JP7549305B1 - Battery deterioration determination support method and device - Google Patents

Abstract

A battery deterioration determination support device is provided that is small, has low power consumption, and can significantly reduce measurement time.
[Solution] A method for assisting in determining deterioration of an electric energy supply medium by acquiring the internal characteristics of the medium using an electronic circuit, comprising: acquiring the internal characteristics of a battery in advance; selecting two or more frequencies required for determining whether the battery is good or bad; storing the selected frequencies; performing discharge control based on the selected frequencies; acquiring voltage data and current data using a voltage measurement means and a current measurement means; performing a correction process on the voltage data to calculate corrected voltage data; calculating resistance data from the corrected voltage data and current data; calculating at least a portion of the Cole-Cole plot data from the resistance data; and performing a fitting process using at least a portion of the Cole-Cole plot data as necessary.
[Selected Figure] Figure 1

Description

The present invention relates to a method for assisting in determining battery deterioration, and more specifically, to a method for assisting in determining battery deterioration by acquiring internal characteristics of batteries such as lithium ion secondary batteries, nickel-hydrogen secondary batteries, solar cells, and fuel cells.

In recent years, various methods have been proposed for diagnosing the deterioration state of electric energy supply media such as secondary batteries. One of the diagnostic methods is the Cole-Cole plot or Nyquist plot, in which the horizontal axis represents the real part of the AC impedance of the secondary battery, etc., and the vertical axis represents the imaginary part of the AC impedance of the secondary battery, etc.

(Explanation of the Cole-Cole plot principle)
A Cole-Cole plot is a plot of values measured by an AC-IR measurement method that measures the internal resistance of a battery using an alternating current. Specifically, the current and voltage values are measured when sine waves of various frequencies are applied to the battery under test, and the resistance values including the phase difference are calculated from these values, and the real and imaginary parts of the resistance values are plotted to create a scatter diagram.
In addition, in the case of AC, the voltage value V is expressed as Vr + Vi×j (where j = (-1) ^0.5 ), the current value I is expressed as Ir + I×j, and the resistance value R is calculated as a complex number called V/I. As an example, when V = -229.585443703293 - 4195.93748617154i and I = -1.47516371438595 - 5.46385648982277i, the calculation is R = 0.0012072190153994 - 0.000304333685571092i.

Also, applying a sine wave means charging and discharging the battery to be measured. The frequency set for measurement is from 0.1Hz to 1kHz, and several tens to hundreds of measurements are required. Measurement time at high frequencies is not an issue, but for example, a measurement at a frequency of 0.1Hz takes 10 seconds, so it takes several minutes to 10 minutes to measure at all planned frequencies.
Furthermore, if you perform smoothing to reduce noise, more measurement time will be required. For example, if you want to take the average of four measurements, the time required for one measurement will be four times longer than in the above example, about 20 to 30 minutes.

An example of generating a Cole-Cole plot from the measurement results by an AC-IR measuring instrument will be described with reference to Figures 8A and 8B. Figure 8A exemplifies the measurement results by an AC-IR measuring instrument. That is, in the figure, the current and voltage values are measured when sine waves of various frequencies are applied to the battery to be measured by the AC-IR measuring instrument, and the resistance values (real resistance and imaginary resistance) including the phase difference calculated from these measurement values are shown for each frequency.
Fig. 8B shows a plot of the real and imaginary parts of the resistance value for each frequency shown in Fig. 8A. In addition, in the Cole-Cole plot shown in Fig. 8B, the zero crossing point (A in the figure) and the change point (depression B in the figure) tend to appear at the same positions for batteries with the same conditions (temperature, battery voltage, etc., the same below) and characteristics. Therefore, if there is a deviation between these points for batteries with the same conditions and characteristics, it can be determined that there is an abnormality or deterioration. This type of determination method can be used, for example, in checks during the manufacture of battery products.

(AC-IR measurement method)
The AC-IR measurement method estimates the internal parameters of a battery by acquiring the AC characteristics of the battery. Conventional methods estimate the deviation of each parameter from its normal value by observing how much it deviates from the normal value. The deterioration of the battery has been estimated by using the DC-IR measurement method. The DC-IR measurement method has also been used in the past, but this is a judgment method based on the resistance value measured by a direct current of the battery's internal resistance. This measurement method, which uses a small current load, causes less damage to the battery and is said to provide more reliable measurements than the DC-IR measurement method, which only reveals some of the characteristics. The AC characteristics obtained by this measurement method can be used to understand the state of various reactions, not only in storage batteries, but also in solar cells, fuel cells, chemical reactions, and more.

Figure 9A conceptually illustrates the physical structure of a lithium-ion secondary battery. In Figure 9A, the battery 900 includes a positive electrode 910, a negative electrode 920, and an electrolyte 930. In one embodiment, the positive electrode 910 uses an aluminum alloy foil with a thickness of about 15 to 30 μm as a positive electrode current collector 911, and the negative electrode 920 uses pure copper, such as rolled copper foil or electrolytic copper foil, as a negative electrode current collector 921. 940 is a Li compound formed on the negative electrode by decomposition of the electrolyte and additives during the initial charging process of the lithium-ion secondary battery, and is called SEI (Solid Electrolyte Interphase). In addition, particles without reference numbers in the figure represent various additives. The hexagonal particles exemplified by 951a and 951b represent Li+ (ions).

The physical structure of a secondary battery as shown in Fig. 9A can be considered as an equivalent circuit 950 shown in Fig. 9B. The circuit shown in Fig. 9B includes at least an RC parallel circuit 961 that is an equivalent circuit of the SEI, an RC parallel circuit 962 that is an equivalent circuit of the anode (positive electrode part), and an RC parallel circuit 963 that is an equivalent circuit of the cathode (negative electrode part). In each RC parallel circuit, CPE _F , CPE _A , and CPE _C are easily affected by temperature, so special symbols are used for the capacitors in the figure.
Based on the above explanation with reference to FIGS. 9A and 9B, the deterioration state diagnosis of a secondary battery using a Cole-Cole plot is established.

In addition, as shown in Figure 9C, the information that can be read from the Cole-Cole plot is that the solution resistance, which has the fastest response, has a frequency band of 1k to 300 Hz, the electric double layer resistance on the electrode surface has a frequency band of 300 Hz to 0.1 Hz, and the diffusion resistance of the active material inside the electrode has a frequency band of 0.1 Hz or less. For this reason, Figure 9D is sometimes used as another model of the equivalent circuit (simple equivalent circuit model). This simple equivalent circuit model also makes it possible to read the characteristic parameter values inside the battery and determine the degree of battery deterioration.
As mentioned above, in the Cole-Cole plot shown in Figure 8B, the zero-crossing point (A in the figure) and the change point (depression B in the figure) tend to appear in the same positions for batteries with the same characteristics under the same conditions (temperature, battery voltage, etc., the same below). This also means that AC-IR has the property of being easily affected by temperature and SoC (State of Charge).

Based on these conventional methods, a battery degradation diagnosis system has been proposed that quickly diagnoses the degradation state of a secondary battery while reducing the computational load on the measuring device (Patent Document 1).

Specifically, a battery degradation diagnostic system is disclosed that includes a measuring device that measures battery state data related to the state of electrical characteristics of a secondary battery, and a diagnostic processing device that diagnoses the degradation state of the secondary battery, the diagnostic processing device being equipped with a storage means that stores in advance battery degradation information for identifying the degradation characteristics of the secondary battery, a receiving means that receives the battery state data from the measuring device, a calculation means that, when the receiving means receives the battery state data, refers to the battery degradation information stored in the storage means to calculate the degradation state of the secondary battery corresponding to the battery state data, and a transmission means that transmits to the measuring device a diagnostic result indicating the degradation state of the secondary battery calculated by the calculation means.

In addition, a method and device for determining the state of a secondary battery that can more accurately determine the state of a secondary battery have been proposed (Patent Document 2, Patent Document 3).

That is, Patent Document 2 discloses a method for determining the state of a secondary battery, which acquires a complex impedance measured by applying an AC voltage or AC current to a secondary battery to be determined, and determines the state of the secondary battery based on the acquired complex impedance, and includes a first capacity deviation determination step for determining the presence or absence of a first capacity deviation based on a comparison between a value of the complex impedance at a predetermined frequency and a first determination value used to determine a negative electrode capacity deviation, and a second capacity deviation determination step for determining the presence or absence of a second capacity deviation based on a comparison between the slope of the complex impedance with respect to the real axis in the diffusion resistance region and a second determination value used to determine a positive electrode capacity deviation when it is determined in the first capacity deviation determination step that the first capacity deviation has not occurred.

Patent Document 3 also discloses a secondary battery state determination method for determining a state in which a micro-short circuit tends to occur in a secondary battery, which is a state in which a micro-short circuit is likely to occur in the secondary battery, comprising: an electron transfer resistance measurement step for measuring the electron transfer resistance Rs of the secondary battery by applying a voltage or current of a predetermined frequency or higher to the electrode system of the secondary battery; an electrode distance quality determination step for comparing the electron transfer resistance Rs obtained in the electron transfer resistance measurement step with a preset lower threshold value Rs min; and a secondary battery state determination step for determining that the electrode distance is good and that the secondary battery is a good product if the electron transfer resistance Rs is equal to or higher than the lower threshold value Rs min in the electrode distance quality determination step.

JP 2020-205253 A JP 2019-049479 A JP 2021-174729 A

However, the conventional AR-IR measurement method has problems such as high cost, high power consumption, and long measurement time, so it has been necessary to judge the quality of a battery based on the battery characteristics measured using the DC-IR measurement method or simple charging and discharging. For example, in the conventional AC-IR measurement method, high power consumption is achieved by applying a sine wave to the secondary battery, which requires charging as well as discharging. Not only does a certain amount of power need to be consumed to charge a battery, but if the amount of charge and discharge current is increased to improve accuracy, further power is required, which results in an AC power supply being required, which makes the device larger. At the same time, this creates the problem of increased manufacturing costs.

In addition, when it comes to extending the measurement time, a particular issue was the accumulation of measurement time in the low-frequency band, as all frequencies must be measured.

Therefore, a method for supporting battery deterioration determination according to one embodiment of the present invention is a method for supporting deterioration determination of a battery by acquiring internal characteristics of the battery using an electronic circuit, the electronic circuit comprising: a control means for controlling ON/OFF discharge of the battery ; a voltage measurement means for measuring the voltage of the ON/OFF controlled battery to acquire voltage measurement data; and a current measurement means for measuring the amount of current discharged by the battery , the electronic circuit acquiring a voltage difference between the start and end of measurement for a period associated with the ON/OFF control, and dividing the acquired voltage difference by a discharge period in the period. The present invention is characterized in that it comprises a step of performing a correction process using correction data, a first conversion process on a result of the correction process, which performs a conversion process from the time domain to the frequency domain (Fourier transform process) or a conversion process from the time domain to the complex domain (Laplace transform process), a second conversion process on the correction data, which performs a conversion process from the time domain to the frequency domain (Fourier transform process) or a conversion process from the time domain to the complex domain (Laplace transform process), and a process for subtracting back the influence of the correction process by applying a result of the second conversion process to the result of the first conversion process.

The first transformation process is a Fourier transform process, and the second transformation process is characterized in that it generates second correction transformation data that takes into account a gain coefficient due to the influence of the window function for the first correction transformation data generated by performing a Fourier transform process by applying a window function to the correction data generated by the correction process.

The device is also characterized in that it selects and controls multiple frequencies for the ON/OFF control of the discharge, and does not perform the correction when controlling a relatively high frequency.

Furthermore, the relatively high frequency is at least 1 Hz or several Hz or more.

The battery deterioration determination support device according to one embodiment of the present invention has advantageous effects such as being able to provide a battery deterioration determination support device that is small, has low power consumption, and can significantly reduce measurement time.

1 is an explanatory diagram illustrating an example of the overall configuration of a system including a battery deterioration determination support device according to an embodiment of the present invention; 1 is an explanatory diagram illustrating an example of measurement of a current value in a battery deterioration determination support device according to an embodiment of the present invention. FIG. 1 is an explanatory diagram illustrating an example of measurement of a voltage value in a battery deterioration determination support device according to an embodiment of the present invention; 4 is a flowchart illustrating a measurement procedure in a battery deterioration determination support device according to one embodiment of the present invention. 4 is a flowchart illustrating a measurement procedure in a battery deterioration determination support device according to one embodiment of the present invention. 1 is an explanatory diagram illustrating an event that should be addressed by a battery deterioration determination support device according to an embodiment of the present invention. FIG. 1 is an explanatory diagram illustrating an example of a voltage graph used for the battery degradation determination support device and the like to take action according to one embodiment of the present invention; FIG. 1 is an explanatory diagram illustrating the state before and after processing by a battery deterioration determination support device according to one embodiment of the present invention; 1 is an explanatory diagram illustrating the state before and after processing by a battery deterioration determination support device according to one embodiment of the present invention; FIG. 2 is an explanatory diagram illustrating an example of a Cole-Cole plot based on the measurement results obtained by a battery deterioration determination support device according to one embodiment of the present invention. FIG. 11 is an explanatory diagram illustrating a comparison between an example Cole-Cole plot based on measurements using a battery deterioration determination support device according to one embodiment of the present invention and an example Cole-Cole plot based on measurements using a conventional AC-IR measuring instrument. FIG. 11 is an explanatory diagram illustrating an example of a measurement result obtained by an AC-IR measuring instrument. FIG. 11 is an explanatory diagram for explaining an example of generating a Cole-Cole plot from the measurement results obtained by an AC-IR measuring instrument. FIG. 1 is an explanatory diagram conceptually illustrating the physical structure of a conventional secondary battery. FIG. 1 is an explanatory diagram illustrating an example of an equivalent circuit based on the physical structure of a conventional secondary battery. FIG. 1 is an explanatory diagram illustrating the properties of an example Cole-Cole plot generated by measuring a conventional secondary battery with an AC-IR meter. FIG. 1 is an explanatory diagram illustrating an example of a simplified equivalent circuit based on the physical structure of a conventional secondary battery.

The battery deterioration determination support device and the like according to one embodiment of the present invention will be specifically described below with reference to the drawings. Note that the term "battery" is used to facilitate understanding of the present invention, but the present invention includes "batteries" and can be broadly applied to electric energy supply media that supply electricity or power.

Figure 1 shows an example of the overall configuration of a system including a battery deterioration determination support device according to one embodiment of the present invention.

As shown in FIG. 1, a system including a battery deterioration determination support device 100 according to one embodiment of the present invention includes a control unit 110 having a signal control I/F (111) for transmitting control signals for controlling ON/OFF of a switch circuit, etc., a current measurement I/F (112), a voltage measurement I/F (113), and an arithmetic processing unit 114 such as a CPU. The system also includes a switch circuit 120, a shunt resistor 130, a resistor (discharge resistor) 140, an A/D converter 150 for digitally converting the resistance value of the shunt resistor 130 (or the current value flowing through the shunt resistor 130), and an A/D converter 160 for digitally converting the voltage value of the storage battery 190.

In one embodiment of the present invention, all elements, circuits, and devices except for the storage battery 190 in FIG. 1 can be a battery deterioration determination support device or system according to the present invention. Also, the control unit 110 can be a battery deterioration determination support device according to one embodiment of the present invention.

(Operational procedure of the battery deterioration determination support device)
The operational features of the battery deterioration determination support device according to one embodiment of the present invention are as follows, but are not limited to these.
(1) The response characteristics are acquired using a square wave. A square wave includes the fundamental frequency and its higher-order frequency characteristics. For example, by applying a 100 Hz square wave and acquiring its characteristics, it is possible to simultaneously acquire the characteristics of higher-order odd terms such as 300 Hz and 500 Hz. The measurement time can be shortened by acquiring the higher-order terms using Fourier transform processing. The effect of this reduction in measurement time outweighs the increase in the amount of calculation required for the Fourier transform processing of the higher-order terms.
(2) Only the discharge characteristics are acquired for the reaction. In the conventional AC-IR measurement method, both the charge characteristics and the discharge characteristics are acquired, but in the battery deterioration determination support device according to one embodiment of the present invention, only the discharge characteristics are acquired. This feature makes it possible to reduce the burden on power consumption and the complexity associated with the large size of the device.

As a result of being provided with these features, the battery deterioration determination support device according to one embodiment of the present invention exhibits the following advantages.
(A) Compactness At the time of filing this application, a battery deterioration determination support device according to one embodiment of the present invention has been miniaturized to approximately 75.0 mm × 103.0 mm × 50 mm, and further miniaturization is possible.
(B) Low power consumption At the time of filing the present application, a battery deterioration determination support device according to one embodiment of the present invention was able to achieve a low power consumption of 325 mW.
(C) Shortening of measurement time There are four measurement points on the frequency band of the battery deterioration determination support device according to one embodiment of the present invention (although not limited to these, in one embodiment, four points are 0.2 Hz (F1), 1.5 Hz (F2), 12 Hz (F3), and 100 Hz (F4)). As a result, at the time of filing the present application, a measurement time of about 10 seconds and a communication time of about 20 seconds have been achieved (in one embodiment, the communication speed is about 100 Kbps and the data format is ASCII. By increasing the communication speed to, for example, 1 Mbps and changing the data format to binary, the communication time can be further shortened). Furthermore, the measurement points have been improved so that they can be freely selected, and ultimately operation is possible with two measurement points (0.3 Hz and 10 Hz in one embodiment. The upper limit of higher-order terms is thought to be about 11th to 13th orders. At higher-order frequencies than this, the plot will be buried in noise and will spread out).

(System configuration example of a battery deterioration determination support device)
An example of a system configuration of a battery deterioration determination support device according to one embodiment of the present invention is as follows.

(1) Configuration example 1
In the case of single cell measurement, one battery deterioration determination support device is used for one storage cell. A control device (not shown, can be realized by existing technology) is prepared to control the multiple battery deterioration determination support devices, and the multiple battery deterioration determination support devices and the control device are connected to each other via RS-485 or the like for communication processing. The processed result data may be analyzed in a cloud (not shown).

This configuration has the advantage of reducing the cost of the battery deterioration determination support device and eliminating the need for an analysis PC, etc.

(2) Configuration Example 2
Measurements are performed at only the two necessary points, and fitting processing is performed from the data obtained from the measurements. In addition, data in the non-measurement range due to the fact that the number of measurement points is limited to two is estimated from the fitting results described above. The fitting processing here is a processing employed in Cole-Cole plots, and a simulation is performed using an equivalent circuit expressed by a resistor and a plurality of parallel circuits (RC or RL) directly connected, and a set of a predetermined number of frequencies. Any known fitting processing can be employed for this fitting processing.

This configuration can meet the needs for AC-IR measurements and Cole-Cole plot calculations on battery production lines. Until now, meeting such needs required large equipment costs and the hassle of finding space to install the measuring device and securing an AC power source, but with this configuration, the solution can be achieved with equipment costs of just a few tens of thousands of yen, a space the size of a cigarette, and a USB power source.

In one embodiment of the present invention, fitting processing can be performed assuming a circuit such as the one below.

circuit=R0-p(R1,C1)-p(R2,C2)-p(R3,C3)-p(R4,L4)

where R0 etc. are resistors, C1 etc. are capacitors, L4 is an inductor, "p(R1, C1)" etc. are parallel connections of the elements in parentheses, and "-" is a series connection. Frequency sets such as [0.2, 1, 4.5, 7.5, 12.5, 36.5, 61, 97.5, 290, 879, 1465] are used as an example only for understanding the present invention.

Figure 2A shows an example of current value measurement in a battery deterioration determination support device according to one embodiment of the present invention. In the figure, the horizontal axis values from near zero to near 2000 show current values in the section where discharging is stopped. Similarly, the horizontal axis values from near 2000 to near 4000 show current values in the section where discharging is in progress. The waveform shown in the figure shows a square wave.

To be precise, since approximately 20 points are measured during the discharge period, the sampling points are 20 to 2068, and 2068 to 4116. That is, 2048 + 20 and 4096 + 20. The initial value of 20 can be determined as appropriate. In one embodiment of the present invention, it is considered desirable for this value to be around 2-3, or at most around 20.

Figure 2B shows an example of voltage measurement in a battery degradation determination support device according to one embodiment of the present invention. The horizontal axis of voltage in the figure corresponds to the horizontal axis of current in Figure 2A. In Figure 2B, the horizontal axis values from near zero to near 2000 indicate voltage values in the section during which discharging is stopped. Similarly, the horizontal axis values from near 2000 to near 4000 indicate voltage values in the section during discharging. The waveforms shown in the figure indicate voltage values corresponding to the states of discharging stopped and discharging.

Figures 3A and 3B show the measurement procedure in a battery degradation determination support device or the like according to one embodiment of the present invention. In the measurement in a battery degradation determination support device or the like according to one embodiment of the present invention, the flow shown in Figure 3A and the flow shown in Figure 3B are generally processed in parallel. More specifically, in one embodiment, steps S304 to S308 in Figure 3A and steps S351 to S353 in Figure 3B are processed in parallel.

(Overview of measurement process)
First, an overview of the entire measurement process including the measurement circuit and elements will be described. In the process flow of Figures 3A and 3B, a FET (Field Effect Transistor) is used for the SW (switching circuit). In one embodiment, the ON/OFF control of the FET is performed by PWM (Pulse Width Modulation) control, which performs power control using a semiconductor. It is possible to control the FET so that it is turned ON/OFF a specified number of times.

Although the present invention is not limited to this, in one embodiment, sampling of the voltage value/current value starts immediately before the SW turns OFF (first) after it is turned ON for the first time, and this is the starting point. After that, sampling ends immediately before the end of the final ON after a predetermined number of ON/OFF controls are repeated, and this is the end point. For example, if the predetermined number is four and the repeated waveform of four ON/OFF controls is measured, 4.5 waveforms will actually be taken. At this time, the FET is repeatedly processed as ON1/OFF1/ON2/OFF2/ON3/OFF3/ON4/OFF4/ON5, and the measurement starts immediately before the first ON (ON1) turns OFF (OFF1) (first) and ends immediately before the final ON turns OFF (just before the end of ON5), so the data actually used for the measurement is four waves of data.

Next, the measurement target data is read out at a specified cycle according to the I2C (Inter-Integrated Circuit) standard (every 10 μs in one embodiment). The read out data can be averaged as necessary. In one embodiment of the present invention, both the current and voltage are sampled at 4096 points and processed. In one embodiment, the data can be read out when a certain amount of measurement data has been accumulated in the PC. In another embodiment, the data can be processed by the arithmetic processing unit 114 in the control unit 110.
Furthermore, since the voltage and current are A/D values, a process of converting these into voltage and current values is carried out.

In addition, the voltage value gradually decreases as the discharge occurs, so the voltage values at the start and end points will be different.

Therefore, the voltage value is processed as follows to make the values at the start and end points match. First, the difference between the start and end voltages is found. Next, this difference is added evenly over the discharge period. In the above example, this would mean adding it evenly only to the ON section of four waveforms (adding correction voltage data). This process makes the voltage at the start and end points match.
Then, for the voltage value, a 4096-point fast Fourier transform (hereinafter, FFT) process is performed on the above-mentioned correction voltage data (this result is referred to as the correction voltage FFT process result), and for the current value, a 4096-point FFT process is performed as is.

Next, a window function is applied to the voltage-added signal (equally divided correction data) to align the start and end points, and FFT processing is performed. Next, gain correction is performed on the result of the FFT processing. For example, if a Hanming window is used, the gain is halved, so the result output using the Hanning window is doubled to perform gain correction. Finally, this correction result is added to the correction voltage FFT processing result, and the obtained result is used as the corrected voltage FFT processing result.

Then, the resistance value is calculated using the corrected voltage FFT processing results and current FFT processing results.

In addition, in other embodiments of the present invention, a general Fourier transform may be used instead of the above-mentioned FFT processing, and may be performed with 1024 points instead of 4096 points. Also, a publicly known alternative Laplace transform process may be used (the same applies to FFT processing below).

(Measurement process flow)
3A, when processing for one frequency is started in step S301, the process proceeds to step S302, where SW control is started. Next, the process proceeds to step S303, where the number of repetitions of the waveform to be measured is set. In one embodiment, when four waveforms are to be measured, the number of repetitions is set to four.

Next, the process proceeds to step S304, where the SW is controlled to be ON. Next, the process proceeds to step S305, where it is determined whether a specified time has elapsed. This time is determined by the frequency, and in the case of processing a frequency of 0.1 Hz, the specified time here is the 5 second ON section. If the answer is No in step S305, the process waits in that step until the specified time has elapsed, but if the answer is Yes in that step, the process proceeds to the next step.

In step S306, the SW is controlled to be OFF. Next, the process proceeds to step S307, where it is determined whether a specified time has elapsed. This time is determined by the frequency, and in the case of processing a frequency of 0.1 Hz, the specified time here is the OFF section of 5 seconds. If the answer is No in step S307, the process waits in that step until the specified time has elapsed, but if the answer is Yes in that step, the process proceeds to the next step.

In step S308, it is determined whether the number of repetitions of the waveform being measured has reached the number set in step S303. If the answer is No, the process returns to step S304, but if the answer is Yes, the process proceeds to step S309, where the measurement for one frequency is terminated.

On the other hand, as a process performed in parallel with the processes from step S304 to step S308, when the process is started in step S351 in FIG. 3B, measurement data is acquired in step S352, voltage values/current values are acquired as data in step S353, and the acquired data is saved in step S354.

The process of Fig. 3A may be automatically controlled by a microcomputer as PWM process. In this case, ON/OFF is automatically performed by the PWM control unit. The only settings made by the microcomputer are the repeat period and start and stop instructions. Once the repeat period is set and PWM is started, it will end after a predetermined time has elapsed. This end determination time corresponds to S308.
On the measurement side, while the ON/OFF control is performed in steps S305 to S308, steps S351 to S354 are repeatedly executed 4096×N times in this example, where N is the number of times of averaging.

The above-mentioned operating procedure is summarized in the table below.

Taking the second from the bottom (F2) in the table above as an example, the FET is turned ON/OFF at a cycle of 1.531 Hz (actually, it is controlled to be ON/OFF/ON). Looking at it on the time axis, it is ON for 327.68 ms, OFF for the next 327.68 ms, and then ON for a final 327.68 ms. Measurement starts (for example) about 7.68 ms after the first ON, and ends 320 ms after the last ON. During this time, measurements are taken every 10 us, so 65,536 measurements are taken over 655.36 ms, but by taking the average of 16, the total becomes 4096. This number 16 corresponds to N mentioned above. In other words, S351 to S354 are repeated a specified number of times, averaged as necessary, and the desired 4096 points (this number can also be freely selected) are obtained.

Figure 4A explains the events that should be handled by a battery deterioration determination support device according to one embodiment of the present invention. Figure 4A shows a state in which discharging (ON section of SW control) and discharging stop (OFF section of SW control) are repeated just under two times. Note that in the figure, the number of repetitions is set to one, and 4096 sampling points are performed during that repetition section (one ON/OFF control of the SW).

What is noteworthy about Figure 4A is that, first, the voltage at the start of the discharge in the second cycle is slightly lower than the voltage at the start of the discharge in the first cycle (part p in the figure). Secondly, the voltage at the end of the discharge section of the second cycle is lower by D than the voltage at the end of the discharge section of the first cycle.

The reason why the voltage drops every cycle is due to the SoC drop caused by the repeated discharge and cessation of discharge. Specific symptoms are as follows:
In Fig. 4A, if the SoC of the battery being measured at time t0 is x%, the SoC is maintained at x% during the period when discharge is stopped from time t0 to time t1. However, when discharge starts at time t1, the SoC gradually decreases during the period until time t2 when discharge is stopped. Then, at time t2, a voltage drop (D) occurs due to the decrease in SoC.

As mentioned in (Overview of measurement processing), this voltage drop (D) is the difference between the voltage value at the start point and the voltage value at the end point, so appropriate correction is required when performing FFT processing. The reason for this is that when FFT is processed over a finite period, if there is a discrepancy between the values at the start and end of the FFT transformation, unnecessary frequency components are generated (the problem is that the frequency characteristics of higher-order terms are calculated). Therefore, both values must be the same.

In general, when there is a discrepancy between the signal values at the start and end points, a method is used to apply a window function and make the start and end points zero to make them match. In other words, the window function is a function that makes only the center larger and brings the periphery closer to zero, and the discrepancy at both ends is treated as "deemed zero" to eliminate jumps. Note that this method only works when the signal characteristics are contained in the center. For example, it does not work well when important signals are contained in large amounts at the start and end points, as important information is lost. In the case of transient responses, applying a window function will eliminate much of the response signal obtained near the start and end points, so it is necessary to introduce a signal processing method that does not use window functions.

Based on the above considerations, in one embodiment of the present invention, the following correction process is performed to match the voltage value at the start point and the voltage value at the end point.

(Correction process)
(1) Calculate the difference between the start point voltage and the end point voltage (D).

(2) The voltage difference D calculated in (1) above is added evenly during the discharge period. Specifically, the correction voltage data is added evenly only to the ON section of the waveform (one waveform in FIG. 4) for the number of repetitions (one in FIG. 4).

(3) For the voltage value, a 4096-point FFT process is performed on the above correction voltage data to obtain the correction voltage FFT process result.

(4) For the current value, no correction is performed, and 4096-point FFT processing is performed on the acquired data to obtain the correction result.

(5) To align the start and end points, a window function is applied to the voltage-added signal (equal division correction data) and FFT processing is performed.

(6) Gain correction is performed on the results of FFT processing. For example, if a Hanming window is used, the gain is halved, so the result output using the Hanning window is doubled to perform gain correction.

(7) The correction result performed in (6) above is subtracted from the correction voltage FFT processing result, and the obtained result is regarded as the corrected voltage FFT processing result.

(8) The resistance value is calculated using the corrected voltage FFT processing results and current FFT processing results. This resistance value is a complex impedance consisting of a real resistance part (R) and an imaginary resistance part (X) for each frequency.

From another perspective, the above processes (1) to (8) can be divided into a (battery transient response processing phase) and a (post-processing phase), in which case they can also be explained as follows:

(Battery transient response processing phase)
(A1) The cell voltage immediately before the FET is turned off for the first time (time t0) is set to V0 (the voltage at the start of FFT). At this time, since the time from the start of FFT to the first discharge stop is short, it can be considered that there is no change in the SoC at this time (the cell voltage in this vicinity is constant at V0).

(A2) The time that the FET was turned OFF ends (slightly after time t1), and the FET is turned ON again (discharge begins). Then, the cell voltage begins to decrease rapidly. If the cell voltage at the end of the FFT (time t2) is V1, then if the SoC does not fluctuate due to discharge, V1 should be higher. It is not clear whether this voltage value is higher or lower than V0, but here we will assume it is V0.

(A3) Then, the difference (V1-V0) is proportionally subtracted from immediately after the FET is turned ON (in the diagram, the time when the voltage starts to drop a little after time t1), and a correction is applied so that the cell voltage becomes V0 at the end of the FFT (time t2). In this way, when applying the FFT, the start and end values can be made to match.

In this way, it is assumed that at time t0 the voltage is V0 and this value does not change until the discharge stops (because the time between these two is short). At this time, D = V0 - V1. In one embodiment of the present invention, this D is subtracted proportionately. Theoretically, there should be 2048 points, but in actual measurement, if the time between t0 and the end of discharge is 1 ms, 100 points are required (because it takes 10 us to measure one point), so 1948 points are effectively apportioned.

(Post-processing phase)
(B1) For the result of FFT processing, FFT processing is performed on a certain voltage drop from immediately after the FET is turned on to the end point of FFT. At this time, a window function is applied to points where the start point and end point do not coincide.

(B2) Although not limited to this, in one embodiment of the present invention, the function applied is a simple voltage drop function (a function in which the voltage decreases linearly over time).

The above is a comprehensive explanation of the measurement and correction embodiments, but below we will provide additional explanations to further deepen your understanding.

[Supplementary Note 1] The significance of matching the voltages at the start and end points As mentioned above, the correction process matches the voltages at the start and end points by equally dividing the difference value in the discharge period. If we further explain this process using an example, it will be as follows (1-1) to (1-4).

(1-1)
First, the starting voltage is V0, the ending voltage is V1, and the intermediate voltage is Vx. The times at which the voltages are respectively set are t0, t1, and tx. The respective sample numbers are 0, 4095, and Cx.

(1-2)
When the discharge is stopped, the intermediate voltage Vx increases from V0 to Vx. When the discharge is started at time tx, the intermediate voltage Vx decreases to V1.

(1-3)
Here, Vd = V0 - V1 (since the SoC decreases due to discharging, V0 > V1 can be set). In one embodiment of the present invention, 4096 points are acquired, but 20 to 30 samples from t0 to the end of discharging are also acquired, and these data are also included in the 4096 points. Therefore, the end point time is 20 to 30 samples before the second discharge stop.

(1-4)
If the number of samples in the section from Vx to V1 is S (=4095-Cx), the voltage at time ti is corrected as V=Vi+Vd/S*(Ci-Cx).
Furthermore, if ti = tx, then it is corrected as Vi = Vx + Vd/S * (Cx - Cx) = Vx, and if ti = t1, then Vi = V1 + Vd/S * (C1 (= 4095) - Cx) = V1 + Vd = V1 + V0 - V1 = V0, and the voltages at the start and end points are combined.

[Supplement 2] Necessity of processing to offset the influence of the difference When processing to match the voltages of the start point and end point is performed as explained in Supplement 1, processing to offset the influence of the difference is required. This is explained with examples in (2-1) to (2-3).

(2-1)
The processing in Addendum 1 is a method of proportionally adding the difference in the second discharge period. However, the voltage value proportionally added in this way is not the original voltage. It is merely a measure to avoid the occurrence of errors in higher-order terms when FFT processing is performed.

(2-2)
Therefore, after the final FFT processing, it is necessary to restore this corrected amount. The result of the addition process is a straight line where Vd is zero from t0 to tx, and decreases linearly from tx to t1. However, even if this is processed by FFT, it is still a straight line, so the desired result cannot be obtained.

(2-3)
Therefore, a window function is applied to the voltage graph data as shown in FIG. 4B to perform FFT processing, and then a gain is applied to calculate the amount of influence, which is then subtracted from the voltage FFT to return to an uninfluenced state.

[Supplementary Note 3] The significance of not performing correction processing at high frequencies Although the frequency of the characteristic to be obtained varies depending on the battery or discharge body to be measured, measurements are usually made by dividing the frequency range from 0.1 Hz to 1 kHz into 100 parts. In contrast, the battery deterioration determination support device according to one embodiment of the present invention has four measurement points on the frequency band (although not limited to these, in one embodiment, four points are used: 0.2 Hz (F1), 1.5 Hz (F2), 12 Hz (F3), and 100 Hz (F4)). No correction processing is performed for frequencies above 10 Hz, and the technical justification for this is as follows.

In other words, if you input a square wave of one wavelength at 100Hz, it will be ON for only 0.005 seconds when converted to real time (to be precise, since it goes ON/OFF/ON → end, it will be ON for 0.01 seconds). How much does the SoC decrease during this time? There is almost no decrease (see the slight voltage drop at part p in Figure 4A).

In fact, when measured at 100Hz, the starting voltage was 3.26431V, while the ending voltage was 3.26400V. During this time, there was a voltage drop of approximately 0.00031V, but the effect of this drop on the FFT processing was negligible.

Furthermore, in one embodiment, the actual measured values at frequencies less than 100 Hz were as follows: when measured at 12 Hz, the starting voltage was 3.26393 V and the end voltage was 3.26393 V; when measured at 1.5 Hz, the starting voltage was 3.26294 V and the end voltage was 3.26256 V; when measured at 0.2 Hz, the starting voltage was 3.26050 V and the end voltage was 3.25943 V.

While it can be determined that either of these has little effect on FFT processing, in one embodiment of the present invention, it is acceptable not to perform correction processing for frequencies of at least 1 Hz or a few Hz or higher. For example, the ON time at 1 Hz is 0.5 seconds, and for this length of time, the impact of a decrease in SoC is considered to be small.

[Supplementary Note 4] Significance of Calculating Parameters by Fitting There are four measurement points on the frequency band of the battery deterioration determination support device according to one embodiment of the present invention. In one embodiment, there are four measurement points: 0.2 Hz (F1), 1.5 Hz (F2), 12 Hz (F3), and 100 Hz (F4). Based on these measurement values, higher-order terms (first, third, fifth, ...) are calculated. On the other hand, for the slower frequency, 0.2 Hz, 0.6 Hz, 1.0 Hz, 1.4 Hz ... are obtained, but under the above conditions, no frequencies lower than 0.2 Hz are obtained. For the faster frequency, 300 Hz, 500 Hz, and 700 Hz can be obtained based on about 100 Hz as a reference, but for example, when it goes up to around 2 kHz or 3 kHz, the gain becomes too small and the width of the oscillation appears due to the influence of noise.

Therefore, the above-mentioned fitting process is performed, and the circuit used for this is as follows:
circuit=R0-p(R1,C1)-p(R2,C2)-p(R3,C3)-p(R4,L4)
It is.

The values of the fitting results can be substituted into the above circuit, and the frequency can be numerically shifted outside the measurement range to obtain the values of the higher-order terms by calculation. Although not limited to this, in one embodiment of the present invention, only two points, 0.5 Hz and 20 Hz, are measured, and the gap is filled to some extent with higher-order terms before parameter calculations are performed by fitting, making it possible to create a Cole-Cole plot sufficient for determining the deterioration of the battery being measured.

It is thought that the extent to which the above-mentioned 0.2Hz can be made faster depends on the characteristics of the battery. 0.2Hz means that the FET ON/OFF cycle is 0.2Hz, which means that one cycle takes 5 seconds. In some cases, it may be desirable to measure at 0.1Hz to create an accurate Cole-Cole plot that is useful for determining battery deterioration, but in that case it will take 10 seconds. If there is not much difference in the fitting results, processing from 0.5Hz can significantly reduce the processing time to 2 seconds, and in many cases it is more convenient overall. In that case, after performing the measurement at a frequency of 0.5Hz, it is acceptable to fill in the higher-order terms such as 1.5Hz, 7.5Hz, etc., and then perform further fitting to calculate the parameters, from which a Cole-Cole plot including the low-speed range (such as 0.1Hz) can be generated.

5A and 5B show the state before and after processing by the battery deterioration determination support device according to one embodiment of the present invention. Fig. 5A shows an example of the measurement of the voltage value before processing by the battery deterioration determination support device according to one embodiment of the present invention, and Fig. 5B shows an example of the result of performing correction processing by the battery deterioration determination support device according to one embodiment of the present invention on the measurement example of the voltage value shown in Fig. 5A. As shown in Fig. 5B, it can be confirmed that the difference D between the start point voltage and the end point voltage recognized in the measurement example of the voltage value shown in Fig. 5A is raised to zero with respect to the entire voltage during the discharging period by the above-mentioned voltage correction processing.
The above-mentioned FFT process is performed on the corrected voltage value shown in Fig. 5B. In other words, if the calculation is performed on the premise that the start point and the end point coincide with each other and this waveform is repeated, the inconvenience of calculating high-frequency abnormal data does not occur.

(Example of a Cole-Cole plot)
FIG. 6 shows an example of a Cole-Cole plot based on the measurement results obtained by a battery deterioration determination support device according to one embodiment of the present invention. In FIG. 6, plots a1 to a4 are measurement points of the first-order term. As shown in the scattering of the plots in the r1 region, the plots tend to scatter as the order increases. The plots in the r1 region can be derived by arithmetic operations as higher-order terms. The plots in the r2 region can be derived by arithmetic operations as higher-order terms of a4, and here, the third, fifth, seventh, and ninth orders are shown. Higher-order terms can also be obtained, but they overlap with the high-frequency first-order term a3, so the notation stops here.

Figure 7 shows a comparison between an example Cole-Cole plot based on measurements using a battery degradation determination support device according to one embodiment of the present invention and an example Cole-Cole plot based on measurements using a conventional AC-IR measuring device. In the figure, the plots concentrated on side A are the Cole-Cole plot (○ plot) for Battery 1 (a good battery) based on measurements using a conventional AC-IR measuring device, and the Cole-Cole plot (△ plot) for Battery 1 based on measurements using a battery degradation determination support device according to one embodiment of the present invention. The plots concentrated on side B are the Cole-Cole plot (○ plot) for Battery 2 (a deteriorated battery) based on measurements using a conventional AC-IR measuring device, and the Cole-Cole plot (△ plot) for Battery 2 based on measurements using a battery degradation determination support device according to one embodiment of the present invention.

Both measurements using a conventional AC-IR measuring device and measurements using a battery degradation determination support device according to one embodiment of the present invention show the characteristics of a good battery and a degraded battery in the same way, indicating that measurements using a battery degradation determination support device according to one embodiment of the present invention are in no way inferior to measurements using a conventional AC-IR measuring device.

In other words, a person skilled in the art can observe a Cole-Cole plot based on measurements made using a battery deterioration determination support device according to one embodiment of the present invention and make an accurate battery deterioration determination in the same manner as in the past.

The battery deterioration determination support device according to one embodiment of the present invention has been described above based on a specific example, but the present invention can also be embodied as a method or program for implementing the system or device, or as a storage medium on which a program is recorded (for example, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a CD-RW, a magnetic tape, a hard disk, or a memory card).

In addition, the implementation form of the program is not limited to application programs such as object code compiled by a compiler or program code executed by an interpreter, but may be in the form of a program module incorporated into an operating system.

Furthermore, the program does not necessarily have to be executed entirely by the CPU on the control board, but can also be configured so that some or all of the program is executed by another processing unit (such as a DSP) implemented on an expansion board or expansion unit added to the board as necessary.

All of the features described in this specification (including the claims, abstract, and drawings) and/or all steps of any method or process disclosed herein may be combined in any combination, except combinations in which the features are mutually exclusive.

Furthermore, each feature described in this specification (including the claims, abstract, and drawings) may be replaced with alternative features serving the same, equivalent, or similar purpose, unless expressly denied. Thus, unless expressly denied, each disclosed feature is merely one example of a generic series of identical or equivalent features.

Furthermore, the invention is not limited to the specific configuration of any of the embodiments described above. The invention extends to any novel feature or combination of features described herein (including the claims, abstract, and drawings), or any novel method or process step or combination of features described herein.

10 Battery deterioration determination support device 110 Control unit 111 Signal control I/F
112 Current measurement I/F
113 Voltage measurement I/F
120 Switch circuit, switching circuit

Claims (5)

A method for assisting in determining deterioration of a battery by acquiring internal characteristics of the battery using an electronic circuit, comprising:
The electronic circuit includes:
A control means for controlling ON/OFF discharging of the battery ;
a voltage measuring means for measuring a voltage of the battery under the ON/OFF control to obtain voltage measurement data;
a current measuring means for measuring the amount of current discharged from the battery ;
The electronic circuit
a voltage difference between a start time and an end time of measurement for a period associated with the ON/OFF control is obtained, and a correction process is performed using correction data obtained by dividing the obtained voltage difference by a discharge period in the period;
A first transformation process is performed on the result of the correction process, which is a transformation process from the time domain to the frequency domain (Fourier transform process) or a transformation process from the time domain to the complex domain (Laplace transform process);
A second transformation process is performed on the correction data, the second transformation process being a transformation process from a time domain to a frequency domain (Fourier transform process) or a transformation process from a time domain to a complex domain (Laplace transform process);
performing a process for canceling the influence of the correction process by applying a result of the second conversion process to a result of the first conversion process;
A method comprising: the first transformation process is a Fourier transformation process,
The method according to claim 1, characterized in that the second transformation process generates second corrected transformation data taking into account a gain coefficient due to the influence of a window function, for first corrected transformation data generated by applying a window function to the correction data generated by the correction process and performing a Fourier transform process. Selecting and controlling a plurality of frequencies for ON/OFF control of the discharge,
2. The method according to claim 1, wherein said correction is not performed during control of relatively high frequencies. The method according to claim 3, characterized in that the relatively high frequency is at least 1 Hz or several Hz or more. Apparatus for carrying out the method according to any one of claims 1 to 4 .

Images