Battery multi-scale equivalent analysis method and device

Abstract

The present invention discloses a multi-scale equivalent analysis method and apparatus for batteries. The method comprises: constructing a microscale model using the core structure of a power battery as a representative volume unit; constructing a macroscale model based on the specific connection relationship between the representative volume unit and all the core structures; solving the microscale model to obtain the equivalent stiffness tensor and equivalent stress corresponding to the microscale model; and performing a coupled calculation on the macroscale model based on the equivalent stiffness tensor and equivalent stress corresponding to the microscale model, updating the equivalent stiffness tensor and equivalent stress at the integration points in the macroscale model corresponding to the microscale model. This method can effectively improve the computational efficiency of power battery simulation analysis and fully reflect the mechanical properties of the power battery.

Description

Technical Field

[0001] The present invention relates to the technical field of power battery analysis, and in particular to a battery multi-scale equivalent analysis method and device. Background Art

[0002] At present, with the widespread popularity of electric vehicles, more and more people are beginning to pay attention to the structural safety of electric vehicles, and the structural safety of power batteries is the top priority of electric vehicle safety. Power batteries are an important component of electric vehicles. Their structure is complex and there are many components, which brings great difficulty to the simulation analysis of power batteries. The two mainstream analysis methods for power battery simulation analysis are as follows: The first is to construct all the components in the power battery. Due to the complex structure and large number of components of the power battery, the equivalent model constructed is too large and the calculation efficiency is low. The second is to simplify the single cell or battery module into a representative volume unit and model based on the representative volume unit, which helps to improve the calculation efficiency. However, in the process of simplifying the single cell or battery module, many micro-scale performance characteristics will be ignored, and the characteristic points of structural failure may be missed. In addition, the constructed equivalent model cannot reflect the mechanical properties of the power battery. Summary of the Invention

[0003] Embodiments of the present invention provide a multi-scale equivalent analysis method and device for a battery to solve the problem that power battery simulation analysis cannot take into account both computational efficiency and reflection of its mechanical properties.

[0004] The present invention provides a battery multi-scale equivalent analysis method, comprising:

[0005] A micro-scale model is constructed using the core structure in the power battery as a representative volume unit, and a macro-scale model is constructed based on the specific connection relationship between the representative volume unit and all the core structures;

[0006] Solving the mesoscale model to obtain an equivalent stiffness tensor and an equivalent stress corresponding to the mesoscale model;

[0007] Based on the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model, the macro-scale model is coupled to calculate and update the equivalent stiffness tensor and equivalent stress of the integral point in the macro-scale model corresponding to the micro-scale model.

[0008] Preferably, the winding core structure in the power battery is used as a representative volume unit to construct a meso-scale model: the winding core structure in the power battery is used as a representative volume unit, and periodic boundary conditions and vertex constraints are imposed to obtain a meso-scale model.

[0009] Preferably, solving the meso-scale model to obtain the equivalent stiffness tensor and equivalent stress corresponding to the meso-scale model includes:

[0010] Obtaining strain information corresponding to the mesoscale model;

[0011] Based on the strain information corresponding to the meso-scale model, obtaining a stress response function corresponding to the meso-scale model;

[0012] Based on the stress response function corresponding to the meso-scale model, the equivalent stiffness tensor and equivalent stress corresponding to the meso-scale model are obtained.

[0013] Preferably, the coupling calculation of the macroscale model based on the equivalent stiffness tensor and equivalent stress corresponding to the mesoscale model, and updating the equivalent stiffness tensor and equivalent stress of the integration point corresponding to the mesoscale model in the macroscale model, comprises:

[0014] In the current increment, strain information corresponding to the mesoscale model is generated, and the strain information is transmitted to the mesoscale model;

[0015] receiving an equivalent stiffness tensor and an equivalent stress corresponding to the micro-scale model, solving the equivalent stiffness tensor and the equivalent stress using a model convergence criterion, and determining whether a convergence condition corresponding to a current increment is satisfied;

[0016] If the convergence condition corresponding to the current incremental step is satisfied and the current incremental step is not the end incremental step, the next incremental step is updated to the current incremental step, and the steps of generating strain information corresponding to the mesoscale model in the current incremental step are repeated, and the strain information is transmitted to the mesoscale model.

[0017] If the convergence condition corresponding to the current incremental step is not satisfied, the step of generating strain information corresponding to the meso-scale model in the current incremental step and transmitting the strain information to the meso-scale model will be repeatedly executed.

[0018] Preferably, the macro-scale model includes a real macro-model and a false macro-model;

[0019] The coupling calculation of the macroscale model based on the equivalent stiffness tensor and equivalent stress corresponding to the mesoscale model, and updating the equivalent stiffness tensor and equivalent stress of the integration point corresponding to the mesoscale model in the macroscale model, includes:

[0020] In the real macro model, generating a load value of a current incremental step, and applying the load value of the current incremental step to the false macro model;

[0021] In the false macro-model, the equivalent stiffness tensors and equivalent stresses corresponding to all the micro-scale models are received, and the model is solved according to the load value of the current incremental step and the equivalent stiffness tensors and equivalent stresses corresponding to all the micro-scale models to determine whether the equilibrium condition corresponding to the current incremental step is satisfied;

[0022] If the equilibrium condition corresponding to the current incremental step is not satisfied, the false macro-model generates strain information corresponding to all the meso-scale models, and transmits the strain information corresponding to all the meso-scale models to all the meso-scale models;

[0023] If the equilibrium condition corresponding to the current incremental step is satisfied, the false macro-model transmits the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model to the real macro-model;

[0024] In the real macroscopic model, solving the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model to determine whether the convergence condition corresponding to the current increment step is satisfied;

[0025] If the convergence condition corresponding to the current incremental step is satisfied and the current incremental step is not the end incremental step, the next incremental step is updated to the current incremental step, and the load value of the current incremental step is generated in the real macro-model repeatedly, and the load value of the current incremental step is applied to the false macro-model;

[0026] If the convergence condition corresponding to the current incremental step is not satisfied, the steps of generating the load value of the current incremental step in the real macro model and applying the load value of the current incremental step to the false macro model are repeated.

[0027] The present invention provides a battery multi-scale equivalent analysis device, comprising:

[0028] A model building module is used to construct a micro-scale model using the core structure in the power battery as a representative volume unit, and to construct a macro-scale model based on the specific connection relationship between the representative volume unit and all the core structures;

[0029] A microscopic parameter solving module, used to solve the microscopic scale model and obtain the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model;

[0030] A macroscopic coupling calculation module is used to perform coupling calculations on the macroscopic scale model based on the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model, and to update the equivalent stiffness tensor and equivalent stress of the integral points in the macroscopic scale model corresponding to the microscopic scale model.

[0031] Preferably, the model building module is also used to take the core structure in the power battery as a representative volume unit, impose periodic boundary conditions and vertex constraints to constrain it, and obtain a micro-scale model.

[0032] Preferably, the microscopic parameter solving module includes:

[0033] a strain information acquisition unit, configured to acquire strain information corresponding to the mesoscale model;

[0034] a stress function acquisition unit, configured to acquire a stress response function corresponding to the meso-scale model based on strain information corresponding to the meso-scale model;

[0035] A mesoscopic parameter acquisition unit is used to obtain an equivalent stiffness tensor and an equivalent stress corresponding to the mesoscopic scale model based on a stress response function corresponding to the mesoscopic scale model.

[0036] Preferably, the macro coupling calculation module includes:

[0037] a first strain information transmission unit, configured to generate strain information corresponding to the meso-scale model in a current increment, and transmit the strain information to the meso-scale model;

[0038] a first model convergence judgment unit, configured to receive an equivalent stiffness tensor and an equivalent stress corresponding to the micro-scale model, solve the equivalent stiffness tensor and the equivalent stress using a model convergence criterion, and judge whether a convergence condition corresponding to a current incremental step is satisfied;

[0039] a first update processing unit, configured to update the next incremental step to the current incremental step if a convergence condition corresponding to the current incremental step is satisfied and the current incremental step is not an ending incremental step, repeatedly generate strain information corresponding to the meso-scale model in the current incremental step, and transmit the strain information to the meso-scale model;

[0040] The second update processing unit is configured to repeatedly generate strain information corresponding to the meso-scale model in the current incremental step and transmit the strain information to the meso-scale model if the convergence condition corresponding to the current incremental step is not satisfied.

[0041] Preferably, the macro-scale model includes a real macro-model and a false macro-model;

[0042] The macroscopic coupling calculation module includes:

[0043] a load value applying unit, configured to generate a load value of a current incremental step in the real macro model, and apply the load value of the current incremental step to the false macro model;

[0044] a model balance judgment unit, configured to receive, in the false macro-model, the equivalent stiffness tensors and equivalent stresses corresponding to all the micro-scale models, solve the model according to the load value of the current incremental step and the equivalent stiffness tensors and equivalent stresses corresponding to all the micro-scale models, and judge whether the balance condition corresponding to the current incremental step is satisfied;

[0045] a second strain information transmission unit, configured to, if the equilibrium condition corresponding to the current incremental step is not satisfied, generate strain information corresponding to all the meso-scale models by the false macro-model, and transmit the strain information corresponding to all the meso-scale models to all the meso-scale models;

[0046] a microscopic parameter transfer unit, configured to transfer the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model to the real macroscopic model by the false macroscopic model if the equilibrium condition corresponding to the current incremental step is satisfied;

[0047] a second model convergence judgment unit, configured to solve the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model in the real macro-model, and judge whether the convergence condition corresponding to the current incremental step is satisfied;

[0048] a first iterative processing unit, configured to, if a convergence condition corresponding to the current incremental step is satisfied and the current incremental step is not an ending incremental step, update the next incremental step to the current incremental step, repeatedly execute the steps of generating a load value of the current incremental step in the real macro-model, and applying the load value of the current incremental step to the false macro-model;

[0049] The second iterative processing unit is used to repeatedly execute the step of generating a load value of the current incremental step in the real macro model and applying the load value of the current incremental step to the false macro model if the convergence condition corresponding to the current incremental step is not met.

[0050] The above-mentioned multi-scale equivalent analysis method and device for batteries uses the core structure in the power battery as a representative volume unit to construct a micro-scale model and a macro-scale model. This makes the constructed micro-scale model smaller in scale and can reflect more micro-scale mechanical properties, avoiding the omission of characteristic points of structural failure during the analysis process based on the micro-scale model. The equivalent stiffness tensor and equivalent stress obtained by solving the micro-scale model are transferred to the macro-scale model for coupling calculation to update the equivalent stiffness tensor and equivalent stress of the integral point corresponding to the micro-scale model in the macro-scale model. This allows the micro-scale model and the macro-scale model to cooperate to effectively reflect the mechanical properties of the power battery, helping to improve the analysis efficiency of power battery simulation analysis and provide an effective equivalent model for power battery damage and failure analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0052] Figure 1 This is a flow chart of a multi-scale equivalent analysis method for a battery according to an embodiment of the present invention;

[0053] Figure 2 is another flow chart of a multi-scale equivalent analysis method for a battery according to one embodiment of the present invention;

[0054] Figure 3 is another flow chart of a multi-scale equivalent analysis method for a battery according to one embodiment of the present invention;

[0055] Figure 4 is another flow chart of a multi-scale equivalent analysis method for a battery according to one embodiment of the present invention;

[0056] Figure 5 FIG. 1 is a schematic diagram of a multi-scale equivalent analysis device for a battery according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort shall fall within the scope of protection of the present invention.

[0058] The multi-scale equivalent analysis method for batteries provided in an embodiment of the present invention can be applied on a computer device to construct an equivalent model of a power battery so as to simulate and analyze the performance of the power battery based on the constructed equivalent model.

[0059] Generally speaking, a power battery comprises at least one battery module, each of which includes at least one single cell, each of which includes at least one core structure, and each core structure comprises multiple components. This means that a power battery comprises multiple core structures, each of which exhibits periodic repetitive features and specific relationships. Periodic repetitive features refer to the repetitive characteristics found within multiple core structures. Specific relationships refer to the connections between all core structures in the power battery that need to be modeled.

[0060] In one embodiment, if Figure 1 As shown, a battery multi-scale equivalent analysis method is provided. The method is applied to computer equipment as an example for explanation. The battery multi-scale equivalent analysis method includes the following steps:

[0061] S101: Take the core structure in the power battery as the representative volume unit to build a micro-scale model, and build a macro-scale model based on the specific connection relationship between the representative volume unit and all core structures.

[0062] The Representative Volume Element (RVE) is a unit used to construct micro-scale structures and performance. A micro-scale model is a model related to the micro-scale, constructed using the roll-core structure in a power battery as a representative volume unit. A macro-scale model is a model related to the macro-scale, constructed using the representative volume units corresponding to all roll-core structures in a power battery and their specific relationships.

[0063] As an example, since the power battery includes at least one battery module, each battery module includes at least one single cell, each single cell includes at least one core structure, each core structure includes multiple components, and multiple components in the core structure have periodic repetition characteristics, therefore, taking the core structure as a representative volume unit to construct a meso-scale model can effectively improve the efficiency of meso-scale model construction. Compared with the meso-scale model constructed based on components, the model scale is smaller; moreover, compared with the meso-scale model constructed based on battery modules and single cells, the constructed meso-scale model can reflect more meso-scale mechanical properties, avoiding the omission of characteristic points of structural failure in the analysis process based on the meso-scale model.

[0064] As an example, a power battery comprises at least one battery module, each module comprises at least one single cell, and each single cell comprises at least one roll-up structure. This means that a power battery comprises multiple roll-up structures, each of which has specific relationships. Therefore, a macroscale model of the power battery can be constructed based on the representative volume units constructed by the roll-up structures and their specific relationships. The macroscale model can reflect the specific relationships between all roll-up structures. In this example, the representative volume unit constructed by the roll-up core structure is used as an integration point in the macroscale model.

[0065] S102: Solve the micro-scale model to obtain the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model.

[0066] Among them, the equivalent stiffness tensor and equivalent stress are the main factors for analyzing the micro-scale model in the multi-scale analysis process.

[0067] As an example, in the process of multi-scale analysis of a meso-scale model, the general analysis step and the linear perturbation analysis step can be used in combination for analysis, that is, a linear perturbation analysis step is added between the general analysis steps. The linear perturbation analysis step can be executed from time to time during the fully nonlinear analysis, and the linear perturbation analysis step has no effect on the general analysis step. Its step time can be arbitrarily taken as a smaller value and will not be accumulated in the total analysis time. Therefore, by using the general analysis step and the linear perturbation analysis step in combination, meso-scale calculations of the meso-scale model can be realized, and the equivalent stiffness tensor and equivalent stress corresponding to the meso-scale model can be solved quickly and effectively. The equivalent stiffness tensor and equivalent stress are information determined by analyzing the representative volume unit after homogenization.

[0068] S103: Based on the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model, a coupling calculation is performed on the macro-scale model to update the equivalent stiffness tensor and equivalent stress of the integral points in the macro-scale model corresponding to the micro-scale model.

[0069] As an example, the finite element method can be used to achieve coupling between multiscale models—that is, coupling between microscale and macroscale models—providing technical support for multiscale analysis. For material and geometric nonlinear problems, as long as the periodicity assumption holds, linear multiscale analysis methods can be applied to nonlinear multiscale analysis. Because the core structure contains periodic repetitive features, multiple increments can be set when solving the macroscale model, each of which contains multiple iterations for analysis. In this example, multiple incremental steps are executed sequentially. In each incremental step, the strain information of the corresponding integration point in the macro-scale model is iteratively sent from the meso-scale model to the macro-scale model until the equivalent stiffness tensor and equivalent stress returned by the meso-scale model meet the preset convergence conditions. Then, the equivalent stiffness tensor and equivalent stress of the integration point corresponding to the meso-scale model in the macro-scale model are updated, that is, the equivalent stiffness tensor and equivalent stress of the integration point corresponding to each representative volume unit in the macro-scale model are determined, so as to iteratively update the equivalent stiffness tensor and equivalent stress in the meso-scale model using the updated equivalent stiffness tensor and equivalent stress of the integration point corresponding to each representative volume unit.

[0070] In the multi-scale equivalent analysis method for batteries provided in this embodiment, the core structure in the power battery is used as a representative volume unit to construct a micro-scale model and a macro-scale model. This ensures that the constructed micro-scale model is smaller in scale and can reflect more micro-scale mechanical properties, avoiding the omission of characteristic points of structural failure during the analysis process based on the micro-scale model. The equivalent stiffness tensor and equivalent stress obtained by solving the micro-scale model are transferred to the macro-scale model for coupling calculation to update the equivalent stiffness tensor and equivalent stress at the integral points in the macro-scale model corresponding to the micro-scale model. This allows the micro-scale model and the macro-scale model to cooperate to effectively reflect the mechanical properties of the power battery, helping to improve the analysis efficiency of power battery simulation analysis and providing an effective equivalent model for power battery damage and failure analysis.

[0071] In one embodiment, the core structure in the power battery is used as a representative volume unit to construct a micro-scale model: the core structure in the power battery is used as a representative volume unit, and periodic boundary conditions and vertex constraints are imposed to obtain a micro-scale model.

[0072] Among them, periodic boundary conditions (PBCs) are conditions used to achieve stress continuity on the boundaries of adjacent representative volume elements, and vertex constraints are conditions used to constrain the vertices of representative volume elements.

[0073] As an example, the periodically repetitive features of the core structure are modeled in detail to make them have the characteristics of a representative volume unit. In order to make the mechanical properties of the representative volume unit equivalent, the representative volume unit needs to be homogenized and periodic boundary conditions applied to it. For example, if the integral point coordinates of a representative volume unit in the microscopic scale model are 、 and ,in 、 and is the side length of the representative volume element, and under periodic boundary conditions (PBCs) the periodic displacement tensor When applied, the governing equations for the six symmetry planes are as follows (1):

[0074] The above periodic boundary conditions can ensure that stress continuity is achieved simultaneously at the boundaries between adjacent representative volume units.

[0075] As an example, in order to avoid rigid body motion, the vertices of the representative volume element should be constrained. Therefore, vertex constraints need to be applied. For example, a Python script can be used to apply linear multi-point constraint (MPC) technology to apply displacement PBC to the representative volume element to achieve vertex constraint. Its control equation is shown in the following formula (2):

[0076] in, is the correlation coefficient; is the number of terms in the linear multi-point control equation; displacement variables Superscript Denotes degrees of freedom, subscript is the position of the node. In order to facilitate the application of MPCs, each node of the boundary is defined as a node set.

[0077] Furthermore, in order to improve the efficiency of constructing the mesoscale model, after taking the core structure in the power battery as a representative volume unit, it is necessary to perform homogenization calculations on it, and then impose periodic boundary conditions and vertex constraints to obtain the mesoscale model.

[0078] In one embodiment, if Figure 2 As shown, the micro-scale model is solved to obtain the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model, including:

[0079] S201: Obtain strain information corresponding to the mesoscale model.

[0080] S202: Based on the strain information corresponding to the meso-scale model, a stress response function corresponding to the meso-scale model is obtained.

[0081] S203: Based on the stress response function corresponding to the meso-scale model, obtain the equivalent stiffness tensor and equivalent stress corresponding to the meso-scale model.

[0082] Strain information refers to the strain applied during the equivalent analysis of a power battery. The stress response function refers to the stress function used in the equivalent analysis of a power battery based on microscale and macroscale models. The stress function is a function of several arbitrary or special types used to represent stress.

[0083] As an example, in step S201, the concept of thermal strain increment can be used to apply strain to the representative volume unit in the mesoscale model through the user-defined subroutine UEXPAN to obtain the strain information corresponding to the mesoscale model, as shown in the following formula (3):

[0084] in, For strain information; is the coefficient of thermal expansion, which can be determined empirically; is the temperature change, for example, .

[0085] As another example, in step S201, strain information corresponding to each integration point in the macroscale model may be received and determined as the strain information of the mesoscale model corresponding to the integration point. That is, when calculating the macroscale model, strain information of the mesoscale model corresponding to the integration point is generated at any iteration step in each current increment, and the strain information is transferred to the mesoscale model for calculation, so that the strain information required by the mesoscale model is obtained during the solution process.

[0086] As an example, in step S202, the function type of the stress response function is predetermined, and then the stress response function corresponding to the mesoscale model is determined based on the strain information corresponding to the mesoscale model. for When the matrix is Each column in can be determined by the strain information corresponding to the mesoscale model, that is, by the corresponding unit strain, as shown in the following formula (4):

[0087] As an example, in step S203, after obtaining the stress response function of the mesoscale model, it is necessary to establish linear perturbation analysis steps and strain loading conditions that match the number of columns in the matrix corresponding to the stress response function. When the strain loading conditions corresponding to all linear perturbation analysis steps are met, the equivalent stiffness tensor corresponding to the mesoscale model is determined according to formula (5), and the equivalent stress corresponding to the mesoscale model is determined according to formula (6):

[0088] in, is the equivalent stiffness tensor, is the stress response function, is the equivalent stress, is the volume of the representative volume unit, is the total number of integration points corresponding to the representative volume unit; and Points The Jacobian determinant and weight of . Equivalent stress is the stress response function in the mesoscale model The average value of the unit strain in each column.

[0089] Understandably, the equivalent stiffness tensor corresponding to the microscopic scale model is calculated and equivalent stress , is an important consideration for micro-scale calculations during multi-scale analysis. In the calculation, there are two types of analysis steps: general analysis steps and linear perturbation analysis steps. By adding linear perturbation analysis steps between general analysis steps, linear perturbation analysis steps can be performed from time to time during the full nonlinear analysis. The linear perturbation analysis step has no effect on the general analysis step, and the step time (which can be arbitrarily taken to be a very small value) is not accumulated into the total time. Since the situation described above is completely suitable for the requirements of micro-scale calculations, the combination of general analysis steps and linear perturbation analysis steps is used for micro-scale calculations. The nonlinear response of the representative volume element is performed in the general analysis step, while the calculation of the instantaneous equivalent stiffness tensor is performed in the linear perturbation analysis step.

[0090] In one embodiment, if Figure 3 As shown, step S103, i.e., performing coupling calculation on the macro-scale model based on the equivalent stiffness tensor and equivalent stress corresponding to the meso-scale model, and updating the equivalent stiffness tensor and equivalent stress of the integration point corresponding to the meso-scale model in the macro-scale model, includes:

[0091] S301: In the current increment, strain information corresponding to the meso-scale model is generated, and the strain information is transferred to the meso-scale model.

[0092] S302: Receive the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model, solve the equivalent stiffness tensor and equivalent stress using the model convergence criterion, and determine whether the convergence condition corresponding to the current increment step is met.

[0093] S303: If the convergence condition corresponding to the current incremental step is met and the current incremental step is not the end incremental step, the next incremental step is updated to the current incremental step, and the current incremental step is repeatedly executed to generate strain information corresponding to the meso-scale model, and the strain information is passed to the meso-scale model.

[0094] S304: If the convergence condition corresponding to the current incremental step is not met, the current incremental step will be repeatedly executed to generate strain information corresponding to the meso-scale model, and the strain information will be transferred to the meso-scale model.

[0095] The current incremental step refers to the incremental step calculated at the current moment, and each incremental step includes multiple iteration steps.

[0096] As an example, in step S301, during the calculation process of the macro-scale model, the strain increment at the integration point corresponding to the meso-scale model in each iteration step of the current increment is used as the strain information of the meso-scale model, and the strain information is passed to the meso-scale model so as to calculate the strain increment as a boundary condition of the general analysis step of the meso-scale model. Specifically, during the coupled calculation process of the macro-scale model, the strain increment at the integration point corresponding to the meso-scale model in each iteration table of the current increment is used as the strain information of the meso-scale model. The sum is written into an external file as the strain information of the meso-scale model for meso-scale model analysis.

[0097] As an example, in step S302, during the calculation process of the macro-scale model, the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model can be obtained based on the strain information corresponding to the micro-scale model. It is understandable that the correct transmission of effective information between different scales must be considered in the multi-scale analysis process to achieve coupled multi-scale analysis. Therefore, based on the strain information of the macro-scale model, the equivalent stiffness tensor and equivalent stress of the micro-scale model can be updated, and the equivalent stiffness tensor determined as the corresponding integral point in the macro-scale model can be updated. and equivalent stress , and its update expressions are shown in formulas (7) and (8):

[0098] Among them, the upper left corner and is the number of iteration steps, lower left corner and is the number of increments. It is important to note that these variables correspond to converged (equilibrium) mesoscale model solutions, which may or may not represent converged macroscale model results.

[0099] Understandably, in the mesoscale model, the material model of the representative volume unit can be customized in the independent user subroutine UMAT. From the perspective of functional implementation, the following two problems need to be solved: (i) Solving the mesoscale model, that is, determining the equivalent stiffness tensor and equivalent stress corresponding to the mesoscale model based on the obtained strain information corresponding to the mesoscale model, as shown in the technical solution of steps S201-S203. (ii) Evaluation of the stress response function, that is, calculating the equivalent stiffness tensor according to formulas (7) and (8) during the calculation process of the mesoscale model. and equivalent stress , specifically through the general analysis step and six linear perturbation analysis steps. Specifically, the strain information corresponding to the micro-scale model can be read through the user-defined subroutine URDFIL, that is, the strain information of the integral point corresponding to the micro-scale model is transmitted to the user-defined subroutine UEXTERNALDB for calculation, and finally the equivalent stiffness tensor is obtained. and equivalent stress .

[0100] The model convergence criterion is a pre-configured criterion used to evaluate whether the model convergence condition is met. The end increment step refers to the last increment step in multiple increment steps.

[0101] As an example, in step S302, during the calculation process of each iteration step of the current incremental step, the corresponding equivalent stiffness tensor and equivalent stress of the micro-scale model need to be received, and the equivalent stiffness tensor and equivalent stress need to be processed using the pre-configured model convergence criterion to determine whether the convergence conditions corresponding to the current incremental step are met, so as to determine whether the calculation of the current incremental step is completed.

[0102] As an example, in step S303, if the convergence condition corresponding to the current increment is satisfied and the current increment is not the final increment, that is, there is a next increment corresponding to the current increment, then the next increment is updated to the current increment, and the current increment is repeatedly executed to generate strain information corresponding to the mesoscale model and send the strain information to the mesoscale model. It can be understood that when the received equivalent stiffness tensor and equivalent stress corresponding to the mesoscale model satisfy the convergence condition corresponding to the current increment, the calculation of the current increment is considered converged. At this time, if the current increment is the final increment, it indicates that the entire macroscale model has converged, and the multi-scale analysis is completed. If the current increment is not the final increment, it is necessary to update the next increment to the current increment. Then, step S301 is repeated, that is, repeatedly executing the current increment to generate strain information corresponding to the mesoscale model and send the strain information to the mesoscale model until the convergence conditions corresponding to all increments are satisfied, and the macroscale model calculation is completed.

[0103] As an example, in step S304, if the convergence condition corresponding to the current incremental step is not met, it means that the equivalent stiffness tensor and equivalent stress corresponding to the currently received meso-scale model do not meet the convergence condition corresponding to the current incremental step. At this time, it is necessary to repeat the execution in the current incremental step to generate the strain information corresponding to the meso-scale model and send the strain information to the meso-scale model, that is, in the next iteration step in the current incremental step, the strain information corresponding to the meso-scale model is passed to the meso-scale model for the next iterative calculation.

[0104] In the framework of ABAQUS / Standard, the calculation of the model integration points is performed one by one, that is, all the integration points are calculated in sequence, and the calculation of the next integration point can be activated only after all the calculations of the previous integration point are completed. It is worth noting that the most time-consuming calculation process in multi-scale analysis is the calculation of the micro-scale model. Due to the limitation of sequential calculation, the running time of each iteration of the macro-scale model is equal to ,in is the total number of integration points corresponding to all integration points in the macroscale model, is the computation time of the micro-scale model corresponding to each integration point of the macro-scale model. Since the more integration points in the macro-scale model, the more computation time it takes, if multiple micro-scale models can be computed simultaneously, i.e., parallel computing, the computation time of multi-scale analysis can be greatly saved. For example, if The calculation time of the micro-scale model can be reduced to the previous Reduce to .

[0105] In fact, within the sequential calculation framework of ABAQUS / Standard, there is still an opportunity to perform parallel calculations. Due to the independence between the micro-scale model and the macro-scale model, if the strain information of each integration point in the macro-scale model can be provided to all micro-scale models at the same time, the parallel calculations of all micro-scale models can be performed separately. However, since the strain information of the integration points is only allowed to be accessed one by one in ABAQUS, rather than providing the strain information of all integration points immediately after each iterative calculation, a new parallel calculation method needs to be provided to ensure that the strain information of all integration points can be accessed simultaneously in the ABAQUS / Standard framework.

[0106] In one embodiment, the macro-scale model includes a real macro-model and a fake macro-model, such as Figure 4 As shown, step S103, i.e., performing coupling calculation on the macro-scale model based on the equivalent stiffness tensor and equivalent stress corresponding to the meso-scale model, and updating the equivalent stiffness tensor and equivalent stress of the integration point corresponding to the meso-scale model in the macro-scale model, includes:

[0107] S401: In the real macro model, generate the load value of the current incremental step, and apply the load value of the current incremental step to the false macro model.

[0108] S402: In the false macro-model, the equivalent stiffness tensors and equivalent stresses corresponding to all micro-scale models are received, and the model is solved according to the load value of the current incremental step and the equivalent stiffness tensors and equivalent stresses corresponding to all micro-scale models to determine whether the equilibrium condition corresponding to the current incremental step is met.

[0109] S403: If the equilibrium condition corresponding to the current incremental step is not satisfied, the false macro-model generates strain information corresponding to all meso-scale models, and transmits the strain information corresponding to all meso-scale models to all meso-scale models.

[0110] S404: If the equilibrium condition corresponding to the current incremental step is satisfied, the false macro-model transfers the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model to the true macro-model.

[0111] S405: In the real macroscopic model, the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model are solved to determine whether the convergence condition corresponding to the current increment step is met.

[0112] S406: If the convergence condition corresponding to the current incremental step is met and the current incremental step is not the end incremental step, the next incremental step is updated to the current incremental step, and the execution is repeated in the real macro model to generate the load value of the current incremental step, and the load value of the current incremental step is applied to the false macro model.

[0113] S407: If the convergence condition corresponding to the current incremental step is not satisfied, the process is repeated in the real macro model to generate the load value of the current incremental step, and the load value of the current incremental step is applied to the false macro model.

[0114] As an example, in step S401, in the real macro model, the load value of the current increment step needs to be obtained in real time. , and the load value of the current increment Applied to the fake macromodel so that the fake macromodel is based on the load value of the current increment Solve the model.

[0115] As an example, in step S402, the false macro model may receive the load value of the current incremental step sent by the real macro model. , and can receive the equivalent stiffness tensor corresponding to the microscopic scale model and equivalent stress , which receives the equivalent stiffness tensor of the corresponding integration point in the macroscale model and equivalent stress In this example, if the current increment is the first increment, the pre-configured equivalent stiffness tensor can be received. and equivalent stress If the current increment is not the first increment, the equivalent stiffness tensors of all micro-scale models connected to the false macro-model can be received. and equivalent stress .

[0116] Furthermore, in step S402, in the false macro-model, the pre-configured model balance criterion is used to solve the model for the load value of the current incremental step, the equivalent stiffness tensor and the equivalent stress corresponding to all micro-scale models, and determine whether the balance condition corresponding to the current incremental step is met, so as to obtain a judgment result corresponding to whether the balance condition corresponding to the current incremental step is met or not.

[0117] As an example, in step S403, when the false macro model solves the model, if it is determined that the equilibrium condition corresponding to the current increment step is not satisfied, the false macro model needs to generate strain information corresponding to all micro-scale models and transmit the strain information corresponding to all micro-scale models to all micro-scale models. In this example, the false macro model updates the strain increments of all integration points. The sum is written to an external file as the strain information corresponding to the corresponding micro-scale model, so that the micro-scale model can be analyzed based on the strain information. It can be understood that the pseudo macro model updates the strain increments of all integration points. Passed to all mesoscale models so that all mesoscale models can be calculated in parallel, thereby quickly iterating to obtain the equivalent stiffness tensors corresponding to all mesoscale models and equivalent stress , so as to repeat step S402.

[0118] As an example, in step S404, during the model solving process of the false macro model, when it is determined that the equilibrium conditions corresponding to all current incremental steps are met, the false macro model transmits the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model to the real macro model so that the real macro model can make a model convergence judgment.

[0119] As an example, in step S405, in the real macro model, the model convergence criterion is used to solve the equivalent stiffness tensor and the equivalent stress model to determine whether the convergence conditions corresponding to the current incremental step are met to determine whether the calculation of the current incremental step is completed.

[0120] As an example, in step S406, in the real macro-model, if it is determined that the convergence conditions corresponding to the current increment are met and the current increment is not the final increment, that is, there is a next increment corresponding to the current increment, then the next increment is updated to the current increment, and the current increment is repeatedly executed to generate strain information corresponding to the meso-scale model, and the strain information is sent to the meso-scale model. It can be understood that when the equivalent stiffness tensor and equivalent stress corresponding to the received meso-scale model meet the convergence conditions corresponding to the current increment, the calculation of the current increment is considered converged. At this time, if the current increment is the final increment, it means that the entire macro-scale model has converged, and the multi-scale analysis is completed. If the current increment is not the final increment, it is necessary to update the next increment to the current increment. Then, step S401 is repeated, that is, the load value of the current increment is repeatedly generated in the real macro-model and applied to the fake macro-model until the convergence conditions corresponding to all increments are met, and the macro-scale model calculation is completed.

[0121] As an example, in step S407, in the real macro model, if it is determined that the convergence conditions corresponding to the current incremental step are not met, it means that the equivalent stiffness tensor and equivalent stress corresponding to the currently received micro-scale model do not meet the convergence conditions corresponding to the current incremental step. At this time, step S401 needs to be repeated, that is, repeatedly executed in the real macro model to generate the load value of the current incremental step, and apply the load value of the current incremental step to the false macro model for the next iterative calculation.

[0122] In this embodiment, multi-scale calculations are still divided into macro-scale model calculations and micro-scale model calculations, and macro-scale model calculations are divided into real macro-model calculations and false macro-model calculations. In the false macro-model, its main task is to collect the strain information of all integral points in the real macro-model after each iterative calculation. In fact, the calculation of the false macro-model is a repeated calculation relative to the calculation of the real macro-model. In addition to consuming a certain amount of calculation time, the calculation of the false macro-model will not have any task impact on other calculations, but the calculation time of the false macro-model is much smaller than that of the micro-scale model, that is, compared with the calculation time saved in the parallel calculation process of the micro-scale model, the additional calculation time increased by using the false macro-model can be ignored. It is worth noting that the false macro-model only performs calculations for the current incremental step, which can save a lot of calculation time. Once the strain information of all integral points in the false macro-model is collected, the strain information of all integral points can be assigned to all micro-scale models so that all micro-scale models can be called for parallel calculation at the same time. After the parallel calculation of all micro-scale models is completed, the equivalent stiffness tensors of all micro-scale models need to be calculated. and equivalent stress Return to the fake macro model. The fake macro model can solve the model according to the load value of the current increment, the equivalent stiffness tensor and equivalent stress corresponding to all micro-scale models to determine whether the equilibrium condition of the current increment is met; if the equilibrium condition of the current increment is met, the equivalent stiffness tensor of all micro-scale models can be and equivalent stress Passed to the real macro model. The real macro model receives the equivalent stiffness tensor of all micro-scale models corresponding to the current increment. and equivalent stress Afterwards, it is necessary to determine whether the convergence condition corresponding to the current incremental step is met. If the convergence condition corresponding to the current incremental step is met and the current incremental step is not the end incremental step, the next incremental step is updated to the current incremental step for iterative calculation. If the convergence condition corresponding to the current incremental step is not met, the step size of the current incremental step is reduced, the current incremental step is updated, and iterative calculation is performed based on the updated current incremental step until all incremental steps meet the corresponding convergence condition. All the calculations mentioned in this example can be automated using multiple calculation and analysis programs written in Fortran, Python scripts, and DOS batch files.

[0123] As an example, the real macromodel, the fake macromodel, and the mesoscale model can be set up on a supercomputer. Because the supercomputer can access multiple CPUs simultaneously for parallel computing, the multiple CPUs can work together to complete the calculation process of the real macromodel, the fake macromodel, and the mesoscale model. Understandably, the powerful computing processing power of the supercomputer makes it relatively simple for multiple CPUs to perform different computing tasks, effectively improving the efficiency of multi-scale analysis.

[0124] As another example, the real macromodel, the false macromodel, and the mesoscale model can be macroscale models installed on multiple computer devices in the same local area network, specifically including a master computer equipped with the real macromodel and the real macromodel, and multiple slave computers equipped with the mesoscale model. The multiple computer devices are used to collaboratively perform multiscale analysis. Specifically, by obtaining access rights between different computer devices, such as using the open source work PsExec, multiple computer devices in the same local area network can cooperate to perform parallel computing of the mesoscale model. Compared with using supercomputers for parallel computing, this can effectively reduce its cost. During the multiscale analysis process, the master computer can automatically assign the computing tasks of different mesoscale models to different false macromodels through a program written in Fortran and DOS batch files. After completing the calculation, the mesoscale model on each slave computer can return the calculation results to the master computer so that the real macromodel on the master computer can perform model calculations. It is worth noting that when the number of units in the false macromodel is larger, the improved computing efficiency is more significant.

[0125] Understandably, in multi-scale analysis, the macro-scale model is usually considered a homogeneous model, while the meso-scale model is usually considered a non-mean model. Since the calculation process of the homogeneous model is simpler and less time-consuming than the non-uniform model, the calculation time of the macro-scale model is much shorter than that of the meso-scale model. In this example, the calculation of the macro-scale model is divided into a real macro-model and a false macro-model for calculation, so that the false macro-model can collect strain information of all integration points and feed it back to the meso-scale model, so that the meso-scale model can perform parallel calculations based on the strain information of all integration points. By increasing the calculation time of the false macro-model, the calculation time of the meso-scale model is reduced, thereby effectively improving the computational efficiency of the multi-scale analysis and reducing the calculation time.

[0126] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0127] In one embodiment, a battery multi-scale equivalent analysis device is provided, which corresponds one-to-one to the battery multi-scale equivalent analysis method in the above embodiment. Figure 5 As shown, the battery multi-scale equivalent analysis device includes a model building module 501, a microscopic parameter solving module 502, and a macroscopic coupling calculation module 503. The functional modules are described in detail as follows:

[0128] The model construction module 501 is used to construct a micro-scale model by taking the core structure in the power battery as a representative volume unit, and to construct a macro-scale model based on the specific connection relationship between the representative volume unit and all the core structures.

[0129] The microscopic parameter solving module 502 is used to solve the microscopic scale model and obtain the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model.

[0130] The macro coupling calculation module 503 is used to perform coupling calculation on the macro scale model based on the equivalent stiffness tensor and equivalent stress corresponding to the micro scale model, and update the equivalent stiffness tensor and equivalent stress of the integral points in the macro scale model corresponding to the micro scale model.

[0131] In one embodiment, the model building module 501 is further used to take the core structure in the power battery as a representative volume unit, impose periodic boundary conditions and vertex constraints to constrain it, and obtain a micro-scale model.

[0132] In one embodiment, the microscopic parameter solving module 502 includes:

[0133] The strain information acquisition unit is used to obtain the strain information corresponding to the micro-scale model.

[0134] The stress function acquisition unit is used to acquire the stress response function corresponding to the mesoscale model based on the strain information corresponding to the mesoscale model.

[0135] The microscopic parameter acquisition unit is used to obtain the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model based on the stress response function corresponding to the microscopic scale model.

[0136] In one embodiment, the macro coupling calculation module 503 includes:

[0137] The first strain information transmission unit is used to generate strain information corresponding to the meso-scale model in the current incremental step, and transmit the strain information to the meso-scale model.

[0138] The first model convergence judgment unit is used to receive the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model, use the model convergence criterion to solve the equivalent stiffness tensor and equivalent stress, and judge whether the convergence conditions corresponding to the current incremental step are met.

[0139] The first update processing unit is used to update the next incremental step to the current incremental step if the convergence condition corresponding to the current incremental step is met and the current incremental step is not the end incremental step, repeatedly execute in the current incremental step, generate strain information corresponding to the meso-scale model, and transmit the strain information to the meso-scale model.

[0140] The second update processing unit is used to repeatedly execute the current incremental step if the convergence condition corresponding to the current incremental step is not met, generate strain information corresponding to the meso-scale model, and transmit the strain information to the meso-scale model.

[0141] In one embodiment, the macro-scale model includes a real macro-model and a false macro-model.

[0142] Macro coupling calculation module 503 includes:

[0143] The load value applying unit is used to generate the load value of the current incremental step in the real macro model and apply the load value of the current incremental step to the false macro model.

[0144] The model balance judgment unit is used to receive the equivalent stiffness tensors and equivalent stresses corresponding to all micro-scale models in the false macro-model, solve the model according to the load value of the current incremental step, the equivalent stiffness tensors and equivalent stresses corresponding to all micro-scale models, and judge whether the balance conditions corresponding to the current incremental step are met.

[0145] The second strain information transmission unit is used to generate strain information corresponding to all meso-scale models from the false macro-model if the equilibrium condition corresponding to the current incremental step is not met, and transmit the strain information corresponding to all meso-scale models to all meso-scale models.

[0146] The microscopic parameter transfer unit is used to transfer the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model to the real macroscopic model if the equilibrium condition corresponding to the current incremental step is met.

[0147] The second model convergence judgment unit is used to solve the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model in the real macro model to determine whether the convergence conditions corresponding to the current incremental step are met.

[0148] The first iterative processing unit is used to update the next incremental step to the current incremental step if the convergence condition corresponding to the current incremental step is met and the current incremental step is not the end incremental step, repeatedly execute in the real macro model, generate the load value of the current incremental step, and apply the load value of the current incremental step to the false macro model.

[0149] The second iterative processing unit is used to repeatedly execute in the real macro model, generate the load value of the current incremental step, and apply the load value of the current incremental step to the false macro model if the convergence condition corresponding to the current incremental step is not met.

[0150] The specific definition of the battery multi-scale equivalent analysis device can be found in the definition of the battery multi-scale equivalent analysis method above, and will not be repeated here. The various modules in the above-mentioned battery multi-scale equivalent analysis device can be implemented in whole or in part by software, hardware, or a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0151] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0152] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0153] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A multi-scale equivalent analysis method for batteries, characterized in that: include: Taking the core structure in the power battery as a representative volume unit, a mesoscale model is constructed, and based on the specific connection relationship between the representative volume unit and all the core structures, a macroscale model is constructed, wherein the macroscale model includes a real macroscale model and a false macroscale model; Solving the mesoscale model to obtain an equivalent stiffness tensor and an equivalent stress corresponding to the mesoscale model; Based on the equivalent stiffness tensor and equivalent stress corresponding to the meso-scale model, coupling calculation is performed on the macro-scale model to update the equivalent stiffness tensor and equivalent stress of the integration point corresponding to the meso-scale model in the macro-scale model, including: In the real macro model, generating a load value of a current incremental step, and applying the load value of the current incremental step to the false macro model; In the false macro-model, the equivalent stiffness tensors and equivalent stresses corresponding to all the micro-scale models are received, and the model is solved according to the load value of the current incremental step and the equivalent stiffness tensors and equivalent stresses corresponding to all the micro-scale models to determine whether the equilibrium condition corresponding to the current incremental step is satisfied; If the equilibrium condition corresponding to the current incremental step is not satisfied, the false macro-model generates strain information corresponding to all the meso-scale models, and transmits the strain information corresponding to all the meso-scale models to all the meso-scale models; If the equilibrium condition corresponding to the current incremental step is satisfied, the false macro-model transmits the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model to the real macro-model; In the real macroscopic model, solving the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model to determine whether the convergence condition corresponding to the current increment step is satisfied; If the convergence condition corresponding to the current incremental step is satisfied and the current incremental step is not the end incremental step, the next incremental step is updated to the current incremental step, and the load value of the current incremental step is generated in the real macro-model repeatedly, and the load value of the current incremental step is applied to the false macro-model; If the convergence condition corresponding to the current incremental step is not satisfied, the steps of generating the load value of the current incremental step in the real macro model and applying the load value of the current incremental step to the false macro model are repeated.

2. The battery multi-scale equivalent analysis method according to claim 1, characterized in that: The method uses the core structure in the power battery as a representative volume unit to construct a micro-scale model: uses the core structure in the power battery as a representative volume unit, imposes periodic boundary conditions and vertex constraints to constrain it, and obtains a micro-scale model.

3. The battery multi-scale equivalent analysis method according to claim 1, characterized in that: Solving the mesoscale model to obtain an equivalent stiffness tensor and an equivalent stress corresponding to the mesoscale model includes: Obtaining strain information corresponding to the mesoscale model; Based on the strain information corresponding to the meso-scale model, obtaining a stress response function corresponding to the meso-scale model; Based on the stress response function corresponding to the meso-scale model, the equivalent stiffness tensor and equivalent stress corresponding to the meso-scale model are obtained.

4. A battery multi-scale equivalent analysis device, characterized in that: include: a model construction module for constructing a mesoscale model using the roll core structure in the power battery as a representative volume unit, and constructing a macroscale model based on the specific connection relationship between the representative volume unit and all the roll core structures, wherein the macroscale model includes a real macroscale model and a false macroscale model; A microscopic parameter solving module, used to solve the microscopic scale model and obtain the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model; A macroscopic coupling calculation module is used to perform coupling calculation on the macroscopic scale model based on the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model, and update the equivalent stiffness tensor and equivalent stress of the integration point corresponding to the microscopic scale model in the macroscopic scale model. The macroscopic coupling calculation module includes: a load value applying unit, configured to generate a load value of a current incremental step in the real macro model, and apply the load value of the current incremental step to the false macro model; a model balance judgment unit, configured to receive, in the false macro-model, the equivalent stiffness tensors and equivalent stresses corresponding to all the micro-scale models, solve the model according to the load value of the current incremental step and the equivalent stiffness tensors and equivalent stresses corresponding to all the micro-scale models, and judge whether the balance condition corresponding to the current incremental step is satisfied; a second strain information transmission unit, configured to, if the equilibrium condition corresponding to the current incremental step is not satisfied, generate strain information corresponding to all the meso-scale models by the false macro-model, and transmit the strain information corresponding to all the meso-scale models to all the meso-scale models; a microscopic parameter transfer unit, configured to transfer the equivalent stiffness tensor and equivalent stress corresponding to the microscopic scale model to the real macroscopic model by the false macroscopic model if the equilibrium condition corresponding to the current incremental step is satisfied; a second model convergence judgment unit, configured to solve the equivalent stiffness tensor and equivalent stress corresponding to the micro-scale model in the real macro-model, and judge whether the convergence condition corresponding to the current incremental step is satisfied; a first iterative processing unit, configured to, if a convergence condition corresponding to the current incremental step is satisfied and the current incremental step is not an ending incremental step, update the next incremental step to the current incremental step, repeatedly execute the steps of generating a load value of the current incremental step in the real macro-model, and applying the load value of the current incremental step to the false macro-model; The second iterative processing unit is used to repeatedly execute the step of generating a load value of the current incremental step in the real macro model and applying the load value of the current incremental step to the false macro model if the convergence condition corresponding to the current incremental step is not met.

5. The battery multi-scale equivalent analysis device according to claim 4, characterized in that: The model building module is also used to take the core structure in the power battery as a representative volume unit, impose periodic boundary conditions and vertex constraints to constrain it, and obtain a micro-scale model.

6. The battery multi-scale equivalent analysis device according to claim 4, characterized in that: The microscopic parameter solving module includes: a strain information acquisition unit, configured to acquire strain information corresponding to the mesoscale model; a stress function acquisition unit, configured to acquire a stress response function corresponding to the meso-scale model based on strain information corresponding to the meso-scale model; A mesoscopic parameter acquisition unit is used to obtain an equivalent stiffness tensor and an equivalent stress corresponding to the mesoscopic scale model based on a stress response function corresponding to the mesoscopic scale model.

Images