Lithium battery surface average temperature prediction method and system

Abstract

The invention discloses a lithium battery surface average temperature prediction method and system, and relates to the technical field of battery thermal management. The lithium battery surface average temperature prediction method comprises the following steps: S1, establishing a lithium battery one-dimensional thermal model; s2, performing meshless discretization solution on the one-dimensional thermal model of the lithium battery by adopting a radial basis function collocation method; s3, based on the lithium battery temperature rise experiment data, dividing the residual electric quantity into continuous intervals, and optimizing the time-varying equivalent heat conductivity coefficient and the time-varying equivalent internal heat source parameters in a segmented manner; s4, based on a generalized regression neural network, establishing a mapping relationship among the discharge rate, the residual electric quantity and the optimized parameters in the step S3; and S5, in combination with the mapping relationship in the step S4 and the discretization solution in the step S2, outputting a predicted value of the surface average temperature of the lithium battery in real time. The calculation efficiency of temperature prediction is improved through a radial basis function collocation method, and the generalization ability and accuracy of the model are improved by combining parameters of a generalized regression neural network optimization model.

Description

Technical Field

[0001] This application relates to the field of battery thermal management technology, and in particular to a method and system for predicting the average surface temperature of a lithium battery. Background Technology

[0002] As an important energy source for new energy vehicles, mobile devices and other fields, the safety and performance of lithium batteries are highly dependent on the operating temperature.

[0003] However, traditional mesh-based methods such as the finite difference method and the finite element method consume a lot of computational resources and have a slow solution speed when dealing with high-dimensional complex modeling. Summary of the Invention

[0004] To address the aforementioned issues, this application provides a method for predicting the average surface temperature of lithium batteries. It improves the computational efficiency of temperature prediction by using the radial basis function collocation method and enhances the generalization ability and accuracy of the model by combining it with a generalized regression neural network to optimize the model's parameters. Correspondingly, it provides a prediction system capable of predicting the average surface temperature of lithium batteries under different conditions.

[0005] The first technical solution adopted in this application is: providing a method for predicting the average surface temperature of a lithium battery, including the following steps:

[0006] S1: Establish a one-dimensional thermal model of a lithium battery, which includes time-varying equivalent thermal conductivity, time-varying equivalent heat capacity and time-varying equivalent internal heat source determined by the discharge rate and remaining charge.

[0007] S2: The radial basis function collocation method is used to perform meshless discretization of the one-dimensional thermal model of lithium battery;

[0008] S3: Based on lithium battery temperature rise experimental data, the remaining power is divided into continuous intervals, and the time-varying equivalent thermal conductivity and time-varying equivalent internal heat source parameters are optimized in segments.

[0009] S4: Establish the mapping relationship between discharge rate, remaining power and optimization parameters in step S3 based on generalized regression neural network;

[0010] S5: Combining the mapping relationship in step S4 and the discretization solution in step S2, the predicted value of the average surface temperature of the lithium battery is output in real time.

[0011] In an optional embodiment, step S1 includes:

[0012] The spatial computational domain of the one-dimensional thermal model is the longitudinal axis direction of the lithium battery;

[0013] The upper and lower surfaces of the lithium battery are thermally isolated to achieve thermally adiabatic boundary conditions.

[0014] In an optional embodiment, step S2 includes:

[0015] Discretize the time derivative term;

[0016] Choose the radial basis function form and configure the computation nodes in the computational domain;

[0017] Temperature distribution is solved by approximating the temperature field using radial basis functions.

[0018] In an optional embodiment, the computing nodes include internal location points and boundary location points distributed along the longitudinal direction of the lithium battery, and the temperature prediction values ​​are generated only based on the temperature values ​​at the internal location points and boundary location points.

[0019] In an optional embodiment, the segmentation optimization in step S3 includes:

[0020] Within the defined remaining power range, a joint optimization objective is set to maximize the correlation coefficient between experimental temperature data and model predictions and minimize the maximum relative error.

[0021] The parameter values ​​that satisfy the joint optimization objective are searched within the preset parameter range and used as the optimization results of the time-varying equivalent thermal conductivity and the time-varying equivalent internal heat source.

[0022] In an optional embodiment, in step S4, the discharge rate and remaining charge are used as input variables, and the time-varying equivalent thermal conductivity and time-varying equivalent internal heat source values ​​are obtained based on the generalized regression neural network mapping to construct the one-dimensional thermal model.

[0023] The second technical solution adopted in this application is: providing a lithium battery surface average temperature prediction system, including:

[0024] A prediction unit that predicts the average surface temperature of a lithium battery based on the method described in any of the preceding claims;

[0025] An alarm unit is connected to the prediction unit to receive the predicted temperature and trigger a thermal runaway early warning command based on the predicted temperature.

[0026] The third technical solution adopted in this application is: providing an electronic device, the electronic device comprising: a memory and a processor coupled to each other, the processor being used to execute program instructions stored in the memory to implement the method as described in any of the preceding claims.

[0027] The fourth technical solution adopted in this application is: providing a computer-readable storage medium that stores program data, the program data being executable by a processor to implement the method described in any of the preceding claims.

[0028] The fifth technical solution adopted in this application is: providing a lithium battery state monitoring device, comprising:

[0029] A temperature sensor array is uniformly arranged along the longitudinal direction of the lithium battery to collect surface temperature data.

[0030] The processing unit is configured to perform the method described in any of the preceding items and output a predicted average surface temperature and a thermal risk level.

[0031] Due to the adoption of the above technical solution, this application has at least one of the following beneficial effects compared with the prior art:

[0032] 1. The radial basis function collocation method is a meshless method, which avoids the complex meshing and large matrix operations of traditional meshing methods, and greatly improves the efficiency of numerical computation.

[0033] 2. By establishing a mapping relationship between discharge rate, remaining power and optimization parameters based on a generalized regression neural network, parameter values ​​under any given conditions can be obtained quickly and accurately without repeating time-consuming experiments or optimization processes.

[0034] 3. By dividing the remaining power into intervals, we can assume that the time-varying equivalent thermal conductivity and the time-varying equivalent internal heat source are constant within each interval, thereby simplifying the optimization problem.

[0035] 4. Parameters are optimized in segments using temperature rise experimental data to ensure that model parameters are closer to the actual physical process and improve prediction accuracy. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] in:

[0038] Figure 1 This is a flowchart illustrating a method for predicting the average surface temperature of a lithium battery according to an embodiment of this application.

[0039] Figure 2 A schematic diagram showing the distribution of absolute error in the temperature field solved by analytical solutions and radial basis function collocation methods at different times;

[0040] Figure 3 A schematic diagram comparing the average temperature change on the surface of a lithium battery with the actual temperature obtained from a one-dimensional equivalent thermal model.

[0041] Figure 4This is a schematic diagram comparing the average temperature change on the surface of a lithium battery with the actual temperature.

[0042] Figure 5 A schematic diagram of the framework of a lithium battery surface average temperature prediction system provided in an embodiment of this application;

[0043] Figure 6 This is a schematic diagram of the structure of a computer device according to an embodiment of this application;

[0044] Figure 7 This is a schematic diagram of the structure of an embodiment of the computer-readable storage medium of this application;

[0045] Figure 8 This is a schematic diagram of the framework of a lithium battery state monitoring device provided in an embodiment of this application. Detailed Implementation

[0046] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It is understood that the specific embodiments described herein are only for explaining this application and not for limiting it. Furthermore, it should be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all structures. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0047] The terms "first," "second," etc., used in this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0048] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0049] Currently, lithium batteries are widely used in new energy vehicles, mobile devices, and other fields. Temperature is a key factor affecting the safety and performance of lithium batteries. Due to the high cost of temperature monitoring, traditional methods are difficult to monitor each individual battery cell in real time. Therefore, constructing a thermal model based on temperature-time change curves combined with parameters such as battery voltage and SOC (state of charge) has become an important monitoring method. Most existing thermal models are based on mesh methods, such as the finite difference method and the finite element method. However, these methods have low computational efficiency and high computational resource requirements. Especially in practical engineering, when multi-dimensional complex modeling is required, the issues of computational efficiency and accuracy become more prominent. In view of this, this application provides a method for predicting the average surface temperature of lithium batteries. It improves the computational efficiency of temperature prediction by using the radial basis function collocation method and combines it with a generalized regression neural network to optimize the model parameters, thereby improving the model's generalization ability and accuracy.

[0050] Example 1

[0051] like Figure 1 As shown, Figure 1 A flowchart illustrating a method for predicting the average surface temperature of a lithium battery according to an embodiment of this application includes the following steps:

[0052] S1: Establish a one-dimensional thermal model of the lithium battery. The spatial computational domain of the one-dimensional thermal model is the longitudinal axis of the lithium battery. During use, the battery mainly experiences temperature gradient changes along its length, and the temperature changes in the longitudinal direction better reflect the internal heat distribution of the battery. By establishing a one-dimensional thermal model, the mathematical description of the heat conduction problem can be simplified. Compared to two-dimensional or three-dimensional models, the one-dimensional model requires less data processing, thereby reducing the demand for computational resources and accelerating the solution speed.

[0053] In this embodiment, the lithium battery is an 18650 lithium battery; it should be noted that this application does not limit the specific model of the lithium battery; in other embodiments, other models of lithium batteries may be selected, and no limitation is made in this regard.

[0054] The upper and lower surfaces of the lithium battery are thermally isolated to achieve adiabatic boundary conditions. In actual modeling, these surfaces of the lithium battery do not exchange heat with the outside world. The adiabatic boundary avoids complex boundary condition settings, which not only simplifies numerical simulation but also reduces errors caused by boundary effects.

[0055] The one-dimensional thermal model includes time-varying equivalent thermal conductivity, time-varying equivalent heat capacity, and time-varying equivalent internal heat source, all determined by the discharge rate and remaining charge. The construction principle of the one-dimensional thermal model is described in detail below:

[0056] The differential equation for heat conduction in one dimension with an internal heat source is shown in the form of equation (1) below:

[0057]

[0058] Wherein, φ(t) is a time-varying, spatially uniformly distributed internal heat source; using this one-dimensional thermal conductivity differential equation with internal heat source and combining it with the data on the average temperature change of the discharge surface under multiple rate conditions of lithium battery, a one-dimensional thermal model is attempted to be constructed to approximate the change of the average temperature of the lithium battery surface; x is the spatial coordinate of the lithium battery.

[0059] The solid and liquid materials inside a lithium battery are distributed layer by layer, and the thermal conductivity of each layer varies with temperature within a certain range. Therefore, the longitudinal and circumferential thermal conductivity of a lithium battery exhibits spatiotemporal non-uniformity. In the one-dimensional equivalent thermal model of a lithium battery, T represents the average surface temperature of the lithium battery; λ represents the equivalent longitudinal thermal conductivity and is time-dependent, becoming... Time-varying equivalent thermal conductivity; ρ and c represent the time-dependent average density and specific heat capacity of the lithium battery, respectively, becoming ρ(t) and c(t); ρ(t)c(t) represents the time-varying equivalent heat capacity; φ(t) represents the one-dimensional time-varying equivalent internal heat source, becoming Therefore, equation (1) becomes equation (2):

[0060]

[0061] Furthermore, since the current discharge time t of a lithium battery can be calculated from the discharge rate C and the remaining charge during the discharge process, equation (2) can be expressed as equation (3). In equation (3), ρ(C,SOC), c(C,SOC), and These are the parameters of the one-dimensional equivalent thermal model.

[0062]

[0063] For ease of subsequent calculations, after rearranging ρ(C,SOC)c(C,SOC), equation (3) can be expressed as equation (4):

[0064]

[0065] in,

[0066] S2: The radial basis function collocation method is used to perform meshless discretization and solution of the one-dimensional thermal model of lithium battery. The radial basis function collocation method for solving the one-dimensional equivalent thermal model is combined with a reduced-order model to achieve data-physics-driven construction of the one-dimensional thermal model of lithium battery. The working principle of the radial basis function collocation method is described in detail below:

[0067] The governing equations, boundary conditions, and initial conditions of the equivalent thermal model are shown in equations (5)-(7):

[0068]

[0069] (x = 0, x = battery height)(6)

[0070] T(x,t0) = Initial surface temperature of the lithium battery (t0 = 0)

[0071] The process of solving the governing equation shown in equation (5) using the radial basis function collocation method is as follows:

[0072] The time derivative term of the control equation is discretized by differential processing to obtain equation (8). In equation (8), T(x,t1) represents the temperature value at point x at the next time moment, and T(x,t0) represents the temperature value at point x at the current time moment.

[0073]

[0074] Equation (8) can be rearranged into equation (9), as shown below:

[0075]

[0076] The radial basis function form is selected and computational nodes are configured within the computational domain; in this embodiment, the following is selected: As a radial basis function, where r j =|xx j |,x j is the coordinate value of the selected base point, and c is the shape parameter; N1 points are selected as internal base points within the calculation region.

[0077] Using radial basis function approximation Where a j (t) represents the coefficients to be determined, and N is the total number of base points;

[0078] We can then obtain:

[0079]

[0080] Substituting equation (10) into equation (9) yields equation (12), and rearranging yields equation (13); substituting equation (11) into the boundary condition equation (6) yields equation (14);

[0081]

[0082] x respectively j Substituting equations (13) and (14) yields five sets of linear equations. Solving these equations using the least squares method yields the coefficient 'a' at the current time. j The numerical solution for any point at time t can be obtained simply by substituting the coordinates of that point into equation (15):

[0083]

[0084] The temperature distribution is solved by approximating the temperature field using radial basis functions. The calculation nodes include internal and boundary points distributed along the longitudinal direction of the lithium battery. The temperature prediction value is generated only based on the temperature values ​​at the internal and boundary points. The partial differential equation is transformed into basic multiplication operations. Each iteration does not need to calculate the values ​​of all points, but only the values ​​of the internal and boundary points, thereby improving the efficiency of numerical calculation.

[0085] To verify the accuracy of the collocation method for solving heat conduction problems, a two-dimensional transient heat diffusion problem without internal heat sources is described as follows, where Lx = Ly = 1, α = 0.1, and f(x,y) = 5:

[0086] Finite region: 0≤x≤Lx, 0≤y≤Ly

[0087] Thermal conductivity differential equation:

[0088] Initial condition: T(x,y,0)=f(x,y)

[0089] Boundary conditions: T(0,y,t)=0, T(Lx,y,t)=0, T(x,0,t)=0, T(x,Ly,t)=0

[0090] The analytical solution to this problem obtained by the method of separation of variables is shown in equation (16), where B mn The form is shown in equation (17):

[0091]

[0092] The temperature field distribution at different times (t = 0.1s, 0.5s, 1s) was solved using Equation (16) and the radial basis function collocation method, respectively. The resulting absolute error distribution is as follows: Figure 2 As shown, Figure 2 A schematic diagram illustrating the absolute error distribution of the temperature field solved using the analytical solution and the radial basis function collocation method at different times; from Figure 2 It can be seen that the maximum absolute error of the two methods is only about 0.04, indicating that the radial basis function collocation method has high accuracy in solving heat transfer problems.

[0093] S3: Based on lithium battery temperature rise experimental data, the remaining charge is divided into continuous intervals, and the time-varying equivalent thermal conductivity and time-varying equivalent internal heat source parameters are optimized in segments; the segmented optimization includes:

[0094] Within the defined remaining power range, a joint optimization objective is set to maximize the correlation coefficient between experimental temperature data and model predictions and minimize the maximum relative error.

[0095] The parameter values ​​that satisfy the joint optimization objective are searched within the preset parameter range and used as the optimization results of the time-varying equivalent thermal conductivity and the time-varying equivalent internal heat source.

[0096] The following is a detailed description with reference to a specific embodiment:

[0097] The surface temperature change of the lithium battery under different discharge rates was divided into 10 intervals according to SOC (remaining charge) = 100%–90%, 90%–80%, ..., 10%–0%. Within each interval, Γ(C,SOC) and Θ(C,SOC) were approximated as constants. The maximum relative error E between the temperature-time change curve obtained from the one-dimensional equivalent thermal model of each interval and the experimentally measured true temperature change was calculated. max Using the correlation R as the optimization objective, the interval optimal values ​​of Γ(C,SOC) and Θ(C,SOC) are obtained by using the radial basis function collocation method in different intervals, as described in detail below:

[0098] Set the optimization range:

[0099] For Γ(C,SOC) and Θ(C,SOC), the optimization interval is defined as [Γ]. min , Γ max ] and [Θ min ,Θ max ], and divide these intervals into n equal sub-intervals:

[0100] Γ i =Γ min +i·ΔΓ,

[0101] Θ j =Θ min +j·ΔΘ,

[0102] Substitute into the model to calculate the temperature change curve:

[0103] Combine (Γ) i ,Θ j Substituting into a one-dimensional equivalent thermal model, the temperature change curves T at different SOC intervals are calculated. model (t,SOC):

[0104] T model (t,SOC)=f(Γ i ,Θ j (21)

[0105] Comparison with experimental data:

[0106]

[0107] E max=max|T model,k T exp,k | (22)

[0108] By iterating through all (Γ) i ,Θ j ) combinations, calculate R and E for each combination. max Choose the option that maximizes R and E max The smallest combination is taken as the SOC interval (Γ) i ,Θ j The optimal solution.

[0109] The optimized values ​​of Γ and Θ under the 2C discharge rate condition are shown in Table 1 below:

[0110]

[0111] Substituting the piecewise obtained Γ and Θ values ​​into equations (12)-(15), the average temperature values ​​at points x=5, 15, 30, 45, and 60 at 1800s are calculated using the collocation method, and compared with the actual experimental results, for example... Figure 3 As shown, Figure 3 This is a schematic diagram comparing the average temperature change on the surface of a lithium battery obtained from a one-dimensional equivalent thermal model with the actual temperature. Figure 3 It can be seen that the values ​​of Γ and Θ obtained by the combined optimization method can be substituted into the equivalent thermal model to represent the temperature change of the battery surface well.

[0112] S4: Establish the mapping relationship between discharge rate, remaining charge and optimization parameters in step S3 based on generalized regression neural network; use discharge rate and remaining charge as input variables, and obtain the time-varying equivalent thermal conductivity and time-varying equivalent internal heat source values ​​for constructing a one-dimensional thermal model based on generalized regression neural network mapping.

[0113] S5: Combining the mapping relationship from step S4 and the discretization solution from step S2, the predicted average surface temperature of the lithium battery is output in real time. A generalized regression neural network (GRN) is a neural network based on radial basis function networks used for data regression analysis. To improve the generalization ability of the constructed one-dimensional equivalent thermal model of the lithium battery, lithium battery temperature rise experiments are conducted under different remaining charge (SOC) and discharge rate (C) conditions. Multiple sets of Γ(C,SOC) and Θ(C,SOC) values ​​are obtained, and the GRN learns the relationship between SOC, C, and Γ(C,SOC) and Θ(C,SOC). Given the remaining charge (SOC) and discharge rate (C), the corresponding Γ(C,SOC) and Θ(C,SOC) values ​​can be quickly calculated. Then, using the calculated Γ(C,SOC) and Θ(C,SOC) values ​​in conjunction with the one-dimensional equivalent thermal model of the lithium battery, the predicted lithium battery temperature is quickly calculated.

[0114] Experiments were designed to verify the improvement of the generalization ability of the generalized regression neural network on the one-dimensional equivalent thermal model of lithium batteries and the accuracy of temperature calculation of the one-dimensional equivalent thermal model. Using a lithium battery surface temperature rise test bench, the average surface temperature rise curves of lithium batteries under varying discharge rates, as shown in Table 2, were obtained.

[0115]

[0116] Simultaneously, a generalized regression neural network was used to learn the relationship between SOC and Γ(C,SOC) and Θ(C,SOC) under 1C, 2C, and 3C discharge rate conditions. The changes in Γ and Θ under these conditions were then predicted using the discharge rate and remaining charge information shown in Table 2. These values ​​were then substituted into a one-dimensional equivalent thermal model and solved using a specific method to obtain the average surface temperature change curve of the lithium battery under these operating conditions.

[0117] like Figure 4 As shown, Figure 4 This is a schematic diagram comparing the average temperature change on the surface of a lithium battery with the actual temperature; it can be seen that the temperature curve obtained by the one-dimensional equivalent thermal model can accurately reflect the average temperature change curve on the surface of the lithium battery.

[0118] Example 2

[0119] This application also provides a lithium battery surface average temperature prediction system, such as... Figure 5 As shown, Figure 5 A schematic diagram of the framework of a lithium battery surface average temperature prediction system provided in an embodiment of this application includes:

[0120] The prediction unit predicts the average surface temperature of the lithium battery based on the method of Example 1;

[0121] An alarm unit is connected to a prediction unit to receive the predicted temperature and trigger a thermal runaway early warning command based on the predicted temperature.

[0122] Example 3

[0123] This application provides a computer device; please refer to [link / reference]. Figure 6 , Figure 6 This is a schematic diagram of the structure of a computer device according to an embodiment of the present application. The computer device includes a memory and a processor, wherein the memory and the processor are coupled to each other. The memory stores program data, and the processor is used to execute the program data to implement any step of the lithium battery surface average temperature prediction method of Embodiment 1.

[0124] In this embodiment, the processor may also be referred to as a CPU (Central Processing Unit). The processor may be an integrated circuit chip with signal processing capabilities. The processor may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor may be a microprocessor or any conventional processor.

[0125] Example 4

[0126] The method of Embodiment 1 can be implemented in the form of a computer program; therefore, this application proposes a computer-readable storage medium. Please refer to [link to relevant documentation]. Figure 7 , Figure 7 This is a schematic diagram of the structure of an embodiment of the computer-readable storage medium of this application. The computer-readable storage medium stores program data that can be executed by a processor to implement any step of the average surface temperature prediction of the lithium battery in Embodiment 1 above.

[0127] In this embodiment, the computer-readable storage medium can be a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or a medium that can store program data. Alternatively, it can be a server that stores the program data. The server can send the stored program data to other devices for execution, or it can run the stored program data itself.

[0128] Example 5

[0129] This application also provides a lithium battery state monitoring device, such as... Figure 8 As shown, Figure 8 A schematic diagram of the framework of a lithium battery state monitoring device provided in an embodiment of this application includes:

[0130] A temperature sensor array is uniformly arranged along the longitudinal direction of the lithium battery to collect surface temperature data.

[0131] The processing unit is configured to perform any step of the lithium battery surface average temperature prediction in Example 1, and output the surface average temperature prediction value and thermal risk level.

[0132] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0133] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0134] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0135] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for predicting the average surface temperature of a lithium battery, characterized in that, Includes the following steps: S1: Establish a one-dimensional thermal model of a lithium battery, which includes time-varying equivalent thermal conductivity, time-varying equivalent heat capacity and time-varying equivalent internal heat source determined by the discharge rate and remaining charge. S2: The radial basis function collocation method is used to perform meshless discretization of the one-dimensional thermal model of lithium battery; S3: Based on lithium battery temperature rise experimental data, the remaining power is divided into continuous intervals, and the time-varying equivalent thermal conductivity and time-varying equivalent internal heat source parameters are optimized in segments. S4: Establish the mapping relationship between discharge rate, remaining power and optimization parameters in step S3 based on generalized regression neural network; S5: Combining the mapping relationship in step S4 and the discretization solution in step S2, the predicted value of the average surface temperature of the lithium battery is output in real time.

2. The method for predicting the average surface temperature of a lithium battery according to claim 1, characterized in that, Step S1 includes: The spatial computational domain of the one-dimensional thermal model is the longitudinal axis direction of the lithium battery; The upper and lower surfaces of the lithium battery are thermally isolated to achieve thermally adiabatic boundary conditions.

3. The method for predicting the average surface temperature of a lithium battery according to claim 1, characterized in that, Step S2 includes: Discretize the time derivative term; Choose the radial basis function form and configure the computation nodes in the computational domain; Temperature distribution is solved by approximating the temperature field using radial basis functions.

4. The method for predicting the average surface temperature of a lithium battery according to claim 3, characterized in that, The computing nodes include internal location points and boundary location points distributed along the longitudinal direction of the lithium battery, and the temperature prediction value is generated only based on the temperature values ​​at the internal location points and boundary location points.

5. The method for predicting the average surface temperature of a lithium battery according to claim 1, characterized in that, Step S3, the segmented optimization, includes: Within the defined remaining power range, a joint optimization objective is set to maximize the correlation coefficient between experimental temperature data and model predictions and minimize the maximum relative error. The parameter values ​​that satisfy the joint optimization objective are searched within the preset parameter range and used as the optimization results of the time-varying equivalent thermal conductivity and the time-varying equivalent internal heat source.

6. The method for predicting the average surface temperature of a lithium battery according to claim 1, characterized in that, In step S4, the discharge rate and remaining charge are used as input variables, and the time-varying equivalent thermal conductivity and time-varying equivalent internal heat source values ​​are obtained based on the generalized regression neural network mapping to construct the one-dimensional thermal model.

7. A lithium battery surface average temperature prediction system, characterized in that, include: The prediction unit predicts the average surface temperature of the lithium battery based on the method described in any one of claims 1-6; An alarm unit is connected to the prediction unit to receive the predicted temperature and trigger a thermal runaway early warning command based on the predicted temperature.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor coupled to each other, the processor being configured to execute program instructions stored in the memory to implement the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program data that can be executed by a processor to implement the method as described in any one of claims 1-6.

10. A lithium battery state monitoring device, characterized in that, include: A temperature sensor array is uniformly arranged along the longitudinal direction of the lithium battery to collect surface temperature data. The processing unit is configured to perform the method of any one of claims 1-6 and output a predicted average surface temperature and a thermal risk level.

Images