Lithium battery echelon utilization method for estimating battery health state

Abstract

The invention discloses a lithium battery echelon utilization method for estimating a battery health state. The lithium battery echelon utilization method comprises the following steps: step 1, obtaining retired single batteries; 2, performing compliance judgment on the appearance of the retired single battery; 3, performing multi-parameter fusion calculation of SOH (state of health) and detection of SOC (state of charge) on the single batteries with the appearance compliance; 4, selecting the single batteries with the SOH state and the SOC state in the step 3 in a significant level in pairs to perform an independent hypothesis experiment, and screening out the single batteries meeting echelon utilization; and step 5, recombining the battery monomers according with echelon utilization into a battery module, and carrying out echelon utilization, so that the utilization rate of the retired battery is improved, and resource waste is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of recycling of retired batteries, and specifically relates to a method for the cascade utilization of lithium batteries for estimating the state of health of batteries. Background Art

[0002] Traditional cars cause great environmental pollution and serious waste of resources. In order to achieve the dual-carbon goal as soon as possible, lithium-ion batteries stand out. They have the advantages of high energy density, high power, low self-discharge rate, environmental protection, etc., and are widely used in new electric vehicles. When the capacity of the lithium battery in an electric vehicle is less than 80%, the battery capacity cannot support its endurance and mileage, and a large number of lithium batteries face retirement. If these retired batteries are not properly disposed of, it will lead to a large amount of resource waste and environmental pollution.

[0003] During the cascade utilization process of lithium batteries, they can be directly used, ignoring the differences in capacity, voltage, and internal resistance in the batteries. However, such use will cause the short-board effect, resulting in low battery utilization efficiency.

[0004] Therefore, it is necessary to provide a method for the cascade utilization of lithium batteries for estimating the state of health of batteries to solve the problems mentioned in the above background art. Summary of the Invention

[0005] To achieve the above object, the present invention provides the following technical solution: A method for the cascade utilization of lithium batteries for estimating the state of health of batteries, including: Step 1: Obtain retired battery cells;

[0006] Step 2: Judge the compliance of the appearance of the retired battery cells;

[0007] Step 3: Respectively perform multi-parameter fusion calculation of the state of health SOH and detection of the state of charge SOC on the battery cells with compliant appearance;

[0008] Step 4: Select the battery cells in pairs of the state of health SOH and the state of charge SOC in Step 3 for an independent hypothesis experiment at a significant level, and screen out the battery cells that meet the requirements for cascade utilization;

[0009] Step 5: Recombine the battery cells that meet the requirements for cascade utilization into a battery module and perform cascade utilization.

[0010] Preferably, in Step 3, the fusion parameters for the multi-parameter fusion calculation of the state of health SOH of the battery cells include: the capacity C, internal resistance R, number of cycles T, and open-circuit voltage V of the battery cells.

[0011] Preferably, the multi-parameter fusion calculation method for the state of health SOH of the battery cells includes: respectively defining the dynamic capacity-internal resistance comprehensive factor F according to the influence of the changes in the capacity, internal resistance, number of cycles, and open-circuit voltage of the battery cells on the state of health CR, the dynamic cycle life factor F T and the dynamic voltage change factor F V , to judge the influence of each parameter on the state of health (SOH) of a single battery cell;

[0012] The formula for calculating the state of health (SOH) of the single battery cell is: SOH = F CR × F T × F V × 100%.

[0013] Preferably, the calculation method of the dynamic capacity - internal resistance comprehensive factor F CR is as follows:

[0014] Using the attenuation of the capacity C and the increase of the internal resistance R of the single battery cell as the indicators to measure the state of health (SOH) of the single battery cell, the attenuation ratio of the capacity C is expressed by ; the increase ratio of the internal resistance R is expressed by ; and a weight coefficient k1 is introduced to adjust the influence degree of the change of the internal resistance R on the state of health (SOH) of the single battery cell;

[0015] The formula for calculating the dynamic capacity - internal resistance comprehensive factor F CR is:

[0016] where C n is the current actual available capacity of the battery, C i is the initial design capacity of the battery, R n is the current internal resistance of the battery, and R i is the initial internal resistance of the battery.

[0017] Preferably, the calculation method of the dynamic cycle life factor F T is as follows:

[0018] Using the increase of the battery cycle number T as the indicator to measure the state of health (SOH) of the single battery cell, the increase ratio of the battery cycle number T is expressed by ; and is used to further emphasize the influence of the increase ratio of the battery cycle number T on the state of health (SOH) of the single battery cell; meanwhile, a weight coefficient k2 is introduced to adjust the influence degree of the increase of the battery cycle number T on the state of health (SOH) of the single battery cell;

[0019] The formula for calculating the dynamic cycle life factor F T is:

[0020] where T is the battery cycle number, and T max is the maximum allowable cycle number of the battery.

[0021] Preferably, the dynamic voltage change factor FV The calculation method is:

[0022] The change of open circuit voltage V is used as an indicator to measure the health status SOH of battery cells. The increase ratio of open circuit voltage V is Indicates and introduces As an index, it indicates the influence of the open circuit voltage V on the battery cell health state SOH as the battery cycle number increases;

[0023] The dynamic voltage variation factor F V The calculation formula is:

[0024] Among them, V ocn is the current open circuit voltage of the battery, V oci is the initial open circuit voltage of the battery, and T is the number of battery cycles.

[0025] Preferably, in step 2, the method for detecting the state of charge (SOC) of the battery cell includes: determining the state of charge (SOC) of the battery cell based on an adaptive extended Kalman particle filter (AEKPF) algorithm.

[0026] Preferably, in step 4, before conducting the independent hypothesis experiment, the experimental data of the battery cells are preprocessed, and the experimental data preprocessing method includes: data normalization, variance homogeneity test, outlier processing and simplified interaction analysis.

[0027] Preferably, in step five, the battery modules are utilized in a cascade manner by combining the battery pack active balancing technology and the data-driven state of health SOH and state of charge SOC method.

[0028] Preferably, the battery module recycling method comprises:

[0029] Monitor each battery cell in the battery module in real time to obtain the real-time health status SOH and state of charge SOC of the battery cell;

[0030] Redistribute the charge of battery cells with abnormal state of charge (SOC) during charging and discharging;

[0031] Collect historical operating data of battery cells, and calculate the health status (SOH) and state of charge (SOC) of the current battery cells based on the historical operating data;

[0032] Based on the calculation results of the health status SOH of the battery cells, the batteries with abnormal health status SOH of the battery cells are removed, and new battery cells are selected to reorganize the battery module;

[0033] Recalculate and evaluate the SOH and SOC of the battery cells periodically or under specific conditions to update the health records of the battery cells.

[0034] Compared with the prior art, the present invention provides a method for the cascade utilization of lithium batteries for estimating the state of health of batteries, which has the following beneficial effects:

[0035] In the present invention, by using a multi-parameter fusion algorithm to comprehensively analyze the changes in parameters such as the capacity, internal resistance, number of charge and discharge cycles, and open-circuit voltage of retired batteries, the SOH value of retired batteries is obtained, making the calculated SOH data more accurate; at the same time, an adaptive weight adjustment mechanism is introduced to dynamically adjust the weights of various parameters in the assessment of the state of health SOH according to the characteristics and usage environments of different batteries, so as to improve the accuracy and adaptability of the assessment.

[0036] In the present invention, a two-factor repeated measures analysis of variance method is adopted to finely evaluate the state of health SOH and state of charge SOC of retired batteries, and identify battery cells with consistent performance and meeting the requirements of cascade utilization, so as to improve the cascade utilization efficiency of retired batteries.

[0037] In the present invention, by combining the active balancing technology and the data-driven method, the consistency of the SOC and SOH of each unit in the battery pack is ensured, and the battery state is predicted to optimize the battery performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic structural diagram of the process of a method for the cascade utilization of lithium batteries for estimating the state of health of batteries. DETAILED DESCRIPTION OF THE INVENTION

[0039] Please refer to Figure 1 , the present invention provides a method for the cascade utilization of lithium batteries for estimating the state of health of batteries, including: Step 1: Obtain retired battery cells;

[0040] Obtaining method: Select several batteries of the same model with different vehicle ages, mileage, and usage conditions from multiple retired electric vehicle batteries to cover different retired states and potential battery attenuation situations.

[0041] Step 2: Judge the compliance of the appearance of the retired battery cells;

[0042] Judging method: Observe whether the battery cells and battery cores are free of swelling, deformation, dislocation, leakage, wear, corrosion, etc., and measure the diameter of the battery cells with a vernier caliper, and eliminate the batteries that do not meet the requirements.

[0043] Step 3: Perform multi-parameter fusion calculation of the state of health SOH and detection of the state of charge SOC on the battery cells with compliant appearance respectively;

[0044] The fusion parameters for the multi-parameter fusion calculation of the state of health (SOH) of the battery cell include: the capacity C, internal resistance R, number of cycles T, and open-circuit voltage V of the battery cell.

[0045] The multi-parameter fusion calculation method for the state of health (SOH) of the battery cell includes: defining the dynamic capacity-internal resistance comprehensive factor F CR , dynamic cycle life factor F T , and dynamic voltage change factor F V respectively according to the influence of the capacity, internal resistance, number of cycles, and open-circuit voltage change of the battery cell on the state of health, so as to judge the influence of each parameter on the state of health (SOH) of the battery cell;

[0046] Based on the physical characteristics and aging mechanism of the battery, the calculation formula for the state of health (SOH) of the battery cell is derived as follows:

[0047] The state of health (SOH) of the battery is closely related to the attenuation of the battery capacity C and the increase of the internal resistance R. As the battery is used, its actual available capacity C n will gradually decrease, while the initial design capacity C i remains unchanged; at the same time, the increase of the internal resistance R will lead to an increase in the energy loss during the charge and discharge process of the battery, thereby affecting the performance of the battery. The calculation method of the dynamic capacity-internal resistance comprehensive factor F CR is as follows:

[0048] Using the attenuation of the battery cell capacity C and the increase of the internal resistance R as indicators to measure the state of health (SOH) of the battery cell, the attenuation ratio of the capacity C is expressed by ; the increase ratio of the internal resistance R is expressed by , and a weight coefficient k1 is introduced to adjust the influence degree of the change of the internal resistance R on the state of health (SOH) of the battery cell, so that the calculation of the state of health (SOH) of the battery cell is more in line with the actual performance change of the battery;

[0049] To sum up, the calculation formula of the dynamic capacity-internal resistance comprehensive factor F CR is:

[0050]

[0051] wherein, R n is the current internal resistance of the battery, and R i is the initial internal resistance of the battery.

[0052] Furthermore, the cycle life of the battery is another key factor for its state of health (SOH). As the number of cycles T of the battery increases, the performance of the battery will gradually decline. Therefore, the calculation method of the dynamic cycle life factor F T is:

[0053] Using the increase in the number of battery cycles T as an indicator to measure the state of health (SOH) of a single battery cell, the increase ratio of the number of battery cycles T is expressed by and is used to further emphasize the influence of the increase ratio of the number of battery cycles T on the state of health (SOH) of a single battery cell, indicating that as the number of battery cycles T approaches the maximum value, the decline rate of the state of health (SOH) of the battery accelerates; meanwhile, a weight coefficient k2 is introduced to adjust the degree of influence of the increase in the number of battery cycles T on the state of health (SOH) of a single battery cell;

[0054] The dynamic cycle life factor F T is calculated by the formula:

[0055] where T is the number of battery cycles, and T max is the maximum allowable number of battery cycles.

[0056] Furthermore, the calculation method of the dynamic voltage change factor F V is as follows:

[0057] Using the change in the open-circuit voltage V as an indicator to measure the state of health (SOH) of a single battery cell, the increase ratio of the open-circuit voltage V is expressed by and is introduced as an exponent to represent that as the number of battery cycles T increases, the degree of influence of the open-circuit voltage V on the state of health (SOH) of a single battery cell gradually weakens, so as to more accurately simulate the influence of voltage changes on the state of health at different usage stages of the battery.

[0058] The dynamic voltage change factor F V is calculated by the formula:

[0059] where V ocn is the current open-circuit voltage of the battery, V oci is the initial open-circuit voltage of the battery, and T is the number of battery cycles.

[0060] Combining the above three factors, a comprehensive battery state of health assessment algorithm is provided. The calculation formula for the state of health (SOH) of a single battery cell is: SOH = F CR × F T × F V × 100%. By comprehensively considering the influence of multiple performance parameters on the state of health of the battery, the accuracy and applicability of the calculation result of the state of health (SOH) of a single battery cell are improved.

[0061] Verify the calculation formula for the dynamic state of health (SOH):

[0062] For example, for a battery with an initial capacity C i= 100 Ah, current capacity C n = 80 Ah, initial internal resistance R i = 0.1 Ω, current internal resistance R n = 0.2 Ω, number of battery cycles T = 500 times, maximum allowable number of cycles T max = 1000 times, initial open-circuit voltage V oci = 4.2 V, current open-circuit voltage V ocn = 4.0 V, k1 = 0.1, k2 = 0.3.

[0063] Calculate according to the battery state of health SOH calculation formula of the present invention;

[0064] First, calculate the capacity-internal resistance comprehensive factor F CR

[0065]

[0066] Then calculate the cycle life factor F T

[0067]

[0068] Next, calculate the voltage change factor F V

[0069]

[0070] Finally, calculate SOH

[0071] SOH = 0.72 × 0.925 × 0.9999 × 100% ≈ 66.59%.

[0072] It should be noted that the determination of the weight coefficients k1 and k2 needs to be adjusted according to the specific type of battery (such as lithium battery, lead-acid battery, etc.), application scenarios (such as electric vehicle, energy storage system, etc.) and a large amount of experimental data to obtain a more accurate assessment of the battery state of health.

[0073] Calculate the battery state of health SOH using the conventional calculation method:

[0074]

[0075] Among them, SOH is the state of health of the lithium battery, and the value range is 0 - 100%; k is the attenuation coefficient, generally 0.1 - 0.3;

[0076] C n is the current capacity of the lithium battery, in Ah; C i is the initial capacity of the lithium battery, in Ah, and when the initial capacity C of a battery i = 100 Ah, current capacity Cn = 80 Ah, when k takes 0.1

[0077]

[0078] That is to say, through the above calculations, when the battery capacity is 80%, the SOH of this battery is 92%, indicating that the battery is in good health condition and the capacity has only decayed by 8%. However, generally, the remaining capacity of retired batteries is used as the basis for cascade utilization. ① When the remaining available capacity of retired lithium batteries is above 80% of the rated capacity, it can be considered that the capacity of the lithium battery can meet the performance requirements of electric vehicles; ② When the remaining capacity of retired lithium batteries is between 20% and 80%, its remaining capacity can meet the performance requirements of low-power scenarios and should be subject to cascade utilization. However, the SOH values calculated by ordinary conventional formulas involve fewer calculations and reference values, resulting in a higher final result and not conforming to actual applications; ③ When the remaining capacity of retired lithium batteries is below 20%, they should be disassembled for material recycling.

[0079] Taking the value of the lithium battery coefficient as an example (the evaluated value of SOH may vary due to abnormal resistance and state of charge. We select batteries with SOH ≥ 69% for cascade utilization in the use of normal electrical appliances according to the formula, 47% ≤ SOH < 69% for electrical appliances with low power, and SOH < 47% no longer meets the usage requirements and should be disassembled to recycle useful materials)

[0080] Battery capacity ≥80％ 80％～20％ ≤20％ k1 0.1 0.3 0.5 k2 0.1 0.3 0.5 SOH ≥69％ ≥47％ ＜47％

[0081] Furthermore, in step two, the method for detecting the state of charge SOC of the battery cell includes: determining the state of charge SOC of the battery cell based on the adaptive extended Kalman particle filter AEKPF algorithm.

[0082] Specifically, it is carried out based on Matlab software. The process of the AEKPF algorithm is roughly as follows:

[0083] Initialization: Set the initial state and covariance matrix, and select an appropriate number of particles and their distribution.

[0084] Prediction step: Use the AEKF algorithm to predict the system state. At the same time, use particle filter to sample the predicted state.

[0085] Update step: According to the new observation data, use the AEKF algorithm to update the state estimate and covariance. At the same time, update the particle weights and perform resampling to avoid particle degradation.

[0086] SOC: Combining the degradation characteristics of the battery, use the updated state estimate value to correct the estimate of SOC.

[0087] Output the simulation results in the Matlab environment.

[0088] The experimental results show that after adopting this combination method, the multi-dimensional inconsistency evaluation indexes, including the maximum available capacity, terminal voltage, and SOC (state of charge of the battery), are all reduced by more than 20%. This reduction is beneficial to improving the reuse rate of lithium-ion battery recombination, enabling the battery to be more effectively utilized in application scenarios such as backup power supplies for communication base stations, distributed energy storage systems, and photovoltaic power stations.

[0089] The specific calculation process is as follows:

[0090] (1) State prediction:

[0091] State prediction equation:

[0092] Among them, X f is the prior state estimate at time f, is the state transition function, U f-1 is the control input, X f-1 is the posterior state estimate at time f-1.

[0093] (2) Error covariance prediction:

[0094] Error covariance prediction equation

[0095] Among them, D f is the prior error covariance at time f, F f is the state transition Jacobian matrix, P f is the process noise covariance.

[0096] (3) Kalman gain calculation:

[0097] Kalman gain equation:

[0098] Among them, K f is the Kalman gain, C f is the observation Jacobian matrix, Q f is the observation noise covariance.

[0099] (4) State update:

[0100] State update equation: X f =X f-1 +K f (y k -C f X f-1 )

[0101] Among them, X f is the posterior state estimate at time f, fk is the observed value at time k.

[0102] (5) Error covariance update:

[0103] Error covariance update equation: D f = (I - K f C f )D f-1

[0104] where D f is the posterior error covariance at time f, and I is the identity matrix.

[0105] Substitution steps:

[0106] ① Initialization:

[0107] Set the initial state estimate X0 and the initial error covariance D0.

[0108] ② Prediction step:

[0109] Use the state prediction equation and the error covariance prediction equation to predict the state and covariance.

[0110] ③ Update step:

[0111] Calculate the Kalman gain, and then use the state update equation and the error covariance update equation to update the state and covariance.

[0112] ④ Particle filtering step (a step unique to AEKPF):

[0113] Generate a set of particles, each particle representing a possible state.

[0114] Update the weights of the particles according to the importance sampling weights.

[0115] Resample the particles according to the particle weights to reduce the particle degeneracy problem.

[0116] ⑤ Iteration step (a step unique to AEKPF):

[0117] Iteratively update the state estimate until convergence or the maximum number of iterations is reached.

[0118] ⑥ Output SOC estimate:

[0119] Use the final state estimate as the estimate of SOC.

[0120] When using the general steps and formulas of the AEKPF algorithm, the specific implementation needs to vary according to different battery models and application scenarios. In practical applications, the algorithm needs to be adjusted and optimized to adapt to specific battery characteristics and measurement noise characteristics. The batteries that meet the conditions are used as the batteries to be sorted, preparing for the next step of battery screening.

[0121] Step 4: Select the battery cells with the state of health SOH and state of charge SOC in pairs at a significant level in Step 3 for independent hypothesis testing, and screen out the battery cells that meet the requirements for cascade utilization;

[0122] Screen the retired batteries based on the state of health SOH and state of charge SOC of the batteries (equivalent to screening the retired batteries based on the two-way ANOVA), and the specific method of the independent hypothesis testing is as follows:

[0123] Denote the state of charge SOC of the battery as A and the state of health SOH as B.

[0124] Factor A has r different levels A1, A2,..., A r , and factor B has s different levels B1, B2,..., B s , so there are a total of r×s different level combinations. For each combination A i ×B j conduct l (l﹥1) independent repeated tests, and a total of n = s×r×l observed values are obtained. The list is as follows:

[0125]

[0126] Regard the index under the level combination A i ×B j as a population, denoted as X ij , assume X ij ~N(μ ij , σ 2 ), and X ij are independent of each other. Denote ε ijk -μ ij , ε ijk the random error of the k-th test under the level combination A i ×B j . Generally, assume ε ijk ~N(0, σ 2 ), and ε ijk are independent of each other.

[0127] The notations are as follows:

[0128] The average value μ of the expected levels of each population is:

[0129]

[0130] A i The average value of each overall expectation under it is μ i· :

[0131]

[0132] B j The average level value μ of each overall expectation under it ·j is:

[0133]

[0134] Mathematical model

[0135] X ijk = μ + a i + b j +(ab) ij + ε ijk

[0136] ε ijk ~N(0, σ 2 ), each ε ijk is independent

[0137]

[0138] i = 1, 2,..., r; j = 1, 2,..., s; k = 1, 2,..., l.

[0139] Main task

[0140] Under the given significance level α (when the left and right significance levels are the same, α is generally taken as 0.05), test the hypotheses:

[0141] H 01 : a1 = a1 =... = a r = 0

[0142] H 11 : a1, a2,..., a r are not all zero

[0143] H 02 : b1 = b2 =... = b s = 0

[0144] H 12 : b1, b2,..., b s are not all zero

[0145] H 03 : (ab) 11 = (ab) 12 =... = (ab) rs = 0

[0146] H 13 : (ab) 11 , (ab) 12 ,..., (ab) rs not all zero

[0147] Then: total sum of squared deviations

[0148] Sum of squared errors:

[0149] Sum of squared interaction effects;

[0150]

[0151] Test statistic and its distribution

[0152] Under the assumption conditions of the two-way repeated measures ANOVA mathematical model, the following conclusions are obtained:

[0153]

[0154] When H 03 is true, S l and S e are independent of each other, and there is

[0155] Thus, there is

[0156] When F1 ≥ F α ((r - 1)(s - 1), rs(l - 1)), reject H 03 , that is, it is considered that the interaction effect of SOH and SOC has a significant impact on the battery, that is, it is considered that after repeated detection experiments on multiple retired batteries, the effects of SOH and SOC of the selected battery are not ideal and cannot be used for recombination;

[0157] When F1 < F α ((r - 1)(s - 1), rs(l - 1)), accept H 03 , that is, it is considered that the interaction effect of SOH and SOC on the battery is not significant, that is, it is considered that after repeated detection experiments on multiple retired batteries, the SOH and SOC of the selected battery are more in line with cascade utilization and can be used for recombination.

[0158] Furthermore, in step four, before conducting the independent hypothesis experiment, the experimental data of the battery cells are preprocessed. The experimental data preprocessing method includes: data normalization, variance homogeneity test, outlier processing, and simplified interaction analysis.

[0159] Specifically, data normalization: By means of data transformation methods such as logarithmic transformation, Box-Cox transformation, etc., the data is made closer to a normal distribution, thereby improving the applicability and accuracy of variance analysis.

[0160] Homogeneity of variance test: Before conducting variance analysis, a homogeneity of variance test is first performed, such as the Levene test. If the test result shows that the variances are not homogeneous, Welch's t-test or non-parametric methods such as the Kruskal-Wallis test can be considered. These methods have lower requirements for homogeneity of variance and can provide more reliable statistical test results in the case of non-homogeneous variances.

[0161] Outlier handling: Identifying and handling outliers: Using statistical methods such as Z-score, IQR (interquartile range), etc., to identify outliers in the data. For the identified outliers, different handling strategies can be adopted, such as replacing them with the median or mean, or directly removing them. These methods can reduce the impact of outliers on the analysis results and improve the robustness of the analysis.

[0162] Simplifying interaction analysis: First, analyze the main effects separately, that is, analyze the effects of SOH and SOC on the battery state separately, and then analyze their interaction. If the interaction is not significant, the model can be simplified to only consider the main effects. This method can reduce the complexity of the analysis and improve the interpretability of the results.

[0163] Step Five: Recombine the battery monomers that meet the requirements for cascade utilization into battery modules and perform cascade utilization.

[0164] The cascade utilization of the battery module is carried out by combining the active equalization technology of the battery pack and the method based on data-driven state of health SOH and state of charge SOC.

[0165] Furthermore, the method for cascade utilization of the battery module includes:

[0166] 5.1 Monitor each battery cell in the battery module in real time to obtain the real-time state of health SOH and state of charge SOC of the battery cell, and ensure that both are within a safe and effective range;

[0167] 5.2 Re-distribute the charge during charging and discharging for the battery cells with abnormal state of charge SOC.

[0168] Specifically, the active equalization technology is used to re-distribute the charge from the battery cells with high SOC to the battery cells with low SOC, while considering SOH to avoid further damage to the battery cells with low health status. Specifically, it is implemented by an equalization controller, such as the active battery equalization controllers LT8584 and LTC3300 of ADI Corporation, to achieve the charge transfer between battery cells, and at the same time optimize the controller parameters to adapt to battery cells with different SOH.

[0169] 5.3 Collect the historical operation data of the battery cells, and calculate the state of health (SOH) and state of charge (SOC) of the current battery cells based on the historical operation data;

[0170] Among them, the data collection includes parameters such as voltage, current, and temperature, which are convenient for calculating the SOH and SOC values. Methods such as the improved cooperative particle swarm optimization (CCPSO) algorithm (prior art) are used to extract features related to the battery health state and charging state from the collected data.

[0171] 5.4 Based on the calculation results of the state of health (SOH) of the battery cells, remove the battery cells with abnormal SOH of the battery cells, and select new battery cells for battery module recombination to keep the battery module always in better performance;

[0172] Among them, the selection criteria for the new battery cells are as follows: select according to the calculation results of the SOH of other battery cells in step 5.4, and select battery cells with the same level of SOH.

[0173] 5.5 Re - calculate and evaluate the SOH and SOC of the battery cells regularly or under specific conditions, update the health files of the battery cells, and ensure the effectiveness and functionality of the batteries for cascade utilization. Among them, "regularly" refers to maintenance plans, etc., and "specific conditions" refer to abnormal fluctuations in battery temperature, long - term non - use of the battery, etc.

[0174] The above - mentioned is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A method for the cascade utilization of lithium batteries to estimate the state of health of the battery, characterized in that, Including: Step 1: Obtain retired battery cells; Step 2: Conduct compliance judgment on the appearance of the retired battery cells; Step 3: Conduct multi-parameter fusion calculation of the state of health (SOH) and detection of the state of charge (SOC) for the battery cells with compliant appearance respectively; Step 4: Select the battery cells with SOH and SOC in pairs at a significant level in Step 3 for independent hypothesis testing, and screen out the battery cells suitable for cascade utilization; Step 5: Recombine the battery cells suitable for cascade utilization into a battery module and conduct cascade utilization.

2. The method for hierarchical utilization of lithium batteries for estimating the state of health of a battery according to claim 1, wherein In Step 3, the fusion parameters for the multi-parameter fusion calculation of the SOH of the battery cells include: the capacity C, internal resistance R, number of cycles T, and open-circuit voltage V of the battery cells.

3. A lithium battery cascade utilization method for estimating the state of health of a battery according to claim 1, characterized in that, The multi-parameter fusion calculation method for the state of health (SOH) of a battery cell includes: defining a dynamic capacity-internal resistance comprehensive factor F based on the impacts of the capacity, internal resistance, number of charge-discharge cycles, and open-circuit voltage change of the battery cell on the state of health CR , a dynamic cycle life factor F T , and a dynamic voltage change factor F V to determine the impacts of various parameters on the state of health (SOH) of the battery cell; The calculation formula for the state of health (SOH) of the battery cell is: SOH = F CR × F T × F V × 100%.

4. A method for the cascade utilization of lithium batteries for estimating the state of health of a battery according to claim 3, characterized in that, The calculation method of the dynamic capacity-internal resistance comprehensive factor F CR is as follows: The capacity C decay and internal resistance R increase of a single battery cell are used as indicators to measure the state of health (SOH) of the single battery cell. The capacity C decay ratio is expressed as ; the internal resistance R increase ratio is expressed as , and a weight coefficient k1 is introduced to adjust the influence degree of the change in internal resistance R on the state of health (SOH) of the single battery cell. The dynamic capacity-internal resistance comprehensive factor F CR The calculation formula is as follows: Among them, C n is the current actual available capacity of the battery, and C i is the initial design capacity of the battery. R n is the current internal resistance of the battery, and R i is the initial internal resistance of the battery.

5. The method for the hierarchical utilization of lithium batteries for estimating the state of health of a battery according to claim 3, characterized in that, The calculation method of the dynamic cycle life factor F T is as follows: Using the increase in the number of battery cycles T as an indicator to measure the state of health SOH of a single battery cell, the increase ratio of the number of battery cycles T is represented by and is used to further emphasize the impact of the increase ratio of the number of battery cycles T on the state of health SOH of a single battery cell; meanwhile, a weight coefficient k2 is introduced to adjust the degree of influence of the increase in the number of battery cycles T on the state of health SOH of a single battery cell; The dynamic cyclic life factor F T The calculation formula is as follows: Among them, T is the number of battery cycles, and T max is the maximum allowable number of battery cycles.

6. A method for lithium battery cascade utilization for estimating the state of health of a battery according to claim 3, characterized in that, The calculation method of the dynamic voltage change factor F V is as follows: Using the change in the open-circuit voltage V as an indicator to measure the state of health (SOH) of a battery cell, the increase ratio of the open-circuit voltage V is represented by and introduce as an exponent to represent the degree of influence of the open-circuit voltage V on the state of health (SOH) of the battery cell as the number of battery cycles increases; The dynamic voltage change factor F V The calculation formula is as follows: Among them, V ocn is the current open-circuit voltage of the battery, V oci is the initial open-circuit voltage of the battery, and T is the number of battery cycles.

7. A method for lithium battery cascade utilization for estimating the state of health of a battery according to claim 1, characterized in that, In Step 2, the method for detecting the SOC of the battery cells includes: determining the SOC of the battery cells based on the adaptive extended Kalman particle filter (AEKPF) algorithm.

8. A method for the hierarchical utilization of lithium batteries for estimating the state of health of a battery according to claim 1, characterized in that, In Step 4, before conducting the independent hypothesis testing, preprocess the experimental data of the battery cells. The experimental data preprocessing method includes: data normalization, variance homogeneity test, outlier processing, and simplified interaction analysis.

9. A method for the cascade utilization of lithium batteries for estimating the state of health of a battery according to claim 1, characterized in that, In Step 5, conduct cascade utilization of the battery module by combining the battery pack active balancing technology and the methods based on data-driven SOH and SOC.

10. The method for cascaded utilization of lithium batteries for estimating the state of health of a battery according to claim 9, characterized in that, The method for cascade utilization of the battery module includes: Monitor each battery unit in the battery module in real time to obtain the real-time SOH and SOC of the battery unit; Re-distribute the charge during charging and discharging for the battery unit with abnormal SOC; Collect the historical operation data of the battery unit and calculate the current SOH and SOC of the battery unit based on the historical operation data; Based on the calculation result of the SOH of the battery unit, remove the battery with abnormal SOH of the battery unit and select a new battery unit to recombine the battery module; Periodically or under specific conditions, re-calculate and evaluate the SOH and SOC of the battery unit to update the health file of the battery unit.

Images