Parameter identification method and identification device of battery equivalent circuit and battery unit

Abstract

The invention discloses a parameter identification method and identification device of a battery equivalent circuit and a battery unit, belongs to the technical field of battery models, and is suitable for a second-order RC equivalent circuit of a battery, the parameter identification method comprises the following steps: at a target temperature and a target SOC, carrying out an electrochemical impedance spectroscopy (EIS) test on the battery to obtain an electrochemical impedance spectroscopy under high-frequency excitation, identifying the accurate value of the internal resistance R0 of the battery; under the target temperature and the target SOC, a hybrid pulse power characteristic HPPC test is carried out on the battery, voltage response data and current response data under short-time pulse current excitation and long-time pulse current excitation are obtained respectively, and accurate parameter values of a polarization RC branch and a diffusion RC branch are identified. The weighted identification method comprises the following steps of: acquiring a plurality of groups of accurate parameter values by adopting the parameter identification method, and carrying out weighted summation to obtain a final accurate parameter value. The method has the advantages of being high in precision, simple in calculation and high in working condition adaptability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery model, and particularly relates to a parameter identification method and device of a battery equivalent circuit and a battery unit. BACKGROUND

[0002] An accurate battery model is the basis for battery state (such as SOC, SOH) estimation and energy management. A battery second-order RC equivalent circuit model as shown in FIG. 1 is widely used due to its good balance between precision and complexity. Model parameters (Ohmic resistance R0, polarization resistance and polarization capacitance R1 / C1, diffusion resistance and diffusion capacitance R2 / C2) will change with factors such as temperature, SOC, and aging degree, and therefore need to be accurately identified. Figure 1

[0003] The existing parameter identification technology mainly has the following deficiencies:

[0004] 1. Only applicable to specific working conditions: HPPC test is mainly for pulse current working conditions of seconds to tens of seconds, and cannot accurately reflect the low-frequency response characteristics of the battery in the long-time charging and discharging process, resulting in inaccurate identification of R2 / C2 parameters.

[0005] 2. Parameter aliasing problem: In the HPPC test, the R0 response (millisecond level) is coupled with the R1 and R2 responses (second to hundred-second level) in the time scale, resulting in that the identified value of R0 actually contains part of the un-damped medium-high frequency polarization resistance, instead of pure Ohmic resistance, introducing systematic errors. This error will be passed on and affect the identification accuracy of R1, C1, R2, and C2.

[0006] 3. Limitations of online identification algorithms: Although online identification algorithms based on recursive least squares (RLS), extended Kalman filter (EKF), etc. can update parameters in real time, the algorithms are complex and have high computational overhead, requiring high algorithmic power of microcontrollers (MCU), increasing the hardware cost and software complexity of the BMS, and having convergence and stability problems.

[0007] 4. Poor working condition adaptability: The parameters identified by the traditional offline method in a specific short-time pulse working condition cannot accurately describe the low-frequency characteristics of the battery in the long-time charging and discharging (such as constant-current charging, actual driving working condition) process, resulting in a decrease in the prediction accuracy of the model in complex working conditions.

[0008] Therefore, there is an urgent need for a parameter identification method with high precision, simple calculation, and strong working condition adaptability. SUMMARY

[0009] ​The purpose of the present application is to overcome the deficiencies in the prior art, provide a battery equivalent circuit parameter identification method, identification device and battery unit, solve the technical problems of low precision, high complexity and weak working condition adaptability of the existing parameter identification method.

[0010] To achieve the above purpose, the present application is realized by adopting the following technical solutions:

[0011] In the first aspect, the present application provides a battery equivalent circuit parameter identification method, which is suitable for a battery second-order RC equivalent circuit, the battery second-order RC equivalent circuit includes a battery internal resistance R0, a polarization RC branch and a diffusion RC branch in series; the parameter identification method includes:

[0012] Under the target temperature and the target SOC, an electrochemical impedance spectroscopy (EIS) test is performed on the battery to obtain an electrochemical impedance spectroscopy under high-frequency excitation;

[0013] Based on the electrochemical impedance spectroscopy, the accurate value of the battery internal resistance R0 under the target temperature and the target SOC is identified;

[0014] Under the target temperature and the target SOC, a hybrid pulse power characteristic (HPPC) test is performed on the battery to obtain voltage and current response data under short-time pulse current excitation and long-time pulse current excitation, respectively;

[0015] Based on the voltage and current response data, the accurate parameter values of the polarization RC branch and the diffusion RC branch under the target temperature and the target SOC are identified.

[0016] The parameter identification method provided by the present application uses the characteristics that the polarization element has not responded under high-frequency excitation, ensures the purity of the battery internal resistance R0, avoids the mixing of the middle-low frequency polarization internal resistance, and obtains the accurate value of the battery internal resistance R0. Based on the voltage change curve in the pulse duration stage, the function fitting method is adopted to obtain the accurate parameter values of the polarization RC branch and the diffusion RC branch. The overall method has the characteristics of high precision, simple calculation and strong working condition adaptability.

[0017] Optionally, the battery is pretreated before the electrochemical impedance spectroscopy (EIS) test and the hybrid pulse power characteristic (HPPC) test, the pretreatment includes placing the battery in a constant temperature environment at the target temperature, charging / discharging at a constant current to the target SOC, and standing for a predetermined time.

[0018] Through the pretreatment, the battery is in a stable state before the test, the accuracy of the test result is improved, and the accuracy of the subsequent identification calculation is ensured.

[0019] Optionally, the obtaining of the electrochemical impedance spectroscopy under high-frequency excitation includes:

[0020] An AC excitation signal is applied in the high-frequency region to ensure that the battery is in a linear response state, and the corresponding electrochemical impedance spectrum is obtained.

[0021] The high-frequency region has a frequency range of 1 kHz to 5 kHz, and the amplitude of the AC excitation signal is ≤10mV.

[0022] In the high-frequency region, the polarization effect of the battery can be ignored. At this time, the impedance is mainly composed of the battery internal resistance R0, which ensures the purity of the R0 value and avoids the mixing of mid- and low-frequency polarization internal resistance, which would affect the identification accuracy of the battery internal resistance R0.

[0023] Optionally, the precise value of the battery internal resistance R0 is the impedance value at the intersection of the high-frequency region and the real axis in the Nyquist plot of the electrochemical impedance spectrum.

[0024] In the high-frequency region, the polarization effect of the battery can be ignored. At this time, the impedance is mainly composed of the battery's internal resistance R0. Therefore, the impedance value at the intersection of the high-frequency region and the real axis in the Nyquist plot can be extracted as the battery's internal resistance R0.

[0025] Optionally, the duration of the short-time pulse current excitation is 10s to 60s, and the duration of the long-time pulse current excitation is 5min to 1h.

[0026] The purpose of short-time pulsed current excitation is to capture the voltage jump at the moment of pulse loading, and to roughly identify the parameter values ​​of the polarized RC branch and the diffused RC branch. Therefore, applying a pulsed current on the order of several seconds to tens of seconds is sufficient. The purpose of long-time pulsed current excitation is to apply a long-term charge-discharge condition. Under this condition, the mid-frequency polarization process has basically reached a steady state, while the low-frequency ion diffusion process is fully manifested, thus allowing for precise identification of the parameter values ​​of the diffused RC branch. Therefore, applying a pulsed current on the order of several minutes to one hour is necessary. By simulating the actual operating conditions of the battery, the test results are made closer to actual operating conditions, improving the accuracy of subsequent parameter identification.

[0027] Optionally, identifying the precise parameter values ​​of the polarized RC branch and the diffused RC branch at the target temperature and target SOC based on the voltage and current response data includes:

[0028] Construct corresponding response curves based on voltage and current response data under short-time pulse current excitation and long-time pulse current excitation;

[0029] A second-order exponential fit is performed on the zero-input response segment of the response curve under short-time pulse current excitation to calculate the rough parameter values ​​of the polarized RC branch and the diffused RC branch.

[0030] A second-order exponential fit is performed on the zero-input response segment of the response curve under long-term pulse current excitation. The coarse parameter values ​​of the polarized RC branch are added as known quantities to the fitting process to calculate the precise parameter values ​​of the diffused RC branch.

[0031] A second-order exponential fit is performed on the zero-input response segment of the response curve under short-time pulse current excitation. The precise parameter values ​​of the diffused RC branch are added as known quantities to the fitting process to calculate the precise parameter values ​​of the polarized RC branch.

[0032] Under short-time pulsed current excitation, the coarse parameter values ​​of the polarization RC branch and the diffusion RC branch are identified by the voltage jump at the moment of pulse loading. Then, using the coarse parameter value of the polarization RC branch as a known quantity, the precise parameter value of the diffusion RC branch is identified separately under long-time pulsed current excitation. Under this condition, the mid-frequency polarization process has basically reached a steady state, while the low-frequency ion diffusion process is fully manifested. Therefore, the precise parameter value of the diffusion RC branch is highly accurate. Then, using the precise parameter value of the diffusion RC branch as a known quantity, the precise parameter value of the polarization RC branch is re-identified under short-time pulsed current excitation, thus improving its identification accuracy.

[0033] Optionally, the fitting formula for the second-order exponential fit is:

[0034]

[0035] In the formula, For the coordinate points on the response curve, These are the start and end times of the zero-input response segment, respectively. These are the fitting coefficients. For constant terms;

[0036] The formula for the zero-input response segment is:

[0037]

[0038] In the formula, The voltage value of the zero-input response segment. Let R1 be the resistance value and C1 be the capacitance value in the polarized RC branch. Let R2 be the resistance value and C2 be the capacitance value in the diffused RC branch. The time within the zero input response segment, This is the current value of the pulse current. This is the battery open-circuit voltage;

[0039] Combine the fitting equation and the operational equation for the zero-input response segment, and then apply the fitting coefficients... Calculate the parameter values ​​for the polarized RC branch and the diffused RC branch:

[0040]

[0041] .

[0042] The main advantage of second-order exponential fitting lies in its ability to precisely characterize dynamic processes at different time scales, balancing the model's accuracy and practicality. Compared to first-order models, it more accurately describes the voltage relaxation process; compared to higher-order models or intelligent algorithms, its mathematical form is relatively simple, its computational burden is lighter, and it is beneficial for implementation in embedded systems (such as BMS).

[0043] Secondly, the present invention provides a parameter weighting identification method for a battery equivalent circuit, comprising:

[0044] Using the target temperature and target SOC as a baseline, adjust the target temperature and target SOC to obtain the neighboring temperature and neighboring SOC;

[0045] Construct a neighbor group of the benchmark group based on the neighboring temperature and the target SOC, and the target temperature and the neighboring SOC;

[0046] Using the parameter identification method described above, the accurate parameter values ​​of the battery internal resistance R0, polarized RC branch, and diffused RC branch in the reference group and adjacent group are obtained.

[0047] The precise parameter values ​​of the baseline group and the neighboring group are weighted and summed to obtain the final precise parameter values ​​of the battery internal resistance R0, the polarized RC branch, and the diffused RC branch at the target temperature and the target SOC.

[0048] The parameter weighted identification method provided by this invention can integrate the parameter values ​​of the benchmark group and the neighboring group, reducing the random error of single-point identification. At the same time, it can construct the functional relationship between parameters and SOC and temperature more smoothly and accurately, rather than just the mean of discrete points, thus improving the engineering application value of the final parameters.

[0049] Thirdly, the present invention provides a parameter identification device for a battery equivalent circuit, adapted to a second-order RC equivalent circuit of a battery, wherein the second-order RC equivalent circuit of the battery includes a battery internal resistance R0 connected in series, a polarized RC branch, and a diffused RC branch; the parameter identification device includes:

[0050] The EIS test module is configured to perform electrochemical impedance spectroscopy (EIS) tests on the battery at the target temperature and target SOC to obtain the electrochemical impedance spectrum under high-frequency excitation.

[0051] An internal resistance determination module is configured to identify the precise value of the battery internal resistance R0 at the target temperature and target SOC based on the electrochemical impedance spectroscopy.

[0052] The HPPC test module is configured to perform hybrid pulse power characteristic HPPC tests on the battery at the target temperature and target SOC, and to acquire voltage and current response data under short-time pulse current excitation and long-time pulse current excitation, respectively.

[0053] The branch determination module is configured to identify the precise parameter values ​​of the polarized RC branch and the diffused RC branch at the target temperature and target SOC based on the voltage and current response data.

[0054] Optionally, the precise value of the battery internal resistance R0 is the impedance value at the intersection of the high-frequency region and the real axis in the Nyquist plot of the electrochemical impedance spectrum.

[0055] Optionally, the branch determination module is specifically configured as follows:

[0056] Construct corresponding response curves based on voltage and current response data under short-time pulse current excitation and long-time pulse current excitation;

[0057] A second-order exponential fit is performed on the zero-input response segment of the response curve under short-time pulse current excitation to calculate the rough parameter values ​​of the polarized RC branch and the diffused RC branch.

[0058] A second-order exponential fit is performed on the zero-input response segment of the response curve under long-term pulse current excitation. The coarse parameter values ​​of the polarized RC branch are added as known quantities to the fitting process to calculate the precise parameter values ​​of the diffused RC branch.

[0059] A second-order exponential fit is performed on the zero-input response segment of the response curve under short-time pulse current excitation. The precise parameter values ​​of the diffused RC branch are added as known quantities to the fitting process to calculate the precise parameter values ​​of the polarized RC branch.

[0060] Fourthly, the present invention provides a battery cell, including one or more of a battery module, a battery pack, or an energy storage battery, wherein the battery cell uses the parameter identification device, the parameter identification method, or the parameter weighting identification method described above to identify the parameters of the battery equivalent circuit.

[0061] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0062] This invention provides a parameter identification method, device, and battery cell for a battery equivalent circuit. The parameter identification method and device utilize the characteristic that the polarization stage has not yet responded under high-frequency excitation, ensuring the purity of the battery's internal resistance R0 and avoiding the mixing of mid-to-low-frequency polarization internal resistances, thus obtaining an accurate value for the battery's internal resistance R0. Based on the voltage change curve during the pulse duration, a function fitting method is used to obtain accurate parameter values ​​for the polarization RC branch and the diffusion RC branch. The overall method features high accuracy, simple calculation, and strong adaptability to operating conditions. The weighted identification method can integrate parameter values ​​from the benchmark group and neighboring groups, reducing random errors in single-point identification. Simultaneously, it can more smoothly and accurately construct the functional relationship between parameters and SOC and temperature, rather than simply the mean of discrete points, enhancing the engineering application value of the final parameters. In summary, this invention significantly improves the accuracy of model parameters and the overall prediction accuracy of the model by employing strategies such as time-scale decoupling, step-by-step parameter iteration, moving window mean filtering, and temperature interval identification for second-order RC equivalent circuits, while maintaining the advantages of offline algorithms such as simple computation and ease of engineering deployment. Attached Figure Description

[0063] Figure 1 This is a schematic diagram of the second-order RC equivalent circuit of the battery provided by the present invention;

[0064] Figure 2 This is a flowchart illustrating the parameter identification method provided by the present invention;

[0065] Figure 3 These are the electrochemical impedance spectra of the lithium battery provided by this invention at 10% and 20% SOC.

[0066] Figure 4 This is a flowchart of parameter identification for the polarized RC branch and the diffused RC branch provided by the present invention;

[0067] Figure 5 This is a schematic diagram of the response curve for HPPC testing provided by the present invention;

[0068] Figure 6 This is a flowchart illustrating the parameter weighting identification method provided by the present invention. Detailed Implementation

[0069] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0070] Example 1

[0071] like Figure 1 As shown, this embodiment of the invention provides a parameter identification method for a battery equivalent circuit, applicable to a second-order RC equivalent circuit of a battery, such as... Figure 1As shown, the second-order RC equivalent circuit of the battery includes the battery internal resistance R0 connected in series, the polarized RC branch (resistor R1 and capacitor C1 in parallel), and the diffused RC branch (resistor R2 and capacitor C2 in parallel).

[0072] like Figure 2 As shown, the parameter identification method includes the following steps:

[0073] Step S1: At the target temperature and target SOC, perform electrochemical impedance spectroscopy (EIS) on the battery to obtain the electrochemical impedance spectrum under high-frequency excitation.

[0074] Electrochemical impedance spectroscopy (EIS) tests were performed on the battery using an electrochemical workstation. A small AC excitation signal with an amplitude ≤10mV was applied in the high-frequency region (e.g., 1kHz-5kHz) to ensure the battery was in a linear response state. The corresponding electrochemical impedance spectrum was then obtained. The frequency and amplitude can be adjusted by the operator as needed. In the high-frequency region, the battery polarization effect can be ignored, and the impedance is mainly composed of the battery's internal resistance R0.

[0075] Step S2: Identify the precise value of the battery internal resistance R0 at the target temperature and target SOC based on electrochemical impedance spectroscopy.

[0076] The impedance value at the intersection of the high-frequency region and the real axis in the Nyquist plot of the electrochemical impedance spectroscopy is extracted and used as the accurate value of the battery's internal resistance R0. This method utilizes the characteristic that the polarization circuit has not yet responded under high-frequency excitation, ensuring the purity of the R0 value and avoiding the incorporation of mid- and low-frequency polarization internal resistance.

[0077] like Figure 3 The figure shows the electrochemical impedance spectroscopy of a lithium battery at 10% and 20% SOC. The ultra-high frequency part (such as 1kHz-5kHz and above) represents the internal resistance of the battery. The points where the imaginary part is 0 in the figure are the values ​​of R0.

[0078] Step S3: At the target temperature and target SOC, perform hybrid pulse power characteristic (HPPC) tests on the battery to obtain voltage and current response data under short-time pulse current excitation and long-time pulse current excitation, respectively.

[0079] Under target temperature and target SOC, the battery is subjected to HPPC (hybrid pulse power characterization) testing, applying pulse current conditions ranging from several seconds to tens of seconds (e.g., 0.2C-2C pulse current, lasting 10-60 seconds). During the HPPC test, the battery's voltage and current response data under short-term pulse current excitation are recorded.

[0080] Under target temperature and target SOC, the battery is subjected to HPPC (hybrid pulse power characterization) testing, applying pulse current conditions ranging from several minutes to one hour (e.g., 0.1C-1C pulse current, lasting 5-60 minutes). During the HPPC test, the voltage and current response data of the battery under long-term pulse current excitation are recorded.

[0081] The purpose of short-time pulsed current excitation is to capture the voltage jump at the moment of pulse loading, and to roughly identify the parameter values ​​of the polarized RC branch and the diffused RC branch. Therefore, applying a pulsed current on the order of several seconds to tens of seconds is sufficient. The purpose of long-time pulsed current excitation is to apply a long-term charge-discharge condition. Under this condition, the mid-frequency polarization process has basically reached a steady state, while the low-frequency ion diffusion process is fully manifested, thus allowing for precise identification of the parameter values ​​of the diffused RC branch. Therefore, applying a pulsed current on the order of several minutes to one hour is necessary. By simulating the actual operating conditions of the battery, the test results are made closer to actual operating conditions, improving the accuracy of subsequent parameter identification.

[0082] Step S4: Identify the precise parameter values ​​of the polarized RC branch and the diffused RC branch at the target temperature and target SOC based on the voltage and current response data.

[0083] like Figure 4 As shown, the specific process includes:

[0084] Step S4.1: Construct the corresponding response curves based on the voltage and current response data under short-time pulse current excitation and long-time pulse current excitation.

[0085] like Figure 5 The figure shows a schematic diagram of the response curve for an HPPC test. The voltage jumps between U1 and U2 and between U3 and U4 are related to the battery's internal resistance R0.

[0086]

[0087] The gradual voltage change process from U2 to U3 and from U4 to U5 is related to R1 / C1 / R2 / C2. According to circuit principles, the current-voltage relationship of an RC circuit is divided into zero-state response and zero-input response. The U2-U3 segment can be considered as the zero-state response, and the U4-U5 segment can be considered as the zero-input response.

[0088] In this embodiment, RC parameter identification is performed using the U4-U5 segment (zero-input response) as an example, and the calculation formula for the zero-input response segment is constructed:

[0089]

[0090] In the formula, The voltage value of the zero-input response segment. Let R1 be the resistance value and C1 be the capacitance value in the polarized RC branch. Let R2 be the resistance value and C2 be the capacitance value in the diffused RC branch. The time within the zero input response segment, This is the current value of the pulse current. This is the battery open-circuit voltage.

[0091] Step S4.2: Perform second-order exponential fitting on the zero-input response segment of the response curve under short-time pulse current excitation, and calculate the rough parameter values ​​of the polarized RC branch and the diffused RC branch.

[0092] A second-order exponential fit is performed on the U4-U5 segment (zero input response).

[0093] The fitting formula is:

[0094]

[0095] In the formula, For the coordinate points on the response curve, These are the start and end times of the zero-input response segment, respectively. These are the fitting coefficients. This is a constant term.

[0096] The main advantage of second-order exponential fitting lies in its ability to precisely characterize dynamic processes at different time scales, balancing the model's accuracy and practicality. Compared to first-order models, it more accurately describes the voltage relaxation process; compared to higher-order models or intelligent algorithms, its mathematical form is relatively simple, its computational burden is lighter, and it is beneficial for implementation in embedded systems (such as BMS).

[0097] The fitting coefficients are obtained through fitting calculations. Then, by combining the fitting equation and the operational equation, the parameter values ​​of the polarized RC branch and the diffused RC branch are calculated:

[0098]

[0099] .

[0100] Step S4.3: Perform second-order exponential fitting on the zero-input response segment of the response curve under long-term pulse current excitation, and add the coarse parameter values ​​of the polarized RC branch as known quantities to the fitting process to calculate the precise parameter values ​​of the diffused RC branch.

[0101] The second-order exponential fitting here is the same as in step S4.2, except that the fitting coefficients in step S4.2 are... As known quantities, refit the fitting coefficients. .

[0102] Step S4.4: Perform second-order exponential fitting on the zero-input response segment of the response curve under short-time pulse current excitation, and add the accurate parameter values ​​of the diffused RC branch as known quantities to the fitting process to calculate the accurate parameter values ​​of the polarized RC branch.

[0103] The second-order exponential fitting here is the same as in step S4.2, except that the fitting coefficients in step S4.3 are... As known quantities, only the fitting coefficients are fitted. .

[0104] In this embodiment, under short-time pulsed current excitation, the coarse parameter values ​​of the polarization RC branch and the diffusion RC branch are identified by the voltage jump at the moment of pulse loading. Then, using the coarse parameter value of the polarization RC branch as a known quantity, the precise parameter value of the diffusion RC branch is identified separately under long-time pulsed current excitation. Under this condition, the mid-frequency polarization process has basically reached a steady state, while the low-frequency ion diffusion process is fully manifested. Therefore, the precise parameter value of the diffusion RC branch is highly accurate. Then, using the precise parameter value of the diffusion RC branch as a known quantity, the precise parameter value of the polarization RC branch is re-identified under short-time pulsed current excitation, thereby improving its identification accuracy.

[0105] Furthermore, both the electrochemical impedance spectroscopy (EIS) test and the hybrid pulse power characteristic (HPPC) test require pretreatment of the battery. Pretreatment involves placing the battery in a constant-temperature environment at the target temperature, charging / discharging it to the target state of charge (SOC) with a constant current, and then allowing it to stand for a predetermined time. This pretreatment stabilizes the battery before testing, improving the accuracy of the test results and ensuring the precision of subsequent identification calculations.

[0106] In summary, the parameter identification method provided by this invention utilizes the characteristic that the polarization circuit has not yet responded under high-frequency excitation, ensuring the purity of the battery internal resistance R0 and avoiding the mixing of mid- and low-frequency polarization internal resistances, thus obtaining an accurate value for the battery internal resistance R0. Based on the voltage change curve during the pulse duration phase, a function identification method is used to obtain the accurate parameter values ​​of the polarization RC branch and the diffusion RC branch. The overall method features high accuracy, simple calculation, and strong adaptability to operating conditions.

[0107] Taking the parameter identification of a certain 32Ah lithium iron phosphate battery at 25℃ and 50% SOC as an example, the implementation process of this invention is illustrated as follows:

[0108] 1. Battery pretreatment

[0109] Place the battery in a constant temperature chamber and set the temperature to the target temperature range (25°C). Charge and discharge the battery at a constant current (0.33C) to the target SOC point (50%), and let it stand for 2 hours to allow the battery to reach a stable state.

[0110] 2. High-frequency EIS testing and R0 identification

[0111] The battery was subjected to EIS testing using an electrochemical workstation with a frequency range of 1kHz to 5kHz and an excitation signal amplitude of 5mV. In the high-frequency region (>1kHz), electrochemical impedance spectroscopy was acquired, and the internal resistance of the battery was determined to be R0 = 0.012Ω by identifying the impedance data.

[0112] 3. Short-term HPPC operating conditions and preliminary identification of R1 / C1

[0113] The battery was subjected to HPPC testing under constant temperature conditions at the target temperature (25°C):

[0114] Discharge at 1C for 30 seconds

[0115] Let stand for 1 minute

[0116] Charge at 0.75C for 30 seconds

[0117] Let stand for 1 minute

[0118] Voltage and current data during the HPPC test were recorded at a sampling frequency of 100Hz. Based on R0=0.012Ω, a second-order RC model equation was constructed, and the exponential function was initially identified for the short-time HPPC data, yielding rough values ​​of R1=0.015Ω and C1=2400F.

[0119] 4. Accurate identification of long-term HPPC operating conditions and R2 / C2

[0120] The battery was subjected to HPPC testing under constant temperature conditions at the target temperature (25°C):

[0121] Discharge at 0.5C for 10 minutes

[0122] Let stand for 1 hour

[0123] Charge at 0.25C for 20 minutes

[0124] Let stand for 1 hour

[0125] Record voltage and current data during long-term operation. Based on the approximate values ​​of R0=0.012Ω, R1=0.015Ω, and C1=2400F, identify the long-term HPPC data to obtain the precise values ​​of R2=0.0015Ω and C2=2400F.

[0126] 5. Iterative optimization and accurate R1 / C1 identification

[0127] Substituting R0=0.012Ω, R2=0.0015Ω, and C2=2400F back into the short-time HPPC data, we constructed an updated second-order RC model equation. By re-identifying the short-time HPPC data, we obtained the precise values ​​of R1=0.014Ω and C1=2350F.

[0128] Using the parameter identification method provided in this embodiment, the impedance parameters of the battery equivalent circuit under different temperatures (0℃ / 10℃ / 25℃ / 40℃) and different SOCs (0% / 10% / 20% / 30% / 40% / 50% / 60% / 70% / 80% / 90% / 100%) are obtained, as shown in Table 1.

[0129] Table 1: Impedance parameters of the battery equivalent circuit

[0130] Table 1 provides a mapping relationship between the parameters of the equivalent circuit and temperature and SOC, thereby improving the temperature adaptability of the model.

[0131] Example 2

[0132] like Figure 6 As shown, this embodiment of the invention provides a parameter weighting identification method for a battery equivalent circuit, including the following steps:

[0133] Step S11: Using the target temperature and target SOC as a reference group, adjust the target temperature and target SOC to obtain the neighboring temperature and neighboring SOC.

[0134] Taking a target temperature of 25℃ and a target SOC of 50% as an example, the adjusted neighboring temperatures are 24℃ and 26℃, and the adjusted neighboring SOCs are 49% and 51%. Staff can set these values ​​according to actual needs.

[0135] Step S12: Construct a neighboring group of the baseline group based on the neighboring temperature and the target SOC, and the target temperature and the neighboring SOC.

[0136] Taking the target temperature of the benchmark group as 25℃ and the target SOC as 50% as an example, examples of the benchmark group and neighboring groups are shown in Table 2.

[0137] Table 2: Examples of baseline and neighboring groups

[0138]

[0139] Step S13: Using the parameter identification method provided in Embodiment 1 above, obtain the accurate parameter values ​​of the battery internal resistance R0, the polarized RC branch, and the diffused RC branch in the reference group and the adjacent group.

[0140] The precise parameter values ​​of the reference group are denoted as: R0, R1, C1, R2, C2.

[0141] The precise parameter values ​​of neighboring group 1 are denoted as: R0_1, R1_1, C1_1, R2_1, C2_1.

[0142] The precise parameter values ​​of neighboring group 2 are denoted as: R0_2, R1_2, C1_2, R2_2, C2_2.

[0143] The precise parameter values ​​of neighboring group 3 are denoted as: R0_3, R1_3, C1_3, R2_3, C2_3.

[0144] The precise parameter values ​​of neighboring group 4 are denoted as: R0_4, R1_4, C1_4, R2_4, C2_4.

[0145] Step S14: Weighted summation of the precise parameter values ​​under the benchmark group and the neighboring group to obtain the final precise parameter values ​​of the battery internal resistance R0, the polarized RC branch and the diffused RC branch at the target temperature and the target SOC.

[0146] A moving window weighted average algorithm is used, with the center of the window receiving the highest weight and the edges receiving lower weights, such as 0.4, 0.15, 0.15, 0.15, 0.15. The target SOC is 50%, and the final parameters of the cell's equivalent circuit at a target temperature of 25℃ are:

[0147] R0_final = R0*0.4 + R0_1*0.15 + R0_2*0.15 + R0_3*0.15 + R0_4*0.15

[0148] R1_final = R1*0.4 + R1_1*0.15 + R1_2*0.15 + R1_3*0.15 + R1_4*0.15

[0149] C1_final = C1*0.4 + C1_1*0.15 + C1_2*0.15 + C1_3*0.15 + C1_4*0.15

[0150] R2_final = R2*0.4 + R2_1*0.15 + R2_2*0.15 + R2_3*0.15 + R2_4*0.15

[0151] C2_final = C2*0.4 + C2_1*0.15 + C2_2*0.15 + C2_3*0.15 + C2_4*0.15

[0152] The parameter weighting identification method provided in this invention can integrate the parameter values ​​of the benchmark group and the neighboring group, reducing the random error of single-point identification. At the same time, it can construct a smoother and more accurate functional relationship between parameters and SOC and temperature, rather than just the mean of discrete points, thus improving the engineering application value of the final parameters.

[0153] Using the parameter weighting identification method provided in this embodiment, the impedance parameters of the battery equivalent circuit under different temperatures (0℃ / 10℃ / 25℃ / 40℃) and different SOCs (0% / 10% / 20% / 30% / 40% / 50% / 60% / 70% / 80% / 90% / 100%) can be obtained. A mapping table of the parameters of the equivalent circuit with temperature and SOC can be established to improve the temperature adaptability of the model.

[0154] Example 3

[0155] This invention provides a parameter identification device for a battery equivalent circuit, adapted to a second-order RC equivalent circuit of a battery. The second-order RC equivalent circuit of a battery includes a battery internal resistance R0, a polarized RC branch, and a diffused RC branch connected in series. The parameter identification device includes:

[0156] The EIS test module is configured to perform electrochemical impedance spectroscopy (EIS) tests on the battery at the target temperature and target SOC to obtain the electrochemical impedance spectrum under high-frequency excitation.

[0157] The internal resistance determination module is configured to identify the precise value of the battery internal resistance R0 at the target temperature and target SOC based on the electrochemical impedance spectroscopy. Specifically, the precise value of the battery internal resistance R0 is the impedance value at the intersection of the high-frequency region and the real axis in the Nyquist plot of the electrochemical impedance spectroscopy.

[0158] The HPPC test module is configured to perform hybrid pulse power characteristic HPPC tests on the battery at the target temperature and target SOC, and to acquire voltage and current response data under short-time pulse current excitation and long-time pulse current excitation, respectively.

[0159] The branch determination module is configured to identify the precise parameter values ​​of the polarized RC branch and the diffused RC branch at the target temperature and target SOC based on voltage and current response data, specifically:

[0160] Construct corresponding response curves based on voltage and current response data under short-time pulse current excitation and long-time pulse current excitation;

[0161] A second-order exponential fit is performed on the zero-input response segment of the response curve under short-time pulse current excitation to calculate the rough parameter values ​​of the polarized RC branch and the diffused RC branch.

[0162] A second-order exponential fit is performed on the zero-input response segment of the response curve under long-term pulse current excitation. The coarse parameter values ​​of the polarized RC branch are added as known quantities to the fitting process to calculate the precise parameter values ​​of the diffused RC branch.

[0163] A second-order exponential fit is performed on the zero-input response segment of the response curve under short-time pulse current excitation. The precise parameter values ​​of the diffused RC branch are added as known quantities to the fitting process to calculate the precise parameter values ​​of the polarized RC branch.

[0164] Example 4

[0165] This invention provides a battery unit, including one or more of a battery module, a battery pack, or an energy storage battery. The battery unit uses the parameter identification device provided in Embodiment 3 above, or the parameter identification method provided in Embodiment 1 above, or the parameter weighted identification method provided in Embodiment 2 above to identify the parameters of the battery equivalent circuit.

[0166] Power batteries (battery modules, battery packs) are power supply devices that provide power to electric vehicles, electric trains, electric ships, etc. The main types include lead-acid batteries and lithium iron phosphate batteries; nickel-metal hydride batteries are also used in hybrid vehicles. By accurately identifying the parameters of the battery equivalent circuit of the battery pack, a battery model can be accurately constructed, enabling accurate estimation of battery state (such as SOC and SOH) and energy management.

[0167] Energy storage batteries are key devices for storing electrical energy in renewable energy systems, primarily used in solar and wind power generation equipment. Product types include lead-acid batteries (including vented, valve-regulated, and gel types) and lithium-ion batteries (represented by lithium iron phosphate). By accurately identifying the parameters of the battery's equivalent circuit, a battery model can be accurately constructed, enabling accurate battery state estimation and energy management.

[0168] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0169] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0170] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0171] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0172] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A parameter identification device of a battery equivalent circuit, adapted to a battery second-order RC equivalent circuit, the battery second-order RC equivalent circuit comprising a battery internal resistance R0, a polarization RC branch and a diffusion RC branch in series; characterized in that, The parameter identification device comprises: An EIS test module configured to perform electrochemical impedance spectroscopy (EIS) test on the battery at a target temperature and a target state of charge (SOC) to obtain an electrochemical impedance spectrum under high-frequency excitation; An internal resistance determination module configured to identify an accurate value of the battery internal resistance R0 at the target temperature and the target SOC based on the electrochemical impedance spectrum; An HPPC test module configured to perform hybrid pulse power characteristic (HPPC) test on the battery at the target temperature and the target SOC to obtain voltage and current response data under short-time pulse current excitation and long-time pulse current excitation, respectively; A branch determination module configured to identify accurate parameter values of the polarization RC branch and the diffusion RC branch at the target temperature and the target SOC based on the voltage and current response data.

2. The apparatus according to claim 1, wherein The accurate value of the battery internal resistance R0 is an intersection impedance value of a high-frequency region and a real axis in a Nyquist plot of the electrochemical impedance spectrum.

3. The apparatus according to claim 1, wherein The branch determination module is specifically configured to: construct a corresponding response curve according to the voltage and current response data under the short-time pulse current excitation and the long-time pulse current excitation; perform second-order exponential fitting on a zero-input response segment in the response curve under the short-time pulse current excitation to calculate rough parameter values of the polarization RC branch and the diffusion RC branch; perform second-order exponential fitting on a zero-input response segment in the response curve under the long-time pulse current excitation, add the rough parameter values of the polarization RC branch as known quantities to the fitting process, and calculate accurate parameter values of the diffusion RC branch; perform second-order exponential fitting on a zero-input response segment in the response curve under the short-time pulse current excitation, add the accurate parameter values of the diffusion RC branch as known quantities to the fitting process, and calculate accurate parameter values of the polarization RC branch.

4. A parameter identification method of a battery equivalent circuit, adapted to a battery second-order RC equivalent circuit, the battery second-order RC equivalent circuit comprising a battery internal resistance R0, a polarization RC branch and a diffusion RC branch in series; characterized in that, The parameter identification method comprises: performing electrochemical impedance spectroscopy (EIS) test on the battery at a target temperature and a target state of charge (SOC) to obtain an electrochemical impedance spectrum under high-frequency excitation; identifying an accurate value of the battery internal resistance R0 at the target temperature and the target SOC based on the electrochemical impedance spectrum; performing hybrid pulse power characteristic (HPPC) test on the battery at the target temperature and the target SOC to obtain voltage and current response data under short-time pulse current excitation and long-time pulse current excitation, respectively; identifying accurate parameter values of the polarization RC branch and the diffusion RC branch at the target temperature and the target SOC based on the voltage and current response data.

5. The parameter identification method of a battery equivalent circuit according to claim 4, wherein The battery is pretreated before the electrochemical impedance spectroscopy (EIS) test and the hybrid pulse power characteristic (HPPC) test, and the pretreatment comprises placing the battery in a constant-temperature environment at the target temperature, charging / discharging the battery at a constant current to the target SOC, and standing for a predetermined time length.

6. The parameter identification method of a battery equivalent circuit according to claim 4, wherein The obtaining of the electrochemical impedance spectrum under high-frequency excitation comprises: applying an alternating excitation signal in a high-frequency region to ensure that the battery is in a linear response state and obtain a corresponding electrochemical impedance spectrum; wherein the frequency range of the high-frequency region is 1 kHz to 5 kHz, and the amplitude of the alternating excitation signal is ≤10 mV; The accurate value of the battery internal resistance R0 is the intersection impedance value of the high-frequency region and the real axis in the Nyquist plot of the electrochemical impedance spectrum.

7. The parameter identification method of a battery equivalent circuit according to claim 4, wherein The duration of the short-time pulse current excitation is 10s to 60s, and the duration of the long-time pulse current excitation is 5min to 1h.

8. The parameter identification method of a battery equivalent circuit according to claim 4, wherein The identifying the accurate parameter values of the polarization RC branch and the diffusion RC branch at the target temperature and the target SOC based on the voltage and current response data comprises: Constructing the corresponding response curves according to the voltage and current response data under the short-time pulse current excitation and the long-time pulse current excitation; Performing second-order exponential fitting on the zero-input response section in the response curve under the short-time pulse current excitation to calculate the rough parameter values of the polarization RC branch and the diffusion RC branch; Performing second-order exponential fitting on the zero-input response section in the response curve under the long-time pulse current excitation, adding the rough parameter values of the polarization RC branch as known quantities into the fitting process to calculate the accurate parameter values of the diffusion RC branch; Performing second-order exponential fitting on the zero-input response section in the response curve under the short-time pulse current excitation, adding the accurate parameter values of the diffusion RC branch as known quantities into the fitting process to calculate the accurate parameter values of the polarization RC branch.

9. The parameter identification method of a battery equivalent circuit according to claim 8, wherein, The fitting formula of the second-order exponential fitting is: wherein is a coordinate point on the response curve, are the start time and the end time of the zero-input response section, respectively, is a fitting coefficient, is a constant term; The operation formula of the zero-input response section is: wherein is the voltage value of the zero-input response segment, is the resistance value of the resistance R1 and the capacitance value of the capacitance C1 in the polarization RC branch, is the resistance value of the resistance R2 and the capacitance value of the capacitance C2 in the diffusion RC branch, is the time instant within the zero-input response segment, is the current value of the pulse current, is the open-circuit voltage of the battery; The fitting formula of the zero-input response section and the operation formula are combined, and the fitting coefficient calculating parameter values of the polarization RC branch and the diffusion RC branch: 。 10. A parameter weighted identification method of a battery equivalent circuit, characterized by, Comprises: Taking the target temperature and the target SOC as the benchmark group, adjusting the target temperature and the target SOC to obtain the adjacent temperature and the adjacent SOC; According to the adjacent temperature and the target SOC, the target temperature and the adjacent SOC, constructing the adjacent group of the benchmark group; Using the parameter identification method in any one of claims 4-9, obtaining the accurate parameter values of the battery internal resistance R0, the polarization RC branch and the diffusion RC branch under the benchmark group and the adjacent group; Performing weighted summation on the accurate parameter values under the benchmark group and the adjacent group to obtain the final accurate parameter values of the battery internal resistance R0, the polarization RC branch and the diffusion RC branch at the target temperature and the target SOC.

11. A battery cell characterized by, Comprises one or more of a battery module, a battery pack or an energy storage battery, the battery cell adopts the parameter identification device in any one of claims 1-3 or adopts the parameter identification method in any one of claims 4-9 or adopts the parameter weighted identification method in claim 10 to identify the parameters of the battery equivalent circuit.

Images