Lithium battery electrochemical impedance spectroscopy online measurement and estimation method and system based on charging pile

Abstract

The invention relates to a lithium battery electrochemical impedance spectroscopy online measurement and estimation method and system based on a charging pile, and belongs to a battery detection technology. The system comprises a measurement module, an estimation module and a result output module. The measurement module comprises a signal generation sub-module, a waveform amplification sub-module and a response processing sub-module; a pulse control signal is inserted into the signal generation sub-module during charging, and is amplified into a measurement excitation signal by the waveform amplification sub-module to be applied to a battery; the response processing sub-module samples the response signal and calculates an impedance spectrum. The estimation module comprises a neural network sub-module and an impedance spectrum estimation sub-module; the neural network sub-module is used for training a physical information neural network by using impedance response to output Randles circuit model parameters to the estimation sub-module, and the change of the electrochemical impedance spectrum is predicted. And the result output module displays and stores the measured and estimated values of the impedance spectrum. According to the invention, the impedance values of the to-be-measured battery pack under multiple frequencies can be measured and estimated on line, and the estimation of the impedance spectrum is physically interpretable.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery detection, and particularly relates to a method and system for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile. Background Art

[0002] Lithium batteries have the advantages of high energy density, good cycle performance, low self-discharge rate, etc., and are widely used as the main energy storage components of electric vehicles. To meet the increasing charging demand of in-vehicle lithium batteries, while pursuing the goal of fast charging, the charging pile also needs to take into account the real-time health state (SOH) of the lithium battery. Electrochemical Impedance Spectroscopy (EIS) is an important tool for analyzing the internal electrochemical reaction state of lithium batteries. With its detection in a wide frequency domain (usually in the range of kilohertz to millihertz), it can obtain the coupled kinetic information such as mass transfer, ion diffusion, and charge transfer during the battery charging process.

[0003] For traditional EIS measurement, when the lithium battery is in an equilibrium state, a small-amplitude voltage signal or current signal (so as not to affect the state of charge of the lithium battery during measurement) needs to be applied to the battery under test through an expensive waveform generator. The power amplifier circuit in the electrochemical workstation amplifies the measurement signal into an excitation signal and applies it across the battery under test. Then, the voltage response and current response of the lithium battery under the measurement excitation are sampled, and the time-frequency domain signal conversion is completed through discrete Fourier transform. According to the amplitude ratio and phase angle difference of the response signal of the battery relative to the excitation signal, the impedance response at different frequencies is measured, and then an impedance spectrum curve is plotted. Traditional EIS measurement not only has low efficiency but also is only applicable to measuring the impedance response data of the battery in an offline state.

[0004] Estimating the electrochemical impedance spectrum of a lithium battery usually involves equivalent model analysis, which is called the model-based analysis method here, including the equivalent circuit model and the electrochemical impedance model. The equivalent circuit model characterizes the battery impedance response through circuit elements and has a simple circuit structure and a small number of parameters; the electrochemical impedance model describes the internal mechanism of the battery, such as electrode process kinetics and ion diffusion process.

[0005] The model-based method is usually only applicable to the offline analysis of the electrochemical impedance spectrum of lithium batteries. Although the data-based method eliminates the analysis complexity of the circuit structure and chemical mechanism, it lacks a certain degree of interpretability in physical principles. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide a method and system for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile, which uses the existing control circuit and power conversion circuit of the charging pile for impedance measurement; measures the impedance values of the battery pack to be measured at multiple frequencies at one time, improves the efficiency of impedance measurement, and avoids generating the impedance value at a single measurement frequency; can quickly and accurately perform online prediction of the electrochemical impedance spectrum according to the offline impedance data provided by the battery management system and the measurement excitation signal input to the battery to be measured, and can simultaneously provide physical interpretability for the online prediction of the electrochemical impedance spectrum by the physics-informed neural network and the Randles circuit.

[0007] To achieve the above object, the system of the embodiment of the present invention is realized through the following technical solutions: A system for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile, including an impedance spectrum measurement module, an impedance spectrum estimation module, and a result output module;

[0008] Among them, the impedance spectrum measurement module includes the following sub-modules:

[0009] A signal generation sub-module, which modifies the charging strategy of the battery to be measured at the charging pile end, inserts a small-amplitude pulse control signal required for impedance measurement during the charging process of the battery to be measured, and transmits it to the waveform amplification sub-module;

[0010] A waveform amplification sub-module, which amplifies the small-amplitude pulse control signal into a wide-band large-amplitude measurement excitation signal through the power conversion circuit of the charging pile and applies it to both ends of the battery to be measured;

[0011] A response processing sub-module, which is used to sample the voltage response and current response of the battery to be measured under the measurement excitation signal, compensate for the time delay and voltage drop of the current and voltage responses, and then perform data processing and calculate the impedance response of the battery to be measured;

[0012] The impedance spectrum estimation module includes the following sub-modules:

[0013] A neural network sub-module, which is used to retrieve the measured charge and discharge historical data, extract the impedance parameter vector, offline train the physics-informed neural network model, output the Randles circuit model parameters, and transmit them to the impedance spectrum estimation sub-module;

[0014] An impedance spectrum estimation sub-module, which combines the Randles circuit model parameters and the current measurement excitation response signal to perform online estimation of the electrochemical impedance spectrum of the battery to be measured and predict the future electrochemical impedance spectrum curve;

[0015] The result output module is used to display and save the measured and predicted values of the impedance spectrum.

[0016] On the other hand, the method of the embodiment of the present invention is implemented through the following technical solutions: A method for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile, which is implemented based on the online measurement and estimation system. The online measurement and estimation method includes the following steps:

[0017] The signal generation sub-module modifies the charging strategy of the battery under test at the charging pile end, inserts a pulse control signal required for impedance measurement during the charging process of the battery under test, and transmits the pulse control signal to the waveform amplification sub-module;

[0018] After receiving the pulse control signal input by the signal generation sub-module, the waveform amplification sub-module uses the power conversion circuit of the charging pile to amplify the pulse control signal to obtain a measurement excitation signal, and applies it across the battery under test;

[0019] The response processing sub-module is connected to the battery under test, samples the voltage response and current response of the battery under test under the measurement excitation signal, compensates for the time delay and voltage drop of the current and voltage responses, and calculates the impedance response data of the battery under test under the measurement excitation signal in parallel;

[0020] The neural network sub-module retrieves the neural network model trained with offline impedance data and outputs the resistance and capacitance parameters of the Randles circuit model;

[0021] The impedance spectrum estimation sub-module online estimates the electrochemical impedance spectrum of the battery under test according to the resistance and capacitance parameters of the Randles circuit model and the measurement excitation signal.

[0022] Compared with the prior art, the beneficial effects achieved by the present invention include:

[0023] 1. The present invention can, while the charging pile charges the in-vehicle lithium battery, monitor the health state of the lithium battery and provide a basis for formulating the charging strategy of the lithium battery by online measuring and predicting the current value and future value of the electrochemical impedance spectrum of the lithium battery.

[0024] 2. The present invention uses the existing control circuit of the charging pile to generate a programmable composite wave pulse signal required for impedance measurement. This signal is easier to generate than a conventional periodic sine signal and does not require a traditional complex and expensive waveform generator; the programmable composite wave pulse signal makes it possible to measure the impedance values of the battery pack under test at multiple frequencies at one time, avoiding generating the impedance value at a single measurement frequency, and improving the efficiency of impedance measurement.

[0025] 3. The present invention uses the existing power conversion circuit of the charging pile to amplify the power of the composite wave pulse signal, avoiding the use of an expensive dedicated power amplifier circuit in traditional measuring instruments.

[0026] 4. The present invention can perform fast and accurate online prediction of the electrochemical impedance spectrum based on the offline impedance data provided by the battery management system and the measurement excitation signal input to the battery under test, and can efficiently obtain the health status information of the battery under test. At the same time, the physics-informed neural network and the Randles circuit provide physical interpretability for the online prediction of the electrochemical impedance spectrum, avoiding the problems of inaccurate models in model-based prediction and lack of physical interpretability in data-based prediction, and improving the accuracy and efficiency of online prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present invention will be further described below with reference to the drawings and embodiments:

[0028] Figure 1 FIG. is a schematic structural diagram of a system for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile proposed in an embodiment of the present invention.

[0029] Figure 2 FIG. is a flowchart of a method for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile proposed in an embodiment of the present invention for predicting the electrochemical impedance spectrum of a battery under test.

[0030] Figure 3 FIG. is a waveform diagram of a programmable composite wave pulse and a measurement excitation signal and their relaxation times in an embodiment of the present invention.

[0031] Figure 4 FIG. is a comparison diagram of an online predicted impedance spectrum and a measured impedance spectrum in an embodiment of the present invention.

[0032] Figure 5 FIG. is a schematic structural diagram of a physics-informed neural network in an embodiment of the present invention.

[0033] Figure 6 FIG. is a schematic structural diagram of a Randles circuit model in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] A specific embodiment of the present invention will be described in detail below to make the technical solution of the present invention clearer. The embodiments described below by referring to the drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. The execution order of each step in the embodiment can be adjusted adaptively according to the understanding of those skilled in the art.

[0035] In the description of the present invention, unless otherwise clearly defined, the technical and scientific terms used are the same as those generally understood by those skilled in the technical field to which the present invention belongs. The technical and scientific terms used in the present invention are only for describing the details of specific embodiments and should not be construed as limiting the present invention.

[0036] The present invention takes into account the problems existing in the prior art, and based on the physics-informed neural network, online measurement and prediction of the electrochemical impedance spectrum of the lithium battery of the charging pile are carried out. It not only uses a certain physical model to bring interpretability to solve problems, but also simplifies the analysis complexity by utilizing the advantages of data.

[0037] Next, a method and system for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile according to this embodiment will be described with reference to the accompanying drawings.

[0038] As Figure 1 shown, a system for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile proposed in an embodiment of the present invention includes an impedance spectrum measurement module, an impedance spectrum prediction module, and a result output module.

[0039] Among them, the impedance spectrum measurement module includes the following sub-modules:

[0040] A signal generation sub-module, connected to the battery management system, modifies the charging strategy of the battery under test at the charging pile end, and inserts a small-amplitude programmable composite wave pulse required for measuring impedance as a pulse control signal during the charging process of the battery under test; for example, in the pulse charging condition, the programmable composite wave pulse will be applied within the relaxation time of pulse charging to avoid directly interfering with pulse charging; and in the constant current charging condition, the programmable composite wave pulse is a small-amplitude short current pulse; this pulse control signal contains multiple components with different amplitudes and different frequencies; and the pulse control signal is transmitted to the waveform amplification sub-module.

[0041] A waveform amplification sub-module, affected by the programmable composite wave pulse, during the charging process of the battery under test by the charging pile, uses the power conversion circuit of the charging pile to amplify the programmable composite wave pulse, amplifies the small-amplitude programmable composite wave pulse into a broadband large-amplitude measurement excitation signal, and applies the measurement excitation signal across the battery under test, that is, inputs it to the battery under test.

[0042] A response processing sub-module is used to sample the voltage response and current response of the battery under test under the measurement excitation signal, perform data processing and calculate the impedance response of the battery under test.

[0043] The impedance spectrum prediction module includes the following sub-modules:

[0044] A neural network sub-module is used to, according to the impedance response of the battery under test, retrieve the Randles circuit model parameters output by the neural network model trained with offline impedance data, and transmit them to the impedance spectrum estimation sub-module;

[0045] An impedance spectrum estimation sub-module, which is used to combine the Randles circuit model parameters output by the neural network sub-module and the measured excitation signal to perform online prediction on the electrochemical impedance spectrum of the battery under test, and obtain the future electrochemical impedance spectrum curve.

[0046] The result output module is used to display and save the measured and predicted values of the impedance spectrum.

[0047] Among them, the signal generation sub-module inserts a programmable composite wave pulse during the charging process according to the charging strategy characteristics of the battery under test. The programmable composite wave pulse is easy to create and contains rich harmonic components, which can be directly generated by the charging pile and amplified by the power conversion circuit into an aperiodic measurement excitation signal, without the need for a traditional complex and expensive waveform generator to input a sine excitation signal.

[0048] The measurement excitation signal input to the waveform amplification sub-module for the battery under test can be expressed in the following form:

[0049]

[0050] Among them, A is the amplitude of the programmable composite wave pulse, which is amplified by the power conversion circuit; f is the frequency of the programmable composite wave pulse, and n is the number of harmonics. In terms of timing, a relaxation time will be set before and after each programmable composite wave pulse (measurement excitation signal) to eliminate the influence of the programmable composite wave pulse (measurement excitation signal) on the state of charge of the battery under test.

[0051] The response processing sub-module includes a Chebyshev filter and an analog-to-digital converter, which respectively filter and perform real-time sampling on the measurement excitation signal and the voltage and current response signals of the battery under test, compensate for the time delay and voltage drop of the current and voltage responses through multiple measurements, and then perform data processing, including but not limited to performing Fourier decomposition to obtain the amplitude components of the measurement excitation signal and the response signal at the same frequency, and further calculating the impedance response of the battery under test. The cut-off frequency of the Chebyshev filter is set to 15 kHz to ensure complete coverage of the frequency range (millihertz to kilohertz) of the measurement excitation signal, fully extract the effective information in the high-frequency range, and have a strong attenuation characteristic at the cut-off frequency to avoid introducing high-frequency noise. The analog-to-digital converter samples and discretizes the programmable composite wave pulse and the voltage and current response signals of the battery under test in continuous time. The sampling frequency of the analog-to-digital converter follows the Shannon sampling theorem, that is, to avoid signal distortion, the sampling frequency should be greater than 2 times the frequency of the signal to be sampled. At the same time, to achieve fast and efficient real-time measurement, the computational burden brought by high-frequency sampling should be avoided. In this embodiment, the sampling frequency of the analog-to-digital converter is set to 5 times the maximum value of the frequency f of the measurement excitation signal.

[0052] The analog-to-digital converter is installed at the charging pile end and is independent of the battery management system (BMS), avoiding the filtering of the signal to be sampled due to the low-pass filtering characteristics of the Delta-Sigma analog-to-digital converter. In addition, in order to avoid the storage and processing burden brought by high resolution while improving the quantization accuracy, the resolution of the analog-to-digital converter is more than 16 bits.

[0053] Further, the response processing sub-module converts the sampled voltage and current response signals from the time domain to the frequency domain through an analog-to-digital converter, analyzes the frequency response using the discrete Fourier transform, calculates the impedance response corresponding to the battery under test at multiple measurement frequencies, transfers the impedance response data of the battery under test to the impedance spectrum estimation module, further calculates the impedance value of the battery under test through the impedance spectrum estimation module, and finally transfers the impedance value of the battery under test to the result output module for display and storage. In this embodiment, the calculation formula for the impedance value of the battery under test is:

[0054]

[0055] In the formula, Z is the impedance value of the battery under test in the frequency domain, U is the voltage response of the battery under test in the frequency domain, and I is the current response of the battery under test in the frequency domain. Among them, the response processing sub-module calculates the real and imaginary part values of the impedance response, obtains the electrochemical impedance spectrum with the Nyquist diagram as the standard, and transfers it to the impedance spectrum estimation module.

[0056] The impedance spectrum estimation module includes the following sub-modules:

[0057] The neural network sub-module, whose framework is a Physics-Informed Neural Networks (PINN), fits and trains the offline impedance data provided by the battery management system, and outputs the resistance and capacitance parameters of the Randles circuit model.

[0058] The impedance spectrum estimation sub-module is used to online estimate the electrochemical impedance spectrum of the battery under test by combining the resistance and capacitance parameters of the Randles circuit model output by the neural network sub-module and the measurement excitation signal.

[0059] Among them, this embodiment can obtain charging data from the battery management system, the battery data cloud platform, the charging pile, etc.; the offline impedance data includes the characteristic information of the ohmic impedance, SEI impedance, SEI capacitance, charge transfer impedance, double-layer capacitance, and diffusion impedance extracted from the impedance spectrum curve, as well as the frequency and amplitude of the measurement excitation signal.

[0060] Reference Figure 2 , a method for online measurement and estimation of the electrochemical impedance spectrum of a lithium battery based on a charging pile proposed in the embodiment of the present invention includes the following steps:

[0061] After receiving the information transmitted by the battery management system, the signal generation sub-module will determine the battery under test; generate a programmable composite wave pulse according to the charging strategy characteristics of the battery under test, and transmit the programmable composite wave pulse to the waveform amplification sub-module.

[0062] After receiving the programmable composite wave pulse input by the signal generation sub-module, the waveform amplification sub-module uses the power conversion circuit of the charging pile to amplify the programmable composite wave pulse with a small amplitude into a large amplitude measurement excitation signal including a wide frequency band with multiple different amplitudes A and different frequencies f, and applies it across the battery under test.

[0063] Figure 3 This is the waveform diagram of the programmable composite wave pulse, the measurement excitation signal and their relaxation times in the embodiment of the present invention. Refer to Figure 3 , in terms of timing, after each programmable composite wave pulse (measurement excitation signal), a certain relaxation time will be set. The relaxation time is intended to eliminate the influence of the programmable composite wave pulse (measurement excitation signal) on the state of charge (SOC) of the battery under test, so as to obtain an accurate impedance response at this state of charge.

[0064] The response processing sub-module is connected to the battery under test and calculates the impedance response data of the battery under test under the measurement excitation signal. The specific steps are as follows:

[0065] Step 1.1: When the battery under test is in a stable DC polarization condition (equilibrium condition), input the measurement excitation signal to the battery under test;

[0066] Step 1.2: Sample the measurement excitation signal and the current and voltage responses of the battery under test;

[0067] Step 1.3: Compensate for the time delay and voltage drop of the current and voltage responses to obtain a compensation value;

[0068] Step 1.4: Use the discrete Fourier transform to calculate the impedance response data of the battery under test, including amplitude and frequency, calculate the real part and imaginary part data corresponding to the Nyquist diagram, and draw the impedance spectrum curve.

[0069] The time delay and voltage drop of the current and voltage responses are due to the parasitic capacitance and contact resistance of the battery under test and the charging pile cable. The compensation value is obtained by offline comparing the actual current and voltage responses measured by the response processing sub-module and the electrochemical workstation under the action of the measurement excitation signal; and the time delay and voltage drop are compensated when measuring the current and voltage responses of the battery under test.

[0070] The neural network sub-module retrieves the measured charge and discharge historical data provided by the battery management system and extracts the impedance parameter vector from the impedance spectrum curve; the amplitude and frequency of the measured excitation signal δ, and the impedance parameter vector including ohmic impedance, SEI impedance, SEI capacitance, charge transfer impedance, double-layer capacitance, and diffusion impedance extracted from the impedance spectrum curve are used as offline impedance data.

[0071] The frequency band range of the offline impedance data includes kilohertz to millihertz; the sampling frequency is selected to be the same as that of the battery under test: following the Shannon sampling theorem and avoiding the complexity brought by high-frequency calculations, the sampling frequency is selected to be 5 times the frequency of the measured excitation signal.

[0072] The extraction of impedance spectrum characteristic information is specifically as follows: the characteristic information r of the ohmic impedance is extracted in the frequency band above kilohertz ohm , the characteristic information r of the SEI impedance and the characteristic information cS of the SEI capacitance are extracted in the frequency band from hundred hertz to kilohertz SEI . EI , the characteristic information r of the charge transfer impedance and the characteristic information c of the double-layer capacitance are extracted in the frequency band from hertz to hundred hertz ct . dl , the characteristic information z of the diffusion impedance is extracted in the frequency band from millihertz to hertz w .

[0073] Furthermore, the offline impedance data is input into the physical information neural network of the neural network sub-module to train the physical information neural network and output the parameters of the Randles circuit model. In this embodiment, different characteristic information is extracted from low frequency to high frequency, which is sufficient to reflect the multi-level information of the internal lithium battery from slow material diffusion to fast charge transfer; in the impedance spectrum curve, characteristic parameters such as ohmic impedance, double-layer capacitance, and diffusion impedance are coupled with each other and show dynamic changes in the low-frequency to high-frequency range; therefore, the above characteristic parameters extracted from the impedance spectrum curve can comprehensively reflect the health state of the battery under test and provide sufficient characteristics for the training of the physical information neural network.

[0074] In this embodiment, the training and parameter output process of the physical information neural network is specifically as follows:

[0075] Step 3.1, construct a 7-dimensional offline impedance data vector x(δ, r ohm , r SEI , c SEI , r ct , c dl , z w ) according to the offline impedance data and input it for preprocessing. The preprocessing includes using a sliding time window to replace the missing and incorrect offline impedance data to ensure that the input data can provide effective historical information for model training.

[0076] Step 3.2: Construct a physics-informed neural network and train the constructed physics-informed neural network with the offline impedance data vector.

[0077] Among them, the constructed physics-informed neural network includes an input layer, a hidden layer, and an output layer. The input layer will receive the offline impedance data, including the characteristic information of the measurement excitation signal, ohmic impedance, SEI impedance, SEI capacitance, charge transfer impedance, double-layer capacitance, and diffusion impedance, and input it as a 7-dimensional vector. The hidden layer contains 3 fully connected hidden layers, and each hidden layer is set with 30 neurons. The hidden layer will, under the action of the activation function σ, iteratively update the weight matrix w and the bias vector b to obtain the output, and the weight matrix w and the bias vector b will be obtained during the training process. In the present invention, the activation function σ is the tanh function. The output layer will output a 4-dimensional vector u(R s ,R ct ,C dl ,Z w ), which respectively represent the parameters of the Randles circuit model, including the ohmic impedance R s , the charge transfer impedance R ct , the double-layer capacitance C dl , and the diffusion impedance Z w .

[0078] Step 3.3: Iteratively update the weight matrix w and the bias vector b of the physics-informed neural network to minimize the loss function L of the physics-informed neural network. The calculation formula of the loss function L is as follows:

[0079] L = λ 1 L EIS + λ 2 L p

[0080]

[0081] In the formula, L represents the total loss function of the physics-informed neural network, λ 1 represents the weight of the loss value L EIS of the impedance spectrum, and λ 2 represents the weight of the loss value L p of the output Randles circuit model parameters; in this embodiment, λ 1 = 0.9 and λ 2 = 0.1 are set. n represents the total number of samples, m is the total number of data points of the impedance spectrum, and q is the number of Randles circuit parameters. In this embodiment, q = 4. is the impedance response prediction value of the k-th sample at the j-th impedance spectrum data point, is the measured impedance response value of the k-th sample at the j-th impedance spectrum data point, is the estimated value of the a-th Randles circuit model parameter of the k-th sample, is the true value of the a-th Randles circuit model parameter of the k-th sample.

[0082] The total loss function L characterizes the matching degree between the parameters of the Randles circuit model estimated by the physics-informed neural network and the impedance spectrum. The goal of the physics-informed neural network is to find a set of optimal network parameters (w, b) to minimize the total loss function L.

[0083] Step 3.4: Use the physics-informed neural network to output the resistance and capacitance parameters of the Randles circuit model, and online estimate the electrochemical impedance spectrum of the battery under test according to the resistance and capacitance parameters.

[0084] The output Randles circuit model parameters will be used by the impedance spectrum estimation sub-module to estimate the impedance response Z under the measurement excitation signal with an input frequency of f. The calculation formula is as follows:

[0085]

[0086] Plot the calculated impedance response Z(f) as an electrochemical impedance spectrum with the Nyquist plot as the standard, and the online estimated value of the electrochemical impedance spectrum of the battery under test can be obtained.

[0087] The result output module displays and saves the measured and estimated values of the impedance spectrum.

[0088] The present invention uses the charging pile to generate a measurement excitation signal with rich harmonic components, and measures the impedance values of the battery under test at multiple frequencies at one time, improving the rapidity of impedance measurement; the present invention uses the physics-informed neural network to online estimate the impedance spectrum and predict the change of the electrochemical impedance spectrum in the future period, which is of great significance for the health state monitoring and charging strategy formulation of lithium batteries.

[0089] The above embodiments are one of the implementation manners of the present invention. The implementation manners of the present invention are not limited by the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement manners and are all included in the protection scope of the present invention.

Claims

1. A lithium battery electrochemical impedance spectrum online measurement and estimation system based on a charging pile, characterized in that: It includes an impedance spectrum measurement module, an impedance spectrum estimation module and a result output module; Wherein, the impedance spectrum measurement module includes the following submodules: The signal generation submodule modifies the charging strategy of the battery under test at the charging pile end, inserts a small-amplitude pulse control signal required for measuring impedance during the charging process of the battery under test, and transmits it to the waveform amplification submodule; The waveform amplification submodule amplifies the small-amplitude pulse control signal into a wide-band large-amplitude measurement excitation signal through the power conversion circuit of the charging pile, and applies it to both ends of the battery to be tested; A response processing submodule, used for sampling the voltage response and current response of the battery under test under the measurement excitation signal, performing data processing and calculating the impedance response of the battery under test; The impedance spectrum estimation module includes the following submodules: The neural network submodule is used to retrieve the measured charge and discharge history data, extract the impedance parameter vector, train the physical information neural network model offline, output the Randles circuit model parameters, and pass them to the impedance spectrum estimation submodule; The impedance spectrum estimation submodule combines the Randles circuit model parameters and the current measurement excitation response signal to perform online estimation of the electrochemical impedance spectrum of the battery to be tested and predict the future electrochemical impedance spectrum curve; The result output module is used to display and save the measured and estimated values ​​of the impedance spectrum.

2. The online measurement and estimation system according to claim 1, characterized in that: The pulse control signal inserted by the signal generation submodule during the charging process is a programmable composite wave pulse, including a plurality of signal components with different amplitudes A and different frequencies f, expressed as: Wherein, A is the amplitude of the programmable composite wave pulse, f is the frequency of the programmable composite wave pulse, and n is the number of harmonics; In terms of timing, relaxation time is set before and after each programmable complex wave pulse.

3. The online measurement and estimation system according to claim 1, characterized in that: The programmable composite wave pulse output by the signal generation submodule will be directly amplified by the power conversion circuit of the charging pile in the waveform amplification submodule.

4. The online measurement and estimation system according to claim 1, characterized in that: The response processing submodule samples the voltage response and current response signals through an analog-to-digital converter; compensates for the time delay and voltage drop of the current and voltage responses; uses discrete Fourier transform to convert the time domain to the frequency domain, and calculates the impedance response corresponding to the battery under test at multiple measurement frequencies in parallel, obtains the real and imaginary data of the corresponding Nyquist diagram, and draws the impedance spectrum curve.

5. The online measurement and estimation system according to claim 1, characterized in that: The framework of the neural network submodule in the impedance spectrum estimation module is a physical information neural network, which performs fitting training on offline impedance data and outputs the resistance and capacitance parameters of the Randles circuit model; The impedance spectrum estimation submodule estimates the electrochemical impedance spectrum of the battery under test online based on the resistance and capacitance parameters of the Randles circuit model and the measured excitation signal.

6. The online measurement and estimation system according to claim 1 or 5, characterized in that: The training and parameter output process of the physical information neural network includes the following steps: Retrieving the measured charge and discharge history data, extracting the impedance parameter vector, wherein the impedance parameter vector includes characteristic information of ohmic impedance, SEI impedance, SEI capacitance, charge transfer impedance, double layer capacitance, and diffusion impedance extracted from the impedance spectrum curve, and using the extracted impedance parameter vector and the frequency and amplitude of the measured excitation signal as offline impedance data; Preprocess the offline impedance data; Constructing a physical information neural network, and inputting the offline impedance data into the physical information neural network for training; Iteratively updating the weight matrix and bias vector of the physical information neural network to minimize the loss function of the physical information neural network; The physical information neural network is used to output the resistance and capacitance parameters of the Randles circuit model.

7. The online measurement and estimation system according to claim 6, characterized in that: The frequency range of offline impedance data includes kilohertz to millihertz; the process of extracting impedance parameter vector includes: extracting characteristic information of ohmic impedance in the frequency range above kilohertz ohm , extract the characteristic information of SEI impedance in the frequency range from 100 Hz to 1 kHz SEI and the characteristic information of SEI capacitance c SEI , extract the characteristic information of charge transfer impedance in the frequency range from Hz to 100 Hz ct The characteristic information of double layer capacitance cdl and the characteristic information of diffusion impedance z are extracted in the frequency band from millihertz to hertz w ; Based on the offline impedance data, a 7-dimensional offline impedance data vector x(δ,r ohm ,r SEI ,c SEI ,r ct ,c dl ,z w ) is input and preprocessed before being input into the physical information neural network, where δ is the measurement excitation signal.

8. The online measurement and estimation method according to claim 1, characterized in that: Based on the online measurement and estimation system according to any one of claims 1 to 7, the online measurement and estimation method comprises the following steps: The signal generation submodule modifies the charging strategy of the battery under test at the charging pile end, inserts the pulse control signal required for measuring impedance during the charging process of the battery under test, and transmits the pulse control signal to the waveform amplification submodule; After receiving the pulse control signal input by the signal generation submodule, the waveform amplification submodule uses the power conversion circuit of the charging pile to amplify the pulse control signal to obtain a measurement excitation signal, and applies it to both ends of the battery to be tested; The response processing submodule is connected to the battery to be tested, samples the voltage response and current response of the battery to be tested under the measurement excitation signal, compensates for the time delay and voltage drop of the current and voltage responses, completes the time-frequency domain conversion through Fourier transform, and then calculates the impedance response data of the battery to be tested under the measurement excitation signal in parallel; The neural network submodule calls the neural network model trained using offline impedance data and outputs the resistance and capacitance parameters of the Randles circuit model; The impedance spectrum estimation submodule estimates the electrochemical impedance spectrum of the battery under test online based on the resistance and capacitance parameters of the Randles circuit model and the measured excitation signal.

Images