Lithium battery pack inconsistency evaluation method based on model parameter fuzzy comprehensive evaluation

Abstract

The invention provides a fuzzy comprehensive evaluation method based on model parameters, which is used for evaluating the inconsistency of a lithium battery pack. Firstly, a first-order RC equivalent circuit model of the lithium battery is established, and model parameters (such as ohmic internal resistance, polarization internal resistance and polarization capacitance) are identified by adopting a recursive least square algorithm, so that key characteristic parameters of the battery are extracted. And secondly, by calculating inconsistency indexes such as range coefficients, standard deviation coefficients, variation coefficients and average absolute deviation of all battery monomers in the battery pack and combining a fuzzy comprehensive evaluation method, grade division and comprehensive evaluation are performed on the inconsistency of the battery pack. According to the method, multi-dimensional parameters are integrated, the inconsistency state of the battery pack can be comprehensively and dynamically reflected, a scientific basis is provided for design optimization, management and maintenance of the battery pack, and meanwhile the performance, the service life and the safety of the system are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery management, and in particular relates to a lithium battery pack inconsistency assessment method based on fuzzy comprehensive evaluation of model parameters. Background Art

[0002] In recent years, with the rapid development of new energy vehicles and energy storage systems, lithium-ion batteries, as core energy storage units, play an important role in improving energy utilization efficiency. However, battery packs composed of multiple single cells connected in series or in parallel face serious inconsistency problems in actual operation, which become the key bottleneck limiting their performance, life and safety. The inconsistency of battery packs mainly comes from initial differences in the manufacturing process (such as deviations in parameters such as the capacity and internal resistance of single cells) and environmental differences in the operation process (such as uneven temperature distribution, charge and discharge current distribution, etc.), which will be further amplified as the use time increases. The impact of inconsistency on battery packs is multifaceted. On the one hand, it will cause the actual capacity of the battery pack to be limited by the single cell with the worst performance, thereby significantly reducing the energy utilization of the system; on the other hand, inconsistency will aggravate the uneven degradation between single cells and shorten the overall service life of the battery pack. In addition, inconsistency may also lead to extreme conditions such as overcharging and over-discharging of local single cells, thereby causing thermal runaway and increasing the safety risk of the system.

[0003] Therefore, a new inconsistency assessment method is urgently needed that can comprehensively analyze the multi-dimensional parameters of all battery cells in the battery pack and build a high-precision dynamic assessment model. This method should comprehensively and accurately assess the current inconsistency status of the battery pack, thereby providing a scientific basis for battery pack design optimization, management strategy formulation, and maintenance. This will not only help improve the energy utilization efficiency of the system, but also extend the service life of the battery pack and significantly improve operational safety, meeting the high performance requirements of modern energy storage and new energy vehicle systems. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a lithium battery pack inconsistency assessment method based on fuzzy comprehensive evaluation of model parameters, which adopts the following technical solutions: including modeling, parameter identification, and battery pack inconsistency assessment.

[0005] In a preferred embodiment, step 1: establish a first-order RC equivalent circuit model to describe the behavior and characteristics of the lithium battery, which includes an open circuit voltage source Vocv, an ohmic resistor R0 and an RC network; the RC network consists of a polarization resistor R1 and a polarization capacitor C1;

[0006] According to the circuit principle, the system model is established as follows:

[0007] V t =V ocv+V1+R0I (1)

[0008] Where V1 represents the polarization voltage, which is defined as follows:

[0009]

[0010] Step 2: Identify the parameters of the model. The specific steps are as follows:

[0011] The recursive least squares algorithm with forgetting factor (FFRLS) is used to identify the parameters of the first-order RC equivalent circuit model, so as to obtain the ohmic internal resistance R0, polarization internal resistance R1 and polarization capacitance C1 of the lithium battery; in order to meet the form of the RLS algorithm, the system model needs to be discretized. First, the polarization voltage equation is discretized as follows:

[0012]

[0013] Where Δt represents the sampling time; the voltage difference is defined as follows:

[0014]

[0015] Since ΔV(k-1)=V t (k-1)-V ocv (k-1), so we can further get:

[0016]

[0017] Since Δt is very small, V ocv (k)≈V ocv (k-1), so the above formula can be converted to:

[0018]

[0019] definition:

[0020]

[0021] The first-order RC equivalent circuit model parameters are solved as follows:

[0022]

[0023] Where θ = [θ1, θ2, θ3, θ4];

[0024] Then, the FFRLS algorithm is used to perform parameter identification based on the collected current and voltage data of the battery cell. The specific steps are as follows:

[0025] The input-output model of the system to be identified can be expressed as:

[0026] V(k)=X T(k)θ+e(k) (9)

[0027] Where V(K) is the kth sampling value of the battery terminal voltage, and e(k) is the measurement error of the system;

[0028] The algorithm flow is as follows:

[0029] 1. Initialize the forgetting factor λ, covariance matrix P, parameter matrix θ, and error matrix e;

[0030] 2. Calculate the gain:

[0031]

[0032] 3. Update the covariance matrix:

[0033]

[0034] 4. Calculation error:

[0035] e(k)=V(k)-X T (k)θ(k-1) (12)

[0036] 5. Update parameter matrix:

[0037] θ(k)=θ(k-1)+K(k)e(k) (13)

[0038] 6. Calculate the model parameters according to formula 8;

[0039] 7. Step 3: Battery pack inconsistency evaluation, the specific steps are as follows:

[0040] Step 3-1: Calculate the ohmic internal resistance, polarization internal resistance and polarization capacitance of all battery cells in the battery pack from the collected battery data according to steps 1 and 2; and calculate their range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation and Gini coefficient respectively; the calculation formula of the range coefficient R is as follows:

[0041] R=X max -X min (14)

[0042] Where X max Represents the maximum value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack, X min Represents the minimum value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack;

[0043] The calculation formula of standard deviation coefficient σ is as follows:

[0044]

[0045] Among them, X i are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack, and N is the number of battery cells that make up the battery pack;

[0046] The calculation formula of coefficient of variation CV is as follows:

[0047]

[0048] Where σ represents the standard deviation of the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack. It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack;

[0049] The calculation formula of mean absolute deviation MAD is as follows:

[0050]

[0051] Among them, X i are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack;

[0052] The calculation formula of Gini coefficient G is as follows:

[0053]

[0054] Among them, X i and X j are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack;

[0055] Step 3-2: Establish the inconsistency level of lithium battery pack; set the inconsistency level into four levels: "excellent", "good", "pass", and "fail". The inconsistency indicators include the five indicators of range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation and Gini coefficient introduced in step 3-1. This range can be flexibly adjusted according to the specific evaluation object and actual needs, as shown in Table 1; the level "excellent" means that the consistency performance is excellent and can be used normally; the level "good" means that the consistency performance level is good and can continue to be used. When it is at the operation and maintenance time node, balanced maintenance can be performed; the level "pass" means that the consistency level of the energy storage battery is general at this time, and balanced maintenance should be applied according to the situation; the level "fail" means that the consistency of the energy storage battery is very poor at this time, and the battery should be shut down immediately for balanced maintenance;

[0056] Step 3-3: Calculate the inconsistency index of all battery cells in the battery pack using equations (14) to (18) according to the ohmic internal resistance, polarization internal resistance, and polarization capacitance of all battery cells in the battery pack identified in step 2;

[0057] Step 3-4: Determine the inconsistency evaluation factor set U = {range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, Gini coefficient}, the inconsistency evaluation grade set V = {excellent, good, pass, fail} and the inconsistency evaluation factor weight set A = {w1, w2, w3, w4, w5}, where the weight set can be set according to personal preference, as long as w1+w2+w3+w4+w5=1 is satisfied. Here we set w1=w2=w3=w4=w5=0.25;

[0058] Step 3-5: Perform single factor evaluation on the range coefficient, standard deviation coefficient, coefficient of variation, and mean absolute deviation coefficient to obtain r i = {r i1 ,r i2 ,r i3 ,r i4 ,r i5}, which indicates the proportion of the five evaluation indicators (range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, and Gini coefficient) of all battery cells in the battery pack in the inconsistency evaluation level ("excellent", "good", "pass", and "fail");

[0059] Step 3-6: Construct a comprehensive evaluation matrix R of battery pack inconsistency based on the five single-factor evaluation sets obtained, as shown in Formula 19, where R is a rows and b columns, a is the number of battery pack inconsistency evaluation indicators, which is set to 5 here (range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, Gini coefficient), and b represents the number of battery pack inconsistency evaluation levels, which is set to 4 here ("excellent", "good", "pass", "fail");

[0060]

[0061] Step 3-7: First, determine the value of the fourth column in the matrix R. If any item is greater than 0, the battery pack inconsistency evaluation level is "failed". If all items in the fourth column of the matrix R are 0, the matrix evaluation B is calculated based on the weight matrix A and the fuzzy relationship matrix R, that is, B = A × R = (b1, b2, b3, b4), where b1, b2, b3, b4, b5 represent the probabilities of consistency performance being "excellent", "good", "passed", and "failed", respectively;

[0062] Step 3-8: Calculate the inconsistency degree of the battery pack according to the obtained evaluation matrix B, and the calculation formula is shown in formula (20), where ω1=100, ω2=80, ω3=60, ω4=40;

[0063] μ=ω1b1+ω2b2+ω3b3+ω4b4 (20)

[0064] Then, a comprehensive evaluation is made on the battery pack inconsistency to determine the battery pack inconsistency evaluation score (40-100). After obtaining the battery pack inconsistency evaluation score, the battery pack inconsistency evaluation level is determined according to the comparison table in Table 2; for example, if the matrix B = (0.75 0.25 0 0), the inconsistency evaluation score is μ = 0.75 × 100 + 0.25 × 80 + 0 × 60 + 0 × 40 = 95, and the battery pack inconsistency performance level is defined as "excellent".

[0065] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0066] (1) High-precision evaluation: Extract key battery parameters based on the first-order RC equivalent circuit model and combine it with the recursive least squares algorithm to ensure high accuracy and reliability of model parameter identification.

[0067] (2) Comprehensive and multi-dimensional analysis: The battery pack status is comprehensively evaluated through multi-dimensional inconsistency indicators (range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation, and Gini coefficient), which can fully reflect the inconsistency characteristics of the battery pack.

[0068] (3) Flexibility and dynamic adaptability: The fuzzy comprehensive evaluation method is adopted, which allows the weight distribution to be adjusted according to actual needs, and has the ability to dynamically adjust and output evaluation results in real time under complex operating conditions.

[0069] (4) Intuitiveness and practicality: The evaluation results are made intuitive by grading the inconsistency levels (excellent, good, pass, fail), which is convenient for application in battery pack management and optimization strategy formulation.

[0070] (5) Wide applicability: The method is applicable to a variety of battery pack configurations and operating environments, providing a scientific basis for the design optimization, balanced management, and maintenance of lithium battery packs, and has good engineering application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 is the first-order RC equivalent circuit model of the present invention;

[0072] Figure 2 This is a flow chart for evaluating the present invention.

[0073] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] The following will be combined Figure 1-Figure 2 The lithium battery pack inconsistency evaluation method based on fuzzy comprehensive evaluation of model parameters of the present invention is described in detail. In order to solve the above problems, the present invention provides a lithium battery pack inconsistency evaluation method based on fuzzy comprehensive evaluation of model parameters. The method provided by the present invention is based on physical model parameters, and the evaluation result has high theoretical credibility and accuracy. At the same time, combined with the fuzzy comprehensive evaluation method, it can effectively integrate various information, improve the comprehensiveness and flexibility of the analysis, so that the inconsistency of the battery pack can be accurately evaluated.

[0075] The technical solution adopted by the present invention to solve its technical problem mainly includes the following three steps (modeling, parameter identification, battery pack inconsistency assessment), wherein the technical solution flow chart is as follows Figure 2 shown.

[0076] Step 1: Establish a first-order RC equivalent circuit model to describe the behavior and characteristics of lithium batteries. Its structure is as follows Figure 1 As shown in Figure 1, it contains an open circuit voltage source Vocv, an ohmic resistor R0 and an RC network. The RC network consists of a polarized resistor R1 and a polarized capacitor C1.

[0077] According to the circuit principle, the system model is established as follows:

[0078] V t =V ocv +V1+R0I (1)

[0079] Where V1 represents the polarization voltage, which is defined as follows:

[0080]

[0081] Step 2: Identify the parameters of the model. The specific steps are as follows:

[0082] The recursive least squares algorithm with forgetting factor (FFRLS) is used to identify the parameters of the established first-order RC equivalent circuit model, thereby obtaining the ohmic internal resistance R0, polarization internal resistance R1 and polarization capacitance C1 of the lithium battery. In order to meet the form of the RLS algorithm, the system model needs to be discretized. First, the polarization voltage equation is discretized as follows:

[0083]

[0084] Where Δt represents the sampling time. The voltage difference is defined as follows:

[0085]

[0086] Since ΔV(k-1)=V t (k-1)-V ocv (k-1), so we can further get:

[0087]

[0088] Since Δt is very small, V ocv (k)≈V ocv (k-1), so the above formula can be converted to:

[0089]

[0090] definition:

[0091]

[0092] The first-order RC equivalent circuit model parameters are solved as follows:

[0093]

[0094] Where θ = [θ1, θ2, θ3, θ4].

[0095] Then, the FFRLS algorithm is used to perform parameter identification based on the collected current and voltage data of the battery cell. The specific steps are as follows:

[0096] The input-output model of the system to be identified can be expressed as:

[0097] V(k)=X T (k)θ+e(k) (9)

[0098] Where V(K) is the kth sampling value of the battery terminal voltage, and e(k) is the measurement error of the system.

[0099] The algorithm flow is as follows:

[0100] 1. Initialize the forgetting factor λ, covariance matrix P, parameter matrix θ, and error matrix e.

[0101] 2. Calculate the gain:

[0102]

[0103] 3. Update the covariance matrix:

[0104]

[0105] 4. Calculation error:

[0106] e(k)=V(k)-X T (k)θ(k-1) (12)

[0107] 5. Update parameter matrix:

[0108] θ(k)=θ(k-1)+K(k)e(k) (13)

[0109] 6. Calculate the model parameters according to formula 8.

[0110] Step 3: Battery pack inconsistency evaluation, the specific steps are as follows:

[0111] Step 3-1: Calculate the ohmic internal resistance, polarization internal resistance and polarization capacitance of all battery cells in the battery pack from the collected battery data according to steps 1 and 2. And calculate their range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation and Gini coefficient respectively. The calculation formula of the range coefficient R is as follows:

[0112] R=X max -X min (14)

[0113] Where X max Represents the maximum value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack, X min Represents the minimum value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack.

[0114] The calculation formula of standard deviation coefficient σ is as follows:

[0115]

[0116] Among them, X i are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack, and N is the number of battery cells that make up the battery pack.

[0117] The calculation formula of coefficient of variation CV is as follows:

[0118]

[0119] Where σ represents the standard deviation of the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack. It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack.

[0120] The calculation formula of mean absolute deviation MAD is as follows:

[0121]

[0122] Among them, X i are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack.

[0123] The calculation formula of Gini coefficient G is as follows:

[0124]

[0125] Among them, X i and X j are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack.

[0126] Step 3-2: Establish the inconsistency level of lithium battery pack. The inconsistency level is set to four levels: "excellent", "good", "pass", and "fail". The inconsistency indicators include the five indicators of range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation and Gini coefficient introduced in step 3-1. This range can be flexibly adjusted according to the specific evaluation object and actual needs, as shown in Table 1. Among them, the level of "excellent" means that the consistency performance is excellent and can be used normally; the level of "good" means that the consistency performance level is good and can continue to be used. When it is at the operation and maintenance time node, balanced maintenance can be performed; the level of "pass" means that the consistency level of the energy storage battery is general at this time, and balanced maintenance should be applied according to the situation; the level of "fail" means that the consistency of the energy storage battery is very poor at this time, and the battery should be shut down immediately for balanced maintenance.

[0127]

[0128]

[0129] Table 1 Relationship between inconsistency evaluation index and inconsistency level

[0130] Step 3-3: Based on the ohmic internal resistance, polarization internal resistance, and polarization capacitance of all battery cells in the battery pack identified in step 2, use equations (14) to (18) to calculate the inconsistency indexes of all battery cells in the battery pack.

[0131] Step 3-4: Determine the inconsistency evaluation factor set U = {range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, Gini coefficient}, the inconsistency evaluation grade set V = {excellent, good, pass, fail} and the inconsistency evaluation factor weight set A = {w1, w2, w3, w4, w5}, where the weight set can be set according to personal preference, as long as w1+w2+w3+w4+w5=1 is satisfied. Here we set w1=w2=w3=w4=w5=0.25.

[0132] Step 3-5: Perform single factor evaluation on the range coefficient, standard deviation coefficient, coefficient of variation, and mean absolute deviation coefficient to obtain r i = {r i1 ,r i2 ,r i3 ,r i4 ,r i5}, which indicates the proportion of the five evaluation indicators (range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, and Gini coefficient) of all battery cells in the battery pack in the inconsistency evaluation level ("excellent", "good", "pass", and "fail") respectively.

[0133] Step 3-6: Construct a comprehensive evaluation matrix R of battery pack inconsistency based on the five single-factor evaluation sets obtained, as shown in Formula 19, where R is a rows and b columns, a is the number of battery pack inconsistency evaluation indicators, which is set to 5 here (range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, Gini coefficient), and b represents the number of battery pack inconsistency evaluation levels, which is set to 4 here ("excellent", "good", "pass", and "fail").

[0134]

[0135] Step 3-7: First, determine the value of the fourth column in the matrix R. If any item is greater than 0, the battery pack inconsistency evaluation level is "fail". If all items in the fourth column of the matrix R are 0, the matrix evaluation B is calculated based on the weight matrix A and the fuzzy relationship matrix R, that is, B = A × R = (b1, b2, b3, b4), where b1, b2, b3, b4, b5 represent the probabilities of the consistency performance being "excellent", "good", "pass", and "fail", respectively.

[0136] Step 3-8: Calculate the inconsistency degree of the battery pack according to the obtained evaluation matrix B. The calculation formula is shown in formula (20), where ω1=100, ω2=80, ω3=60, ω4=40.

[0137] μ=ω1b1+ω2b2+ω3b3+ω4b4 (20)

[0138] Then, a comprehensive evaluation is made on the battery pack inconsistency to determine the battery pack inconsistency evaluation score (40-100). After obtaining the battery pack inconsistency evaluation score, the battery pack inconsistency evaluation level is determined according to the comparison table in Table 2. For example, if the matrix B = (0.75 0.25 0 0), the inconsistency evaluation score is μ = 0.75 × 100 + 0.25 × 80 + 0 × 60 + 0 × 40 = 95, and the battery pack inconsistency performance level is defined as "excellent".

[0139] Inconsistency Assessment Level excellent good Pass Fail Inconsistency Assessment Score [100,85) [85,70) [70,55) [55,40)

[0140] Table 2 Inconsistency evaluation score comparison table

[0141] Working principle: The present invention has high-precision evaluation: based on the first-order RC equivalent circuit model, the key parameters of the battery are extracted, combined with the recursive least squares algorithm to ensure the high accuracy and reliability of the model parameter identification; comprehensive and multi-dimensional analysis: through multi-dimensional inconsistency indicators (range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation, Gini coefficient) to comprehensively evaluate the battery pack status, it can comprehensively reflect the inconsistency characteristics of the battery pack. Flexibility and dynamic adaptability: the fuzzy comprehensive evaluation method is adopted, which allows the weight distribution to be adjusted according to actual needs, and has the ability to dynamically adjust and output the evaluation results in real time under complex operating conditions. Intuitiveness and practicality: through the inconsistency level classification (excellent, good, pass, fail), the evaluation results are intuitive, which is convenient for application in battery pack management and optimization strategy formulation. Wide applicability: The method is applicable to a variety of battery pack configurations and operating environments, providing a scientific basis for the design optimization, balanced management and maintenance of lithium battery packs, and has good engineering application prospects.

[0142] In the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", and "fixed" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0143] The standard parts used in the present invention can all be purchased from the market, and special-shaped parts can be customized according to the instructions and the drawings. The specific connection methods of each part adopt conventional means such as mature bolts, rivets, welding, etc. in the prior art. Machinery, parts and equipment all adopt conventional models in the prior art, and the circuit connection adopts the conventional connection method in the prior art, which will not be described in detail here.

[0144] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A lithium battery pack inconsistency assessment method based on fuzzy comprehensive evaluation of model parameters, characterized in that: Includes modeling, parameter identification, and battery pack inconsistency assessment.

2. The lithium battery pack inconsistency assessment method based on fuzzy comprehensive evaluation of model parameters according to claim 1, characterized in that: The method comprises the following steps: Step 1: establishing a first-order RC equivalent circuit model to describe the behavior and characteristics of a lithium battery, which includes an open circuit voltage source Vocv, an ohmic resistor R0 and an RC network; the RC network is composed of a polarization resistor R1 and a polarization capacitor C1; According to the circuit principle, the system model is established as follows: In t =V ocv +V1+R0I (1) Where V1 represents the polarization voltage, which is defined as follows: Step 2: Identify the parameters of the model. The specific steps are as follows: The recursive least squares algorithm with forgetting factor (FFRLS) is used to identify the parameters of the first-order RC equivalent circuit model, so as to obtain the ohmic internal resistance R0, polarization internal resistance R1 and polarization capacitance C1 of the lithium battery; in order to meet the form of the RLS algorithm, the system model needs to be discretized. First, the polarization voltage equation is discretized as follows: Where Δt represents the sampling time; the voltage difference is defined as follows: Since ΔV(k-1)=V t (k-1)-V ocv (k-1), so we can further get: Since Δt is very small, V ocv (k)≈V ocv (k-1), so the above formula can be converted to: definition: The first-order RC equivalent circuit model parameters are solved as follows: Where θ = [θ1, θ2, θ3, θ4]; Then, the FFRLS algorithm is used to perform parameter identification based on the collected current and voltage data of the battery cell. The specific steps are as follows: The input-output model of the system to be identified can be expressed as: V(k)=X T (k)θ+e(k) (9) Where V(K) is the kth sampling value of the battery terminal voltage, and e(k) is the measurement error of the system; The algorithm flow is as follows:

1. Initialize the forgetting factor λ, covariance matrix P, parameter matrix θ, and error matrix e; 2. Calculate the gain:

3. Update the covariance matrix:

4. Calculation error: e(k)=V(k)-X T (k)θ(k−1) (12) 5. Update parameter matrix: θ(k)=θ(k-1)+K(k)e(k) (13) 6. Calculate the model parameters according to formula 8; 7. Step 3: Battery pack inconsistency evaluation, the specific steps are as follows: Step 3-1: Calculate the ohmic internal resistance, polarization internal resistance and polarization capacitance of all battery cells in the battery pack from the collected battery data according to steps 1 and 2; and calculate their range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation and Gini coefficient respectively; the calculation formula of the range coefficient R is as follows: R=X max -X min (14) Where X max Represents the maximum value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack, X min Represents the minimum value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack; The calculation formula of standard deviation coefficient σ is as follows: Among them, X i are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack, and N is the number of battery cells that make up the battery pack; The calculation formula of coefficient of variation CV is as follows: Where σ represents the standard deviation of the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack. It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack; The calculation formula of mean absolute deviation MAD is as follows: Among them, X i are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack; The calculation formula of Gini coefficient G is as follows: Among them, X i and X j are the single cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) of all batteries in the battery pack, It is the average value of all battery cell parameters (ohmic internal resistance, polarization internal resistance and polarization capacitance) in the battery pack; Step 3-2: Establish the inconsistency level of lithium battery pack; set the inconsistency level to "excellent", "good", "pass", "fail" four levels, the inconsistency index includes the five indicators of range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation and Gini coefficient introduced in step 3-1, this range can be flexibly adjusted according to the specific evaluation object and actual needs, as shown in Table 1; the level "excellent" means that the consistency performance is excellent and can be used normally; the level "good" means that the consistency performance level is good and can continue to be used. When it is at the operation and maintenance time node, balanced maintenance can be performed; the level "pass" means that the consistency level of the energy storage battery is general at this time, and balanced maintenance should be applied according to the situation; the level "fail" means that the consistency of the energy storage battery is very poor at this time, and the battery should be shut down immediately for balanced maintenance; Step 3-3: Calculate the inconsistency index of all battery cells in the battery pack using equations (14) to (18) according to the ohmic internal resistance, polarization internal resistance, and polarization capacitance of all battery cells in the battery pack identified in step 2; Step 3-4: Determine the inconsistency evaluation factor set U = {range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, Gini coefficient}, the inconsistency evaluation grade set V = {excellent, good, pass, fail} and the inconsistency evaluation factor weight set A = {w1, w2, w3, w4, w5}, where the weight set can be set according to personal preference, as long as w1+w2+w3+w4+w5=1 is satisfied. Here we set w1=w2=w3=w4=w5=0.25; Step 3-5: Perform single factor evaluation on the range coefficient, standard deviation coefficient, coefficient of variation, and mean absolute deviation coefficient to obtain r i = {r i1 ,r i2 ,r i3 ,r i4 ,r i5 }, indicating the proportion of the five evaluation indicators (range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, Gini coefficient) of all battery cells in the battery pack in the inconsistency evaluation level ("excellent", "good", "pass", "fail") in turn; Step 3-6: Construct a comprehensive evaluation matrix R of battery pack inconsistency based on the five single-factor evaluation sets obtained, as shown in Formula 19, where R is a rows and b columns, a is the number of battery pack inconsistency evaluation indicators, which is set to 5 here (range coefficient, standard deviation coefficient, coefficient of variation, mean absolute deviation coefficient, Gini coefficient), and b represents the number of battery pack inconsistency evaluation levels, which is set to 4 here ("excellent", "good", "pass", "fail"); Step 3-7: First, determine the value of the fourth column in the matrix R. If any item is greater than 0, the battery pack inconsistency evaluation level is "failed". If all items in the fourth column of the matrix R are 0, the matrix evaluation B is calculated based on the weight matrix A and the fuzzy relationship matrix R, that is, B = A × R = (b1, b2, b3, b4), where b1, b2, b3, b4, b5 represent the probabilities of consistency performance being "excellent", "good", "passed", and "failed", respectively; Step 3-8: Calculate the inconsistency degree of the battery pack according to the obtained evaluation matrix B, and the calculation formula is shown in formula (20), where ω1=100, ω2=80, ω3=60, ω4=40; μ=ω1b1+ω2b2+ω3b3+ω4b4 (20) Then, a comprehensive evaluation is made on the battery pack inconsistency to determine the battery pack inconsistency evaluation score (40-100). After obtaining the battery pack inconsistency evaluation score, the battery pack inconsistency evaluation level is determined according to the comparison table in Table 2; for example, if the matrix B = (0.75 0.25 0 0), the inconsistency evaluation score is μ = 0.75 × 100 + 0.25 × 80 + 0 × 60 + 0 × 40 = 95, and the battery pack inconsistency performance level is defined as "excellent".

Images