A Joint Estimation Method of Lithium-ion Battery Internal Resistance and State of Charge Based on Fuzzy Strategy-Guided PI Observer

Abstract

This invention relates to a method for jointly estimating the internal resistance and state of charge (SOC) of a lithium-ion battery based on a fuzzy strategy-guided PI observer, belonging to the field of battery technology. The method includes the following steps: S1: Designing a suitable extended PI observer to transform the internal resistance of the lithium-ion battery into an augmented state, and jointly estimating the internal resistance and SOC; S2: Developing an intelligent strategy based on information fusion and fuzzy logic guidance. Based on the extended PI observer designed in step S1, using the concept of low-density outlier detection (LOD), combined with knowledge-based fuzzy logic, to detect model uncertainties in real time; S3: According to custom rules, changing the control logic to adjust the internal polarization of the battery, achieving timely compensation for modeling defects, and ensuring accurate SOC estimation.

Description

Technical Field

[0001] This invention belongs to the field of battery technology and relates to a method for jointly estimating the internal resistance and SOC of a lithium-ion battery based on a fuzzy strategy-guided PI observer. Background Technology

[0002] The State of Charge (SOC) of a battery is one of the most critical states in a Battery Management System (BMS). Most management and control decisions are based on the SOC value. Inaccurate SOC estimation can easily lead to overcharging or over-discharging, decreased battery balancing capability, capacity reduction, or even permanent damage. Therefore, accurate SOC estimation is crucial for the safe, reliable, and long-life operation of the battery system.

[0003] SOC estimation is mainly categorized into direct calculation methods based on ampere-hour integrals, methods based on electrochemical models, methods based on equivalent circuit models, and data-driven methods. Direct calculation methods suffer from uncertainties in parameter measurements during practical applications, leading to limited accuracy. Electrochemical models suffer from complexity, overfitting, and high computational costs. Data-driven methods require large amounts of high-quality training data and are prone to overfitting. Therefore, only estimation methods based on equivalent circuit models are practically feasible and have significant application potential due to their lower complexity, better accuracy, and robustness. However, equivalent circuit models are highly sensitive to battery parameters, leading to the development of joint estimation methods. Current joint estimation methods primarily combine recursive least squares (RLS) and Kalman filters and their variants for parameter identification and collaborative SOC estimation. These methods still have limitations, such as the assumption of unrealistic Gaussian noise and poor robustness to random scenarios. Furthermore, due to the low accuracy of battery equivalent circuit models, they cannot accurately reflect the true electrochemical response under high-dynamic operating conditions, causing these model defects to be incorrectly estimated as part of the internal resistance, thus affecting the accuracy of battery SOC estimation. In existing research, the uncertainties caused by different parameters are treated equally, without focusing on the physical meaning of each battery parameter, and the specific impact of each parameter on output and accuracy is largely ignored. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a joint estimation method for the internal resistance and state of charge (SOC) of lithium-ion batteries based on a fuzzy strategy-guided PI observer. By utilizing an extended PI observer, the internal resistance is transformed into an augmented state. Based on low-density outlier detection (LOD) and fuzzy logic for real-time detection and compensation of model defects, an adaptive control mechanism that effectively ensures model uncertainty awareness is achieved, enabling high-precision and robust estimation of internal resistance and SOC.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for jointly estimating the internal resistance and state of charge (SOC) of a lithium-ion battery based on a fuzzy strategy-guided PI observer includes the following steps:

[0007] S1: Design a suitable extended PI observer to convert the internal resistance of the lithium-ion battery into an augmented state and perform joint estimation of the internal resistance and SOC.

[0008] S2: Develop an intelligent strategy based on information fusion and fuzzy logic guidance. Based on the extended P observer designed in step S1, use the concept of low-density outlier detection (LOD) and combine it with knowledge-based fuzzy logic to detect the uncertainty of the model in real time.

[0009] S3: Based on custom rules, change the control logic, adjust the internal polarization of the battery, realize timely compensation for modeling defects, and ensure accurate SOC estimation.

[0010] Furthermore, the extended PI observer selected in step S1 is:

[0011]

[0012] Among them, K p and K i These are the proportional gain matrix and the integral gain matrix, S e K is the error signal fi Second integral.

[0013] Furthermore, step S2 specifically includes the following steps:

[0014] S21: For each measured data value, calculate the relative outlier of the selected variable. When the value is close to or greater than 1, it indicates that the fluctuation is large.

[0015] S22: Input the relative outlier value into the fuzzy logic system, and use the fuzzy logic system to divide the relative outlier value into three levels: low, medium, and high; in addition, the input describing the current state is also fed back into the fuzzy logic system;

[0016] S23: The fuzzy logic system provides decisions, which are divided into normal, normal compensation and model uncertainty scenarios.

[0017] Furthermore, in step S22, the fuzzy logic process is divided into three steps: fuzzification, rule formulation, and defuzzification. The fuzzy logic model defines several rules, and the fuzzy rules are in the form of IF_THEN conditions.

[0018] Furthermore, in step S2, the extended PI observer is described using the following expression:

[0019]

[0020] in,

[0021] x eo =[V1V2 soc R in ] T

[0022]

[0023] Among them, K peo K ieo These are the proportional and integral gain matrix functions, respectively.

[0024] Furthermore, in step S3, after confirming the model uncertainty, the control logic is changed according to the custom rules to adjust the internal polarization of the battery to mitigate the impact of model defects, until the system returns to normal and the accumulated error returns to its previous value. The specific approach to correcting modeling defects is as follows:

[0025] Whenever the model-based output e k When fluctuations exceed the threshold, observe the score and... Icr k and dVoc k-1 Look for similar trends in the scores; after confirming the model uncertainty, adjust the internal polarization of the battery until the system returns to normal and the accumulated error returns to its previous value.

[0026] The beneficial effects of this invention are as follows:

[0027] (1) The proposed method takes into account the modeling uncertainty and can reflect the true electrochemical response of the battery under high dynamic conditions. The proposed joint estimation method of battery internal resistance and SOC has higher accuracy and robustness, and is computationally efficient, and has great potential in practical applications.

[0028] (2) The proposed method combines the intelligent strategy of information fusion and fuzzy logic to detect model defects in real time with the extended PI observer, which can detect and compensate for model defects in real time, and can clearly explain the physical meaning of each battery parameter and clarify the specific impact of each parameter on the output and accuracy.

[0029] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0030] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0031] Figure 1This is a schematic diagram illustrating the principle of joint estimation of battery internal resistance and SOC in this invention.

[0032] Figure 2 This is the overall flowchart of the present invention;

[0033] Figure 3 This is a schematic diagram of the equivalent circuit model of a second-order battery in this invention;

[0034] Figure 4 This is a detailed flowchart of step S1 in an embodiment of the present invention;

[0035] Figure 5 This is a detailed flowchart of step S12 in an embodiment of the present invention;

[0036] Figure 6 This is a detailed flowchart of step S3 in an embodiment of the present invention;

[0037] Figure 7 These are the parameters of the fuzzy logic model in step S3 of this invention example. Detailed Implementation

[0038] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0039] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0040] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0041] Please see Figure 1 and Figure 2A method for jointly estimating the internal resistance and state of charge (SOC) of a lithium-ion battery based on a fuzzy strategy-guided PI observer includes the following steps:

[0042] S1: Parameter sensitivity analysis was performed using the second-order RC equivalent circuit model ECM and the OCV-SOC mapping relationship to study the changes in the estimation performance of the battery model under different parameters, revealing that the battery internal resistance is a key parameter affecting the accuracy of lithium-ion state estimation.

[0043] like Figure 3 In step S1, the battery model is a second-order RC ECM. This model has an open-circuit voltage (OCV) source related to the state of charge (SOC). Using Kirchhoff's principle, the dynamics of the charge transfer region can be expressed as:

[0044]

[0045] Where I is the current, R1 and R2 are the polarization resistances, C1 and C2 are the polarization capacitances, and V1 represents the voltage drop generated by the charge transfer polarization resistance R1 and the double-layer capacitance C1. Similarly, the kinetic equation for the diffusion process is:

[0046]

[0047] V2 represents the diffusion process simulated by the diffusion capacitance C2 and the concentration polarization resistance R2. For a given time interval t, the change in the battery's state of charge is equal to the total amount of charge released or charged by the battery during that time interval. Therefore, the battery's state of charge can be expressed as:

[0048]

[0049] Wherein, η and C t These represent coulombic efficiency and total battery capacity, respectively. According to Figure 1 The battery terminal voltage is:

[0050] V o =V oc (soc)+V1+V2+IR in

[0051] Among them, V o R is the terminal voltage. in V represents the internal resistance of the battery. oc (soc) is an open-circuit voltage source. Therefore, the battery ECM system can be represented as a set of the following state equations:

[0052]

[0053] The above expression is converted to the following form:

[0054]

[0055] in,

[0056]

[0057] u = I, x = [V1V2 soc] T D = R in

[0058] h(x) = V oc (soc)+V1+V2

[0059] The internal state of the LIBs system described above can be estimated using the following PI observer:

[0060]

[0061] Among them, K p and K i These are the proportional gain matrix and the integral gain matrix, S e K is the error signal fi Second integral.

[0062] Please see Figure 4 The steps for parameter sensitivity analysis using a second-order RC ECM are as follows:

[0063] S11: Before the experiment begins, all batteries are screened and their capacity is tested to identify those with a capacity fluctuation of less than 3%.

[0064] S12: Define the OCV-SOC relationship mapping. For example... Figure 5 First, the LIB batteries were charged at 0.25C using a constant current constant voltage (CCCV) method until the charging current was less than 0.01C. Then, these fully charged batteries were placed under a discharge current of C / 30 until their V... min (2.75V); After the battery has been left to stand for 2 hours, charge it to V using the same small current C / 30. max (4.35V). Because the extremely low discharge current minimizes battery polarization and hysteresis, the terminal voltage V during charging is... o,ch It can be represented as:

[0065] V o,ch =V oc (soc)+V in +I(R1+R2)

[0066] Similarly, the terminal voltage V during the discharge process o,dis It can be represented as:

[0067] V o,dis =V oc (soc)-V in-I(R1+R2)

[0068] The OCV of the LIB battery can be obtained by averaging the charging and discharging curves.

[0069] V oc (soc)=(V o,ch +V o,dis ) / 2

[0070] After obtaining the OCV for each point, the relationship between SOC and OCV between any two consecutive data points is as follows:

[0071] V oc (soc)=m i .soc+c i

[0072] in,

[0073]

[0074] After obtaining the OCV-SOV relationship, a series of hybrid pulse power characteristic (HPPC) tests were performed on the battery. The main steps are as follows: (1) Fully charge the battery using the CCCV method at 0.25C until the charging current is less than 0.01C; (2) Let it stand for two hours; (3) Discharge the battery at 0.1C for 5% + let it stand for 2 hours until V min (4) Charge at 0.1C to 5% + let stand for 2 hours until V max The measured battery data is processed using the Particle Swarm Optimization (PSO) algorithm to identify battery parameters.

[0075] S13: The measured battery data and parameter sensitivity are provided as time-series inputs to the PI observer. The adverse effects of battery parameter sensitivity are evaluated using metrics such as maximum absolute error (MaxAE), root mean square error (RMSE), and mean absolute error (MAE). The expression is as follows:

[0076]

[0077] Where, x oi and These represent the measured state value and the estimated state value, respectively.

[0078] S2: For the uncertain key parameter of internal resistance, a suitable extended PI observer is selected to transform the battery's internal resistance into an augmented state, and the internal resistance and SOC are jointly estimated; the selected PI observer is:

[0079]

[0080] Among them, K p and K iThese are the proportional gain matrix and the integral gain matrix, S e K is the error signal fi Second integral.

[0081] S3: Develop an intelligent strategy for real-time detection of model defects based on information fusion and fuzzy logic, and combine it with an extended PI observer, using an improved LOD concept, and knowledge-based fuzzy logic to detect model uncertainties in real time; then, according to custom rules, change the control logic to adjust the internal polarization of the battery to achieve timely compensation for modeling defects and ensure the estimation accuracy of state and parameters.

[0082] Please see Figure 6 and Figure 7 The steps for effectively detecting model uncertainty using intelligent strategies based on information fusion and fuzzy logic are as follows:

[0083] S31: For each measured data value, calculate the relative outlier of the selected variable. A value close to or greater than 1 indicates significant fluctuation. The formula for calculating the relative outlier is as follows:

[0084]

[0085] S32: The relative outlier values ​​are input into the fuzzy logic system, which then classifies them into three levels: low, medium, and high. Furthermore, inputs describing the current state (i.e., whether the system has returned to normal, and whether the accumulated error has returned to its previous value) are also fed back into the fuzzy system. The fuzzy logic process can be divided into three steps: fuzzification, rule formulation, and defuzzification. The fuzzy logic system of this invention defines a total of 27 rules, which are in the form of IF_THEN conditions and have been sufficiently verified through experiments.

[0086] S33: The fuzzy logic system provides decisions, which are divided into normal, normal compensation and model uncertainty scenarios;

[0087] S34: Correct for model defects. Whenever the model-based output e... k When fluctuations exceed the threshold, observe the score and... Icr k and dVoc k-1 By searching for similar trends in the scores and confirming model uncertainty, the proposed method is used to adjust the internal polarization of the battery to mitigate the impact of model defects, thereby maintaining the system's accuracy until the system recovers and the accumulated error returns to its previous value. The extended PI observer can be described by the following expression:

[0088]

[0089] in,

[0090] xeo =[V1 V2 soc R in ] T

[0091]

[0092]

[0093]

[0094] Among them, K peo K ieo These are the proportional and integral gain matrix functions, respectively. Furthermore, the output of this design... It is not a linear function of u.

[0095] S4: Verify the observability of the proposed method and experimentally verify its applicability under various stochastic conditions.

[0096] The method for verifying observability in step S4 is as follows: For nonlinear systems, the observability of the system is determined using the Lie derivative theorem. According to the Lie derivative, the battery system can be expressed as:

[0097]

[0098] If Li's gradient matrix is ​​full rank or has independent rows, then in any It can be observed at this location.

[0099]

[0100] in,

[0101]

[0102] according to

[0103] x eo =[V l ,V2,socR in ] T

[0104] u = I, y = V o

[0105] A eo x eo =[-V l / R l C l V2 / R2C2 0 0] T

[0106] B eo =[1 / C l 1 / C2ηi / C t 0] T

[0107] h(x eo ,u)=V oc (soc)+V l +V2+R in u

[0108] The gradient of the Li derivative is calculated as follows:

[0109] dh eo =[1 1dV oc / dsoc I]

[0110]

[0111] The observability matrix is ​​then:

[0112]

[0113] When d k V oc / dsoc k When I ≠ 0 and I ≠ 0, the above matrix is ​​full rank. Since OCV is a complex nonlinear function of battery SOC, the first condition always remains unchanged. However, the unobservability when I = 0 can be physically understood as the enhanced state being multiplied by zero, and its effect is not even reflected in the output voltage. Therefore, it is necessary to set [K] when I = 0. p4eo ;K i4eo The value is set to zero to avoid misleading results.

[0114] In the above embodiments, the reference to "this embodiment" in the specification indicates that a specific feature, structure, or characteristic described in connection with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple appearances of "this embodiment" do not necessarily refer to the same embodiment.

[0115] In the above embodiments, although the invention has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory structures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed. The embodiments of the invention are intended to cover all such substitutions, modifications, and variations falling within the broad scope of the appended claims.

[0116] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the methods in this embodiment.

[0117] This embodiment also provides an electronic terminal, including: a processor and a memory;

[0118] The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to cause the terminal to perform any of the methods in this embodiment.

[0119] As will be understood by those skilled in the art, the computer-readable storage medium described in this embodiment allows for the implementation of all or part of the steps in the above method embodiments by computer program-related hardware. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0120] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication between them. The memory is used to store computer programs, the communication interface is used to perform communication, and the processor and the transceiver are used to run the computer programs, so that the electronic terminal performs the steps of the above method.

[0121] In this embodiment, the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0122] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0123] This invention can be used in a wide range of general-purpose or special-purpose computing system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0124] This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for jointly estimating the internal resistance and state of charge (SOC) of a lithium-ion battery based on a fuzzy strategy-guided PI observer, characterized in that: Includes the following steps: S1: Design a suitable extended PI observer to convert the internal resistance of the lithium-ion battery into an augmented state and jointly estimate the internal resistance and SOC; the designed extended PI observer is as follows: in, and These are the proportional gain matrix and the integral gain matrix, respectively. It is error accumulation; S2: Develop an intelligent strategy guided by information fusion and fuzzy logic. Based on the extended PI observer designed in step S1, use the concept of low-density outlier detection (LOD) and combine it with knowledge-based fuzzy logic to detect model uncertainty in real time. Specifically, this includes the following steps: S21: For each measured data value, calculate the relative outlier of the selected variable. When the value is close to or greater than 1, it indicates that the fluctuation is large. S22: Input the relative outlier value into the fuzzy logic system, and use the fuzzy logic system to divide the relative outlier value into three levels: low, medium, and high; in addition, the input describing the current state is also fed back into the fuzzy logic system; S23: The fuzzy logic system provides decisions, which are divided into normal, normal compensation and model uncertainty scenarios; The extended PI observer is described using the following expression: in, in Indicates the internal resistance of charge transfer polarization and polarization capacitor The resulting voltage drop; Indicates the polarization capacitor and concentration polarization internal resistance Simulated diffusion process; Indicates the battery's state of charge; Indicates the battery's internal resistance; in, , For polarization internal resistance, , Polarizing capacitor; , These are the proportional and integral gain matrix functions, respectively. S3: Based on custom rules, change the control logic, adjust the internal polarization of the battery, realize timely compensation for modeling defects, and ensure accurate SOC estimation.

2. The method for jointly estimating the internal resistance and SOC of a lithium-ion battery based on a fuzzy strategy-guided PI observer according to claim 1, characterized in that: In step S22, the fuzzy logic process is divided into three steps: fuzzification, rule formulation, and defuzzification. The fuzzy logic model defines several rules, and the fuzzy rules are in the form of IF_THEN conditions.

Images