A battery broadband impedance spectrum measurement method, system, device and storage medium

Abstract

The present invention discloses a method, system, device, and storage medium for measuring a battery's broadband impedance spectrum, relating to the technical field of battery impedance spectrum measurement. The method comprises the following steps: collecting battery voltage and current data, and determining multiple broadband impedance frequencies to be calculated based on the application scenario; wherein the voltage and current data are broadband step response signals; and calculating the broadband impedance spectrum based on the voltage and current data and the multiple broadband impedance frequencies based on a generalized S transform. By adjusting two important parameters in the generalized S transform, the present invention can effectively improve problems such as long low-frequency impedance measurement time and low signal-to-noise ratio in high-frequency impedance measurement associated with broadband step signals.

Description

Technical Field

[0001] The present invention relates to the technical field of battery impedance spectrum measurement, and in particular to a battery broadband impedance spectrum measurement method, system, device and storage medium. Background Art

[0002] As an electrochemical diagnostic tool, electrochemical impedance spectroscopy (EIS) can effectively decouple the dynamic processes of different electrochemical reactions within a battery, characterizing the battery's internal characteristics from multiple dimensions. This technology can be applied to battery status assessments, such as state of charge (SoC), state of health (SoH), internal electrode temperature, and remaining life. It is a core concept in the development of next-generation battery management systems (BMS). Therefore, rapidly, accurately, and cost-effectively acquiring battery EIS is fundamental to the widespread application of EIS-based battery status assessment methods.

[0003] Patent No. CN115308621B proposes a multi-signal, embedded lithium-ion battery electrochemical impedance spectroscopy (EIS) testing system and method. This system uses a switching power supply to convert 220V AC power into low-voltage, high-current DC power to power a linear excitation source. The linear excitation source can excite the battery based on a reference signal provided by a signal generator and adjust the excitation amplitude according to the operating conditions. Finally, the signal is processed using a variable intelligent optimization algorithm that processes periodic and time-varying signals to obtain the battery's EIS. Patent No. CN116008835B proposes a battery EIS testing method based on a single pulse response. When the battery has been idle for a set time, it is marked as ready for EIS measurement. The battery management system then initiates voltage and current data acquisition. After a period of time, the battery is discharged with a constant current short-duration pulse through the existing converter within the battery management system. Data acquisition is stopped after a period of time after the discharge ends. Finally, the parameters of the nonlinear analog equivalent circuit are calculated based on the voltage and current data, and the battery's EIS is inferred based on the circuit model. Reference Fast Calculation of Broadband Battery Impedance Spectra Based on S Transform of Step Disturbance and Response. IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION. 2022. pp. 3659-3672. For the first time, the S transform is applied to the calculation of the battery broadband impedance spectrum. By applying a current step excitation to the battery and sampling the battery's voltage and current data with a data acquisition card, the voltage and current response are processed using the S transform to obtain the battery's broadband impedance spectrum. Reference Fast Approach for Battery Impedance Identification Using Pseudo-Random Sequence Signals. IEEE TRANSACTIONS ON POWER ELECTRONICS. compares the linearity of pseudo-random binary sequences and ternary sequences in battery broadband impedance spectrum measurement, and filters the noisy impedance spectrum calculated by fast Fourier transform using a sliding average filter. Existing technologies all design measurement methods from three levels: excitation, measurement, and calculation. The excitation module is used to apply an excitation signal to the electrochemical system to generate a corresponding response signal. The measurement module is used to measure the excitation signal and the response signal to obtain a discrete digital signal that can be processed by the processor. The calculation algorithm processes the discrete response signal and the excitation signal to obtain the impedance spectrum of the electrochemical system.

[0004] Existing calculation algorithms for broadband impedance spectroscopy commonly use Fourier transform and S-transform, but the adjustment parameters of Fourier transform and S-transform are single parameters, which makes it impossible to adjust the time-frequency resolution according to actual conditions. As a result, problems such as long low-frequency impedance measurement time and low signal-to-noise ratio of high-frequency impedance measurement occur during the broadband impedance spectroscopy calculation process. Summary of the Invention

[0005] The present invention provides a battery broadband impedance spectrum measurement method, system, device and storage medium, which solves the problems that the adjustment parameters of existing Fourier transform and S transform are single parameters, resulting in the inability to adjust the time-frequency resolution according to actual conditions, resulting in long low-frequency impedance measurement time and low high-frequency impedance measurement signal-to-noise ratio during the broadband impedance spectrum calculation process.

[0006] The present invention provides a method for measuring a battery's broadband impedance spectrum, comprising the following steps:

[0007] Collecting voltage and current data of the battery and determining multiple broadband impedance frequencies to be calculated based on the application scenario; wherein the voltage and current data are broadband step response signals;

[0008] Based on the generalized S-transform, a broadband impedance spectrum is calculated from voltage and current data and multiple broadband impedance frequencies, including:

[0009] Selecting multiple Gaussian window functions corresponding to multiple broadband impedance frequencies, wherein the quasi-step signal is a signal in which the step signal is attenuated, shifting the mean center of each Gaussian window function to the step instant of the quasi-step signal, and multiplying the shifted Gaussian function with the corresponding voltage or current data to obtain multiple pre-processed voltage and current data;

[0010] Performing time-frequency conversion on the pre-processed voltage and current data to obtain frequency domain data of multiple voltage and current data;

[0011] The battery impedance corresponding to the broadband impedance frequency is obtained according to the frequency domain data of the voltage and current data, and the battery impedances of multiple broadband impedance frequencies constitute an impedance spectrum.

[0012] Preferably, the Gaussian window function is as follows:

[0013]

[0014] Where A is the slope, B is the intercept, and σ(f) is the Gaussian window function with broadband impedance at frequency f.

[0015] The pre-processed voltage or current data is expressed as follows:

[0016]

[0017] Where, is the voltage or current data after preprocessing, x is the voltage or current data before preprocessing, the exponential term is the Gaussian window function, n is the index value of the voltage and current digital data after sampling, the range is [-N, N], T s is the sampling frequency, f i is the i-th broadband impedance frequency;

[0018] The pre-processed voltage and current data are converted into time-frequency using the following formula:

[0019]

[0020] Where x(τ) is the time domain signal, X(f,t) represents the complex-valued frequency domain information of the harmonic of the broadband impedance with frequency f at time t, g(τ-t,f) is the Gaussian function, exp(-j2πfτ) is the sine term, and τ is the time variable of the time domain signal.

[0021] Preferably, the battery impedance is obtained by the following formula:

[0022]

[0023] Where U(f,t) and I(f,t) are the complex-valued frequency domain information of the voltage and current of the harmonic with frequency f at the same time t.

[0024] Preferably, the voltage and current data of the battery are collected through a discharge circuit, wherein the discharge circuit includes a MOSFET, a driver chip, a discharge resistor, an STM32 single-chip microcomputer, a current transformer and a measured battery, and the measured battery, the MOSFET, the resistor and the current transformer are connected in series in sequence; the STM32 single-chip microcomputer is used to output a command signal to control the MOSFET to be turned on and off; the driver chip is used to enhance the command signal, and input the enhanced voltage command signal between the gate and the drain of the MOSFET; the output terminals of the voltage measurement probe and the current transformer are connected to the data acquisition part; the single-chip microcomputer sends a turn-on command at time t0, and starts sampling voltage and current data from t0-T0, and ends at t0+T0, and the single-chip microcomputer sends a turn-off command; each time the MOSFET is turned on to before it is turned off and before it is turned off to before it is turned on is a discharge cycle.

[0025] Preferably, the method for calculating the broadband impedance spectrum at a frequency greater than the maximum broadband impedance frequency includes:

[0026] The MOSFET is turned on and off multiple times to collect voltage and current data over multiple discharge cycles;

[0027] For each broadband impedance frequency, a broadband impedance spectrum is calculated based on a plurality of voltage and current data based on a generalized S-transform to obtain a plurality of impedance values;

[0028] Averaging multiple impedance values ​​corresponding to each broadband impedance frequency, the average value being the impedance of each broadband impedance frequency;

[0029] The battery impedance at multiple impedance frequencies constitutes the impedance spectrum.

[0030] Preferably, for a broadband impedance frequency lower than the minimum broadband impedance frequency, the resistance of the discharge resistor is increased, and voltage and current data of multiple discharge cycles are collected.

[0031] Preferably, before performing time-frequency conversion on the pre-processed voltage and current data, it is necessary to accelerate them through FFT, which specifically includes the following steps:

[0032] The pre-processed voltage and current data are padded with zeros until the data points reach 2 m ;

[0033] Perform fast Fourier transform on the voltage and current data after the end is padded with zeros to obtain frequency domain data of multiple voltage and current data; as shown in the following formula:

[0034]

[0035] in,

[0036] k≈f i T s 2 m

[0037] Where, X MST (f i ) is the frequency of the corresponding signal x(t) is f i The complex-valued frequency domain information of the harmonics of It is a discrete digital signal with zero padding at the end;

[0038] The battery impedance corresponding to the broadband impedance frequency is obtained based on the frequency domain data of the voltage and current data.

[0039] A battery broadband impedance spectrum measurement system, comprising:

[0040] An acquisition module is used to collect voltage and current data of the battery and determine multiple broadband impedance frequencies that need to be calculated based on the application scenario; wherein the voltage and current data are broadband step response signals;

[0041] a calculation module for calculating a broadband impedance spectrum based on voltage and current data and a plurality of broadband impedance frequencies based on a generalized S-transform;

[0042] The calculation module includes:

[0043] A preprocessing module is configured to select multiple Gaussian window functions corresponding to multiple broadband impedance frequencies, wherein the quasi-step signal is a signal in which the step signal is attenuated, and to shift the mean center of each Gaussian window function to the step instant of the quasi-step signal, and to multiply the shifted Gaussian function by the corresponding voltage or current data to obtain multiple preprocessed voltage and current data; wherein the slope of the Gaussian window function is selected;

[0044] A conversion module, used to perform time-frequency conversion on the pre-processed voltage and current data to obtain frequency domain data of multiple voltage and current data;

[0045] The calculation module is used to obtain the battery impedance corresponding to the broadband impedance frequency according to the frequency domain data of the voltage and current data. The battery impedances of multiple broadband impedance frequencies constitute an impedance spectrum.

[0046] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for measuring the wide-band impedance spectrum of a battery is implemented.

[0047] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the above-mentioned battery broadband impedance spectrum measurement method.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention uses a broadband step response signal and calculates the broadband impedance spectrum based on voltage and current data and multiple broadband impedance frequencies through a generalized S-transform. By adjusting two important parameters in the generalized S-transform, the present invention can effectively improve the problems of broadband step signals, such as long low-frequency impedance measurement time and low high-frequency impedance measurement signal-to-noise ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 This is a flow chart of a method for measuring a wide-band impedance spectrum of a battery according to the present invention;

[0052] Figure 2 This is a block diagram of voltage and current signal preprocessing based on generalized S-transformation of the present invention;

[0053] Figure 3This is a block diagram of signal time-frequency analysis based on generalized S transform of the present invention;

[0054] Figure 4 Schematic diagram showing the comparison between the calculated impedance spectrum and the reference impedance spectrum of a 25Ah lithium iron phosphate battery according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0056] The present invention discloses a method for measuring a wide-band impedance spectrum of a battery, comprising the following steps:

[0057] Step 1: Sample the voltage and current data of the battery.

[0058] In the battery-to-resistor discharge circuit, a four-terminal method is used to sample the voltage and current signals across the battery. The discharge circuit consists of a high-speed MOSFET, a high-speed MOSFET driver chip, a discharge resistor, an STM32 microcontroller, and the battery under test. The STM32 microcontroller outputs a command signal to control the MOSFET's conduction and shutdown. The driver chip amplifies the command signal and inputs the amplified voltage command signal between the MOSFET's gate and drain, ensuring reliable conduction and shutdown. The battery under test, MOSFET, resistor, and current transformer are connected in series. The output terminals of the battery voltage measurement probe and current transformer are connected to the data acquisition unit. The device does not require a high-precision excitation source; its main circuitry consists of only a resistor and a MOSFET. Its compact design, minimal switches, low cost, and compact size offer promising applications.

[0059] Assume that at time t0, the MCU sends a turn-on command and starts sampling voltage and current data from t0-T0. The voltage and current data sampling ends at t0+T0, and the MCU sends a turn-off command. T0 depends on the minimum impedance frequency required for calculation, as shown in formula (1):

[0060]

[0061] The voltage and current data recorded in [t0-T0, t0+T0] are used to calculate the frequency greater than f min The battery impedance. Assume that the sampling period of the data acquisition device is T s According to the Nyquist sampling theorem, the maximum frequency that can be analyzed after the data time-frequency conversion is Therefore, when the sampling period is Ts When the voltage and current data are used to calculate the frequency less than f max The battery impedance. The step signal contains an infinite number of frequency components.

[0062] Step 2: Determine multiple impedance frequencies.

[0063] Determine the impedance frequency (f1, f2, ..., f i , f i+1 ,…,f n ), the impedances at different frequencies constitute the final broadband impedance spectrum. Usually, equal intervals of 50Hz power frequency are avoided, or equal intervals of every 10 times the frequency division are selected. If a sampling frequency of 10kHz is used, according to the Nyquist sampling theorem, the frequency range that can be measured by this solution is 0.1Hz-2kHz. Impedances above 2kHz have reduced accuracy due to insufficient signal-to-noise ratio, and impedances below 0.1Hz have significant deviations due to non-negligible changes in the battery SoC caused by excessive discharge time.

[0064] The problem of insufficient signal-to-noise ratio in the ultra-high frequency part can be solved by regularly turning on and off the MOSFET multiple times. The data before each turn-on (turn-off) and before each turn-off (turn-on) can be used to calculate a broadband impedance spectrum. After multiple turns-on-off-on-…-off cycles, multiple broadband impedance spectrum calculations are performed. For each frequency f i , the frequency obtained each time is f i The impedance values ​​are averaged and the average value is f i The corresponding impedance value is calculated by doing the same operation for other frequency impedance calculations to obtain the average impedance value at different frequencies, and finally a high signal-to-noise ratio wide-band impedance spectrum of the battery is formed, so that the maximum frequency of the measurable impedance is close to the theoretical maximum frequency f max , if you still want to increase the maximum frequency of impedance calculation, you must increase the sampling frequency. For the problem of non-negligible SoC changes during ultra-low frequency band measurements, you can improve it by reducing the discharge current, that is, increasing the resistance of the discharge resistor. The current and voltage of the battery are approximately a step response under the action of the switch on and off, with continuous frequency components. Compared with multi-frequency signals such as multi-sine, periodic square waves, triangle waves, etc., more frequency impedances can be obtained within the same measurement time. Therefore, the influence of the resistance of the discharge resistor on the harmonic amplitude of the current signal can be approximately expressed by formula (2):

[0065]

[0066] Where I(f) is the approximate value of the current harmonic amplitude at frequency f, U OCV is the open circuit voltage of the battery, and R is the resistance of the discharge resistor.

[0067] Step 3: The sampled voltage and current data and all impedance frequencies are input into the impedance calculation process based on the generalized S-transform. This process includes data preprocessing, time-frequency conversion of the preprocessed data, and dividing the frequency-domain voltage by the frequency-domain current to obtain the impedance value at the corresponding frequency.

[0068] Whether a single turn-on is used to obtain the broadband impedance spectrum of the battery, or multiple turn-on and turn-off are used to obtain the impedance spectrum, the preprocessing process of the voltage and current data obtained each time the turn-on (turn-off) is exactly the same, such as Figure 2 As shown, the frequency is calculated as f using single conduction data. i The impedance value of is used as an example to illustrate the preprocessing process of voltage and current data. Assume that the time interval of the data recording is [t0-T0, t0+T0]. According to the calculated impedance frequency, a suitable Gaussian window function is selected, that is, a suitable Gaussian function standard deviation parameter is selected. The calculation formula of the standard deviation parameter is shown in the following formula:

[0069]

[0070] The slope A and intercept B should be determined based on the principle of minimizing the deviation between the calculated broadband impedance spectrum and the standard impedance spectrum. The deviation is expressed as the root mean square error (RMSE):

[0071]

[0072] Where Z ref (f i ) represents the frequency f i The standard reference impedance value, Z meas (f i ) represents the frequency f i The calculated broadband impedance value, M represents the total number of frequency points for the calculated broadband impedance.

[0073] When the parameters A = 1 and B = 0 are determined according to the above principles, the generalized S transform at this time is called the standard S transform. In general, A = 0.75 and B = 0.0055 are most appropriate, but calculations need to be made according to specific circumstances. Combined with the analysis of formula (1), it can be concluded that: when B is larger, the required measurement time before and after the conduction (shutdown) is shorter under the condition that the minimum broadband impedance frequency remains unchanged, that is, the discharge time or static time of the battery can be shortened. When A is smaller, the time-frequency resolution of the high-frequency band is improved, as shown in formula (5):

[0074]

[0075] in is the window function added in the signal preprocessing process, which is a Gaussian function in the generalized S transform. It represents the Fourier transform of the analyzed signal, and Δf represents the minimum identifiable frequency interval. The smaller the interval, the higher the frequency resolution, which ultimately leads to improved measurement accuracy of high-frequency impedance.

[0076] Afterwards, the mean center of the Gaussian function is shifted to the instantaneous moment t0 of the quasi-step signal, that is, to coincide with the data point sampled at the instant the MOSFET is turned on (off). After the coincidence, the corresponding Gaussian window function and the value of the sampled data at the same moment are multiplied, and the result of the multiplication is used as the preprocessed data value at that moment. Since the sampled data is discrete data, the process of multiplying the Gaussian window and the sampled data can be expressed by the following formula:

[0077]

[0078] Where, represents the data after preprocessing, x represents the data before preprocessing, the exponential term is the Gaussian window function, and this expression takes t0 = 0 as an example to represent the data preprocessing process, so the mean center of the Gaussian window function in the formula is also shifted to t0 = 0, n is an integer (…, -1, 0, 1, …) and ranges from [-N, N]; N is shown in the following formula, which is obviously derived from formula (1):

[0079]

[0080] The preprocessing process is applied to the voltage and current signal sampling data at the same time, and the preprocessed data is used to calculate the impedance frequency f i If the process is used to calculate the voltage and current data of the frequency f i+1 The impedance of the i Replace with f i+1 .

[0081] After signal preprocessing, time-frequency analysis is performed to convert the time domain signal into the frequency domain, and then processed by formula (10), the frequency f can be obtained. i The battery impedance.

[0082]

[0083] Equation (8) is the general formula for the generalized S-transform, where x(τ) is the time domain signal, X(f,t) represents the complex-valued frequency domain information of the harmonic of signal x(τ) at time t, g(τ-t,f) is a Gaussian function, expressed as shown in Equation (9), and exp(-j2πfτ) is the sine term. In Equation (9), σ(f) is the standard deviation of the Gaussian function, which is a function of the signal harmonic frequency f and corresponds to the frequency of the battery impedance calculated in this scheme. Time-frequency analysis of the voltage and current signals at the battery terminals yields the complex-valued frequency domain information of the voltage and current harmonics at frequency f at the same time t, expressed as U(f,t) and I(f,t), respectively. The impedance at frequency f can then be expressed as Equation (10).

[0084] If the above-mentioned multiple on-off signal-to-noise ratio optimization process is continued, formula (10) becomes formula (11), where N represents the total number of on-off times and j represents the number of each on-off time. The impedance values ​​at different frequencies obtained by the formula (11) are significantly improved in signal-to-noise ratio compared with the impedance values ​​at different frequencies obtained by the formula (10).

[0085]

[0086] Since the impedance frequency calculated by this scheme is relatively wide, and the same on (off) sampling data is used when calculating the impedance of different frequencies, that is, the data sampling frequency for calculating the impedance of different frequencies is constant, and the calculation of high-frequency impedance requires a higher sampling rate. At the same time, the calculation of low-frequency impedance according to formula (1) requires a longer sampling time. In this way, calculating low-frequency impedance based on high-sampling rate data will result in a large number of calculated data points, so the calculation time will be longer. Therefore, it is necessary to accelerate the operation through FFT, and the process is as follows: Figure 3 As shown. Because FFT can only be used to process discrete digital signals with data amounts that are integer powers of 2, the pre-processed signal is padded with zeros until the number of data points reaches 2. m , and 2 m Satisfy inequality (12). After that, the range of n of the preprocessed data becomes [-N,2 m -N-1],[N+1,2 m -N-1] all data values ​​are 0;

[0087] 2(2N+1)>2 m >2N+1 (12)

[0088] Afterwards, for 2 m The data points are fast Fourier transformed, as shown in formula (13), X MST (f i ) is the frequency of the corresponding signal x(t) is f i The complex-valued frequency domain information of the harmonics of is the discrete digital signal obtained by the preprocessing process of S5 and the zero-filling process of S6, and the expression of k is shown in formula (14);

[0089]

[0090] k≈f i T s 2 m (14)

[0091] Formula (14) shows that k is equal to f i T s 2 m The nearest integer; this calculation process is used for preprocessing voltage data and current data at the same time, and the UF and IF frequency domain sequences are obtained respectively; finally, according to formula (15), the frequency f can be obtained i If the process is to be used to calculate the impedance value of the frequency f i+1 The impedance of the above process can be obtained by simply changing f i Replace with f i+1 .

[0092]

[0093] The above process and Figure 1 The impedance calculation process in the calculation frequency is f i Taking the impedance value as an example, the calculated frequency is f i+1 The impedance value still follows the S5 and S6 process, just change all f i Replace with f i+1 The impedance calculation process for other frequencies is the same. The impedance values ​​of all frequencies calculated are plotted on a graph (with the real part as the horizontal axis and the imaginary part as the vertical axis) to obtain the broadband impedance spectrum of the battery. Figure 4 The comparison between the battery broadband impedance spectrum calculated by the method proposed in this scheme and the standard impedance spectrum calculated by the electrochemical workstation is shown to prove the feasibility of this scheme.

[0094] Based on the same concept, the present invention also provides a battery broadband impedance spectrum measurement system, including an acquisition module and a calculation module.

[0095] The acquisition module is used to collect the voltage and current data of the battery and determine multiple broadband impedance frequencies that need to be calculated according to the application scenario; wherein the voltage and current data are broadband step response signals.

[0096] The calculation module is used to calculate the broadband impedance spectrum based on the voltage and current data and multiple broadband impedance frequencies based on the generalized S-transform.

[0097] The calculation module includes a preprocessing module, a conversion module and a calculation module.

[0098] The preprocessing module is used to select multiple corresponding Gaussian window functions according to multiple broadband impedance frequencies. The quasi-step signal is a signal in which the step signal is attenuated. The mean center of each Gaussian window function is shifted to the step moment of the quasi-step signal, and the shifted Gaussian function is multiplied with the corresponding voltage or current data to obtain multiple preprocessed voltage and current data; wherein, the slope of the Gaussian window function is selected.

[0099] The conversion module is used to perform time-frequency conversion on the preprocessed voltage and current data to obtain frequency domain data of multiple voltage and current data.

[0100] The calculation module is used to obtain the battery impedance corresponding to the broadband impedance frequency according to the frequency domain data of the voltage and current data. The battery impedances of multiple broadband impedance frequencies constitute an impedance spectrum.

[0101] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the above-mentioned battery broadband impedance spectrum measurement method is implemented.

[0102] The present invention also provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned battery broadband impedance spectrum measurement method is implemented.

[0103] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0104] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for measuring battery broadband impedance spectrum, characterized in that: The following steps are involved: Collecting voltage and current data of the battery and determining multiple broadband impedance frequencies to be calculated based on the application scenario; wherein the voltage and current data are broadband step response signals; Based on the generalized S-transform, a broadband impedance spectrum is calculated from voltage and current data and multiple broadband impedance frequencies, including: Multiple Gaussian window functions corresponding to multiple broadband impedance frequencies are selected, and two parameters in the Gaussian window function are adjusted at the same time; the Gaussian window function is shown in the following formula: Where A is the slope ，B is the intercept, The broadband impedance frequency is f Gaussian window function; slope A and intercept B It should be determined based on the principle of minimum deviation between the calculated broadband impedance spectrum and the standard impedance spectrum; Shifting the mean center of each adjusted Gaussian window function to the step instant of the quasi-step signal, wherein the quasi-step signal is a signal in which the step signal is attenuated, and multiplying the shifted Gaussian function with the corresponding voltage or current data to obtain a plurality of pre-processed voltage and current data; Performing time-frequency conversion on the pre-processed voltage and current data to obtain frequency domain data of multiple voltage and current data; The battery impedance corresponding to the broadband impedance frequency is obtained according to the frequency domain data of the voltage and current data, and the battery impedances of multiple broadband impedance frequencies constitute an impedance spectrum.

2. A battery broadband impedance spectrum measurement method according to claim 1, characterized in that: The pre-processed voltage or current data is expressed as follows: Where, is the pre-processed voltage or current data, is the voltage or current data before preprocessing, the exponential term is the Gaussian window function, The index value of the voltage and current digital data after sampling, the range is , is the sampling frequency, For the i A broadband impedance frequency; The pre-processed voltage and current data are converted into time-frequency using the following formula: Where, is the time domain signal, Indicates signal In time The broadband impedance frequency is The complex-valued frequency domain information of the harmonics of is a Gaussian function, is the sine term, is the time variable of the time domain signal.

3. A battery broadband impedance spectrum measurement method according to claim 2, characterized in that: The battery impedance is obtained by the following formula: Where, and At the same time The frequency is The complex-valued frequency domain information of the harmonics of voltage and current.

4. A battery broadband impedance spectrum measurement method according to claim 1, characterized in that: The voltage and current data of the battery are collected through a discharge circuit, which includes a MOSFET, a driver chip, a discharge resistor, an STM32 single-chip microcomputer, a current transformer and a battery under test, wherein the battery under test, the MOSFET, the discharge resistor and the current transformer are connected in series in sequence; the STM32 single-chip microcomputer is used to output a command signal to control the MOSFET to be turned on and off; the driver chip is used to enhance the command signal and input the enhanced voltage command signal between the gate and drain of the MOSFET; the output terminals of the voltage measurement probe and the current transformer are connected to the data acquisition part; At this moment, the STM32 microcontroller sends a turn-on command and Start sampling voltage and current data until At the end, the STM32 microcontroller sends a shutdown command; each time the MOSFET is turned on to turned off and before turned off to turned on is a discharge cycle.

5. A battery broadband impedance spectrum measurement method according to claim 4, characterized in that: The calculation method for broadband impedance spectrum with frequency greater than the maximum broadband impedance frequency includes: The MOSFET is turned on and off multiple times to collect voltage and current data over multiple discharge cycles; For each broadband impedance frequency, a broadband impedance spectrum is calculated based on a plurality of voltage and current data based on a generalized S-transform to obtain a plurality of impedance values; Averaging multiple impedance values ​​corresponding to each broadband impedance frequency, the average value being the impedance of each broadband impedance frequency; The battery impedance at multiple impedance frequencies constitutes the impedance spectrum.

6. A battery broadband impedance spectrum measurement method according to claim 4, characterized in that: For broadband impedance frequencies less than the minimum broadband impedance frequency, the resistance of the discharge resistor is increased, and voltage and current data are collected for multiple discharge cycles.

7. A battery broadband impedance spectrum measurement method according to claim 2, characterized in that: Before performing time-frequency conversion on the pre-processed voltage and current data, it is necessary to accelerate it through FFT, which includes the following steps: The pre-processed voltage and current data are padded with zeros until the data points reach ,and The following inequality is satisfied: Perform fast Fourier transform on the voltage and current data after the end is padded with zeros to obtain frequency domain data of multiple voltage and current data; as shown in the following formula: in, Where, For the corresponding signal The frequency is The complex-valued frequency domain information of the harmonics of It is a discrete digital signal with zero padding at the end; The battery impedance corresponding to the broadband impedance frequency is obtained based on the frequency domain data of the voltage and current data.

8. A battery broadband impedance spectrum measurement system, characterized in that: include: The acquisition module is used to collect the voltage and current data of the battery and determine the multiple broadband impedance frequencies that need to be calculated according to the application scenario; The voltage and current data are both wide-frequency step response signals; a calculation module for calculating a broadband impedance spectrum based on voltage and current data and a plurality of broadband impedance frequencies based on a generalized S-transform; The calculation module includes: A preprocessing module is used to select multiple Gaussian window functions corresponding to multiple broadband impedance frequencies, where the quasi-step signal is a signal with attenuation, and to shift the mean center of each Gaussian window function to the step moment of the quasi-step signal. The shifted Gaussian function is multiplied by the corresponding voltage or current data to obtain multiple preprocessed voltage and current data; The Gaussian window function is shown below: Where A is the slope ，B is the intercept, The broadband impedance frequency is f Gaussian window function; slope A and intercept B It should be determined based on the principle of minimum deviation between the calculated broadband impedance spectrum and the standard impedance spectrum; A conversion module, used to perform time-frequency conversion on the pre-processed voltage and current data to obtain frequency domain data of multiple voltage and current data; The calculation module is used to obtain the battery impedance corresponding to the broadband impedance frequency according to the frequency domain data of the voltage and current data. The battery impedances of multiple broadband impedance frequencies constitute an impedance spectrum.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for measuring the wide-band impedance spectrum of a battery according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the battery broadband impedance spectrum measurement method according to any one of claims 1 to 7 is implemented.

Images