cole-Cole plot to the room temperature solution resistance R' on the real axis. nt1 From room temperature solution resistance R' nt1 and room temperature resistance component R' nt2 The remaining capacity of the battery 10 is estimated from the area S calculated by integrating over the interval up to the sum of the two. For this reason, the monitoring unit 4B corrects the estimated state of the battery 10 to the state at room temperature, and since the remaining capacity including the imaginary axis is estimated, the accuracy of estimating the remaining capacity is the room temperature resistance component R'. nt2 Or the current resistance component R' nt3 This is an improvement compared to processes that estimate from only that information.

[0063] [Third Embodiment] The third embodiment will now be described. As shown in Figure 19, the battery state estimation device 1C of this embodiment is equipped with a current application unit 2C in place of the current application unit 2A of the first embodiment. Also, the battery state estimation device 1C of this embodiment is equipped with a monitoring unit 4C in place of the monitoring unit 4A of the first embodiment. Furthermore, it is equipped with a shunt resistor 13 and a current measurement unit 14 similar to the conventional battery state estimation device 1000 shown in Figure 26.

[0064] The current application unit 2C applies a current to the battery 10 with a waveform that falls so that the difference between the maximum and minimum applied current values ​​is greater than a first predetermined value, and the time T1 is the time it takes for the current to fall from the maximum to the minimum applied current value. In addition, the current application unit 2C applies a sinusoidal alternating current to the battery 10 while sweeping the frequency in the range of 1kHz to 0.01Hz, similar to the current application unit 2B of the conventional battery state estimation device 1000 shown in Figure 26. The voltage measurement unit 3 measures the voltage of the battery 10. The current measurement unit 14 measures the current flowing through the shunt resistor 13. The voltage measurement value from the voltage measurement unit 3 and the current measurement value from the current measurement unit 14 are synchronized.

[0065] The monitoring unit 4C includes an EIS analysis unit 15 and an EIS correction unit 20. Similar to the EIS analysis unit 15 of the conventional battery state estimation device 1000 shown in Figure 26, the EIS analysis unit 15 creates a Cole-Cole plot of the battery 10's impedance from the voltage measured by the voltage measurement unit 3 and the current measured by the current measurement unit 14. This provides a first distribution result of the resistance and capacitance of the Cole-Cole plot of the battery 10's impedance. Note that the current measurement unit 14 and the EIS analysis unit 15 do not necessarily need to be built into the monitoring unit 4C. For example, the first distribution result of the battery 10 obtained by electrochemical impedance method using another device outside the battery state estimation device 1C may be used in the process described below.

[0066] The Cole-Cole plot creation unit 12, similar to the first embodiment described above, creates a Cole-Cole plot of the battery 10's impedance from the voltage waveform measured by the voltage measurement unit 3 and the current waveform created by the current waveform creation unit 11. This provides a second distribution result of the resistance and capacitance of the Cole-Cole plot of the battery 10's impedance. The EIS correction unit 20 calculates a correction value from the created second distribution result of the resistance and capacitance of the Cole-Cole plot of the battery 10's impedance to convert the second distribution result to a third distribution result corresponding to the first distribution result. The EIS correction unit 20 converts the second distribution result to a third distribution result from the second distribution result and the correction value.

[0067] The process of converting the second distribution result of the battery state estimation device 1C to a third distribution result corresponding to the first distribution result will be explained. As shown in Figure 20, the EIS analysis unit 15 creates a Cole-Cole plot by electrochemical impedance method and obtains the first distribution result (S501). As shown in S502 to S506 of Figure 20, the Cole-Cole plot creation unit 12 performs the same process as described above in S001 to S005 of Figure 6 until the Cole-Cole plot is created.

[0068] The EIS correction unit 20 calculates a correction value that converts the second distribution result of resistance and capacitance from the Cole-Cole plot of the created battery impedance to a third distribution result corresponding to the first distribution result (S507). As shown in Figure 21, the R of the first distribution result D1 obtained by the electrochemical impedance method 1E , R 2E and R jE In comparison, the current solution resistance R'1, current resistance component R'2, and current capacitance component R' of the second distribution result D2 j The value becomes smaller. The EIS correction unit 20 sets k1=R 1E / R'1, k2=R 2E / R'2 and k3 = R jE / R' j The correction values ​​k1, k2, and k3 are calculated. As shown in Figure 20, the EIS correction unit 20 converts the second distribution result D2 to a third distribution result from the second distribution result D2 and the correction values ​​(S507). As shown in Figure 21, the third distribution result D3, which corresponds to the first distribution result D1, is obtained from the second distribution result D2.

[0069] As shown in Figure 20, Domain 100 calculates the area S by integrating the imaginary axis value f(Re)·k3 in the Cole-Cole plot of the third distribution result D3 converted by the EIS correction unit 20 over the interval from the current solution resistance k1·R'1 on the real axis converted by the EIS correction unit 20 to the sum of the current solution resistance k1·R'1 and the current resistance component k2·R'2 (S508). Domain 100 outputs the current solution resistance k1·R'1 (S509). Domain 100 also estimates and outputs the remaining capacity of the battery 10 from the relationship between the area S and the remaining capacity (S509). Note that in the process shown in Figure 20, after the EIS correction unit 20 calculates the correction values ​​k1, k2, and k3, the process of obtaining the first distribution result by the EIS analysis unit 15 in S501 can be omitted.

[0070] The following describes the process of creating a fourth distribution result by interpolating the first distribution result D1 in the low frequency band with the third distribution result D3 in the low frequency band from the battery state estimation device 1C. In the example of the process described below, the first frequency ω2 = 25 Hz, the second frequency ω0 = 1 kHz, and the third frequency ω1 = 50 Hz. As shown in Figure 22, the current application unit 2C applies an AC wave current to the battery 10 while sweeping the frequency range from 0.01 Hz to 1 kHz (S601). The EIS analysis unit 15 performs an analysis by electrochemical impedance method in the frequency range from 0.01 Hz to 1 kHz (S602). The EIS analysis unit 15 creates a Cole-Cole plot by electrochemical impedance method and obtains the first distribution result D1.

[0071] The EIS analysis unit 15 instructs the EIS correction unit 20 to create a fourth distribution result by interpolating the first distribution result D1 for the frequency band below the first frequency ω2 using the third distribution result D3 for the frequency band below the first frequency ω2, which in the example of Figure 22 is below 50 Hz (S604). As shown in Figure 23, the accuracy of the first distribution result D1 decreases in the low frequency band.

[0072] On the other hand, as shown in S605 to S610 in Figure 22, the EIS correction unit 20, similar to S502 to S507, performs a process to convert the second distribution result D2 to the third distribution result D3. The EIS correction unit 20 interpolates the first distribution result D1 for the frequency band below the first frequency ω2 using the third distribution result D3 for the frequency band below the first frequency ω2 to create a fourth distribution result (S611). As a result, a fourth distribution result D4 with interpolated low-frequency bands is obtained, as shown in Figure 24.

[0073] As shown in Figure 22, Domain 100 is the value of the imaginary axis in the call-call plot of the fourth distribution result D4, f(R j e) R on the real axis 1E From R 1E and R 2E The area S is calculated by integrating over the interval up to the sum of (S612). Domain 100 has R as the current solution resistance. 1E The output is (S613). In addition, domain 100 estimates the remaining capacity of battery 10 from the relationship between area S and remaining capacity and outputs it (S613).

[0074] As mentioned above, the R of the first distribution result D1 obtained by electrochemical impedance method 1E , R 2E and R jE In comparison, the current solution resistance R'1, current resistance component R'2, and current capacitance component R' of the second distribution result D2 j The value becomes smaller. However, according to this embodiment, with respect to the first distribution result D1 obtained by the electrochemical impedance method, the monitoring unit 4C calculates correction values ​​k1, k2, and k3 from the second distribution result D2 of resistance and capacitance in the Cole-Cole plot of the impedance of the created battery 10 to convert the second distribution result D2 to the third distribution result D3 which corresponds to the first distribution result. The monitoring unit 4C also converts the second distribution result D2 to the third distribution result D3 from the second distribution result D2 and the correction values ​​k1, k2, and k3.

[0075] After the correction values ​​k1, k2, and k3 are calculated, there is no need to perform electrochemical impedance analysis. Therefore, synchronization between voltage and current measurements is not required, the time required to estimate the state of battery 10 is reduced, and results similar to those obtained by the widely used electrochemical impedance analysis are obtained. In addition, it becomes easy to make highly accurate comparisons with other batteries 10 whose states have been estimated by electrochemical impedance analysis.

[0076] As described above, in the electrochemical impedance method, measurements at low frequencies using AC waveforms are time-consuming. This causes changes in the state of the battery 10, such as its temperature, making it difficult to obtain accuracy. On the other hand, in this embodiment, the call-call plot creation unit 12 enables stable low-frequency measurements in a short time. Therefore, in this embodiment, the monitoring unit 4C interpolates the first distribution result D1 for the frequency band below the first frequency ω2 using the third distribution result D3 for the frequency band below the first frequency ω2 to create a fourth distribution result D4. As a result, the state of the battery 10 can be estimated with high accuracy even in the low-frequency band using the electrochemical impedance method.

[0077] [Fourth Embodiment] The fourth embodiment will now be described. As shown in Figure 25, the battery state estimation device 1D of this embodiment includes a shunt resistor 13 and a current measurement unit 14, similar to the conventional battery state estimation device 1000 shown in Figure 26. As described above for the first embodiment, the shunt resistor 13 and the current measurement unit 14 are not essential components. However, the shunt resistor 13 and the current measurement unit 14 may remain, as in the case where the existing battery state estimation device 1000 is modified to constitute the battery state estimation device 1D of this embodiment. The battery state estimation device 1D provides the same effects as the battery state estimation device 1A of the first embodiment.

[0078] Although several embodiments of the present invention have been described above, the invention is not limited to the embodiments described above and can be realized in various configurations without departing from its spirit. For example, configurations that arbitrarily combine the configurations of Embodiments 1 to 4 can easily be conceivable. These variations are included within the scope of the invention and its equivalents as described in the claims. [Explanation of symbols]

[0079] 1A, 1B, 1C, 1D Battery State Estimation Device 2A, 2B, 2C current application section 3. Voltage measurement unit 4A,4B,4C Monitoring section 5 BMU 6 Signal Generation Unit 7 MOSFET 8. Load Resistance 9 CMU 10 batteries 11 Current waveform generation unit 12. Call-Call Plot Creation Section 13 Shunt resistor 14 Current measurement section 15 EIS analysis department 16 Temperature estimation section 17 Temperature Estimation Database 18 Room temperature conversion unit 19 Temperature Characteristics Database 20 EIS Correction Unit 100,200,300 domains 400 Central ECU 500 Electric Vehicles 600 Cloud R1 Solution resistance R2 resistance component R3 resistance component C2 capacity component C3 capacity component Z0 diffusion resistance R'1 Current solution resistance R'2 Current Resistance Component R'3 Current Resistance Component R j Current volume ingredients S area D1 1st distribution result D2 2nd distribution result D3 3rd distribution result D4 4th distribution result

Claims

1. The constants of the equivalent circuit are the solution resistance R. １ , the resistance component R of the negative electrode ２ and the capacitance component C of the negative electrode ２ A battery state estimation device that estimates the state of a battery, A current application unit applies a current to the battery with a falling waveform such that the difference between the maximum applied current and the minimum applied current is greater than a first predetermined value. A voltage measuring unit measures the voltage of the battery to which current is applied by the current application unit, A monitoring unit estimates the state of the battery based on the voltage measured by the voltage measurement unit, Equipped with, The current application unit is Time T for the applied current to fall from its maximum value to its minimum value. １ = 1 / ω １ Apply a current with a waveform that is ×2, Time T １ = 1 / ω １ × 2 at the third frequency ω １ is the first frequency ω ２ = 1 / R ２ C ２ and, in the Cole-Cole plot of the impedance of the equivalent circuit, the real axis is the solution resistance R １ at the value of the second frequency ω ０ and with respect to the first frequency ω ２ is greater than, and the second frequency ω ０ is less than, The aforementioned monitoring unit, When a current is applied by the current application unit, the transient change in the waveform of the battery voltage measured by the voltage measurement unit is converted into a complex frequency. The current falls so that the difference between the maximum and minimum current values ​​is greater than a second predetermined value, and the time T is the time it takes for the current to fall from the maximum to the minimum current value. ２ However, time T ０ = 1 / ω ０ Greater than ×2, and the aforementioned time T １ The transient change in the waveform of a current that becomes smaller than a certain value is converted into a complex frequency. A battery state estimation device characterized by estimating the state of a battery by creating a Cole-Cole plot of the battery's impedance from the transient changes of the voltage waveform converted to a complex frequency and the transient changes of the current waveform converted to a complex frequency.

2. The battery state estimation device according to claim 1, characterized in that the monitoring unit performs the conversion of transient changes in the voltage waveform to complex frequencies and the conversion of transient changes in the current waveform to complex frequencies by Fourier transform.

3. the first frequency ω ２ The frequency is 20-80 Hz, and the second frequency ω ０ The battery state estimation device according to claim 1, characterized in that the frequency is 1 kHz.

4. The battery state estimation device according to claim 1, characterized in that the minimum value of the current applied by the current application unit is 0A.

5. The aforementioned monitoring unit, If the difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency is not within a third predetermined value in any frequency band, The difference between the phase with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the phase with respect to frequency during the transient change of the current waveform converted to a complex frequency is set to fall within the third predetermined value in any frequency band. The current waveform over time T ２ The battery state estimation device according to claim 1, characterized by adjusting the battery state.

6. The aforementioned monitoring unit, In the created Cole-Cole plot, the value on the real axis where the imaginary axis is 0 represents the current solution resistance R'. １ Calculated as follows: In the call-call plot created, the first frequency ω ２ The fourth frequency ω is the inflection point of the arc containing it. ４ The value on the real axis and the solution resistance R' １ The difference is the current resistance component R' ２ Calculated as follows: The aforementioned current resistance component R' ２ The battery state estimation device according to claim 1, characterized in that it estimates the remaining capacity of the battery from the above.

7. The aforementioned monitoring unit, In the created Cole-Cole plot, the value on the real axis where the imaginary axis is 0 represents the current solution resistance R'. １ Calculated as follows: In the call-call plot created, the first frequency ω ２ The value on the real axis and the current solution resistance R' １ The difference is the current resistance component R' ３ Calculated as follows: The aforementioned current resistance component R' ３ The battery state estimation device according to claim 1, characterized in that it estimates the remaining capacity of the battery from the above.

8. The aforementioned monitoring unit, In the created Cole-Cole plot, the value on the real axis where the imaginary axis is 0 represents the current solution resistance R'. １ Calculated as follows: In the call-call plot created, the first frequency ω ２ The fourth frequency ω is the inflection point of the arc containing it. ４ The value on the real axis and the current solution resistance 'R'. １ The difference is the current resistance component R' ２ Calculated as follows: The value on the imaginary axis in the created Cole-Cole plot corresponds to the current solution resistance R' on the real axis. １ From the current solution resistance R' １ and the aforementioned current resistance component R' ２ The battery state estimation device according to claim 1, characterized in that it estimates the remaining capacity of the battery from the area calculated by integrating over the interval up to the sum of the two.

9. The aforementioned monitoring unit, In the created Cole-Cole plot, the value on the real axis where the imaginary axis is 0 represents the current solution resistance R'. １ Calculated as follows: In the call-call plot created, the first frequency ω ２ The value on the real axis and the solution resistance R' １ The difference is the current resistance component R' ３ Calculated as follows: The value on the imaginary axis in the created Cole-Cole plot corresponds to the current solution resistance R' on the real axis. １ From the current solution resistance R' １ and the aforementioned current resistance component R' ３ The battery state estimation device according to claim 1, characterized in that it estimates the remaining capacity of the battery from the area calculated by integrating over the interval up to the sum of the two.

10. An AC current is applied to the battery while sweeping the frequency, and the first distribution result of the resistance and capacitance of the Cole-Cole plot of the battery's impedance, obtained from the synchronized current and voltage measurements, The aforementioned monitoring unit, From the second distribution result of resistance and capacitance in the Cole-Cole plot of the impedance of the created battery, a correction value is calculated to convert the second distribution result to a third distribution result corresponding to the first distribution result. The battery state estimation device according to claim 1, characterized in that it converts the second distribution result to the third distribution result from the second distribution result and the correction value.

11. The aforementioned monitoring unit, the first frequency ω ２ Based on the third distribution results in the following frequency band, the first frequency ω ２ The battery state estimation device according to claim 10, characterized in that a fourth distribution result is created by interpolating the first distribution result for the following frequency bands.

12. The aforementioned monitoring unit, The battery state estimation device according to claim 1, characterized in that it estimates the temperature of the battery from the relationship between the phase change with respect to frequency during the transient change of the voltage waveform converted to a complex frequency and the temperature of the battery.

13. The aforementioned monitoring unit, In the created Cole-Cole plot, the value on the real axis where the imaginary axis is 0 represents the current solution resistance R'. １ Calculated as follows: The estimated temperature of the battery and the current solution resistance R' １ Based on the temperature characteristics, the current solution resistance R' １ The room temperature solution resistance R' ｎｔ１ Convert to, In the call-call plot created, the first frequency ω ２ The fourth frequency ω is the inflection point of the arc containing it. ４ The value on the real axis and the current solution resistance R' １ The difference is the current resistance component R' ２ Calculated as follows: The estimated temperature of the battery and the current resistance component R' ２ From the temperature characteristics, the current resistance component R' ２ The room temperature resistance component R' at room temperature ｎｔ２ Convert to, The aforementioned room-temperature solution resistance R' ｎｔ１ From this, the state of the solution in the battery is estimated, The aforementioned room-temperature resistance component R' ｎｔ２ The battery state estimation device according to claim 12, characterized in that it estimates the remaining capacity of the battery from the above.

14. The aforementioned monitoring unit, In the created Cole-Cole plot, the value on the real axis where the imaginary axis is 0 represents the current solution resistance R'. １ Calculated as follows: The estimated temperature of the battery and the current solution resistance R' １ Based on the temperature characteristics, the current solution resistance R' １ The room temperature solution resistance R' ｎｔ１ Convert to, In the call-call plot created, the first frequency ω ２ The fourth frequency ω is the inflection point of the arc containing it. ４ The value on the real axis and the solution resistance R' １ The difference is the current resistance component R' ２ Calculated as follows: The estimated temperature of the battery and the current resistance component R' ２ From the temperature characteristics, the current resistance component R' ２ The room temperature resistance component R' at room temperature ｎｔ２ Convert to, In the call-call plot created, the first frequency ω ２ The value of the imaginary axis in the current capacity component R' ｊ Calculated as follows: The estimated temperature of the battery and the current capacity component R' ｊ Based on the temperature characteristics, the imaginary axis values ​​of the created Cole-Cole plot are transformed. The value on the imaginary axis in the converted Cole-Cole plot corresponds to the room-temperature solution resistance R' on the real axis. ｎｔ１ From the above room temperature solution resistance R' ｎｔ１ and the aforementioned room-temperature resistance component R' ｎｔ２ The battery state estimation device according to claim 12, characterized in that it estimates the remaining capacity of the battery from the area calculated by integrating over the interval up to the sum of the two.

Images