Method and device for calculating PSD curve of battery pack, storage medium and equipment

Abstract

The application discloses a PSD curve calculation method and device of a battery pack, a storage medium and equipment, and belongs to the technical field of automobiles. A frame stiffness parameter of a new vehicle model is acquired; a plurality of regression models previously constructed are acquired, the regression models are obtained by fitting frame stiffness parameters of a plurality of old vehicle models and mode scores of a plurality of vibration modes, the vibration mode is a characteristic vector obtained by performing singular value decomposition on logarithmic domain data obtained by discretizing and logarithmic transforming actual PSD curves of the plurality of old vehicle models, and the mode score is extracted from a decomposition result of the singular value decomposition; the frame stiffness parameter of the new vehicle model is input into the plurality of regression models respectively, and predicted scores of the new vehicle model under the plurality of vibration modes are obtained; an average value sequence of logarithmic domain data of the plurality of old vehicle models is acquired; and a predicted PSD curve of the new vehicle model is calculated according to the predicted scores, the vibration modes and the average value sequence. The application can improve the accuracy of the PSD curve of the new vehicle model.

Description

Technical Field

[0001] This application relates to the field of automotive technology, and in particular to a method, apparatus, storage medium, and device for calculating the PSD curve of a battery pack. Background Technology

[0002] The battery pack is one of the core components of an electric tractor, and its structural reliability directly affects the overall vehicle safety and service life. During vehicle operation, the excitation generated by uneven road surfaces is transmitted to the battery pack through the frame, causing the battery pack to vibrate. Long-term accumulated vibration may lead to structural fatigue, loose connections, or even failure of the battery pack. Therefore, during the development of new models, it is necessary to conduct random vibration simulation analysis and vibration bench tests on the battery pack to verify its vibration intensity and fatigue life.

[0003] In the simulation analysis and bench tests described above, the power spectral density (PSD) curve is the most critical input parameter. The PSD curve describes the distribution characteristics of vibration energy at different frequencies, and its accuracy directly determines the reliability of the analysis and test results.

[0004] In engineering practice, designers often directly use the PSD curves of older models as the PSD curves of new models. However, different models have differences in frame stiffness, chassis structure, suspension parameters, etc., which directly affect vibration transmission characteristics. If the PSD curves of existing older models are directly used, it will lead to a large deviation between the simulation input and the actual situation, thus rendering the simulation analysis results and bench test results worthless. Summary of the Invention

[0005] This application provides a method, apparatus, storage medium, and device for calculating the PSD curve of a battery pack, to solve the problem that directly using the PSD curve of an older vehicle model as the PSD curve of a new vehicle model leads to a large deviation between the simulation input and the actual situation. The technical solution is as follows: According to a first aspect of this application, a method for calculating the PSD curve of a battery pack is provided, the method comprising: Obtain the frame stiffness parameters of the new vehicle model; Multiple pre-constructed regression models are obtained. The regression models are obtained by fitting the frame stiffness parameters of multiple old models and the mode scores of multiple vibration modes. The vibration modes are obtained by discretizing and logarithmically transforming the actual PSD curves of multiple old models, and then performing singular value decomposition on the obtained logarithmic domain data to obtain feature vectors. The mode scores are extracted from the decomposition results of the singular value decomposition. The frame stiffness parameters of the new vehicle model are input into multiple regression models to obtain the predicted scores of the new vehicle model under multiple vibration modes. Obtain the average sequence of logarithmic field data for multiple old car models; The predicted PSD curve of the new vehicle model is calculated based on the predicted score, the vibration mode, and the average value sequence.

[0006] In one possible implementation, the method further includes: Obtain the actual PSD curves of multiple older car models; After discretizing and logarithmically transforming multiple actual PSD curves, singular value decomposition is performed on the obtained logarithmic domain data, and multiple vibration modes are determined based on the decomposition results. For each vibration mode, a mode score for each old model under the vibration mode is generated based on the decomposition results. The mode scores are then fitted with the frame stiffness parameters of multiple old models to obtain the corresponding regression model.

[0007] In one possible implementation, for each vibration mode, a mode score for each older vehicle model under that vibration mode is generated based on the decomposition results. The mode scores are then fitted with the frame stiffness parameters of multiple older vehicle models to obtain a corresponding regression model, including: For each vibration mode, the left singular matrix and singular value matrix in the decomposition result are multiplied to obtain the mode score of each old model in each vibration mode; Create a regression model corresponding to the vibration mode, wherein the regression model contains unknown fitting coefficients; The regression model is calculated by using the pattern score as the output parameter and the frame stiffness parameters of multiple old models as the input parameter.

[0008] In one possible implementation, after discretizing and logarithmically transforming multiple actual PSD curves, singular value decomposition is performed on the obtained logarithmic domain data, and multiple vibration modes are determined based on the decomposition results, including: Each actual PSD curve is discretized into multiple frequency points, and the power spectral density value at each frequency point is logarithmically transformed to obtain logarithmic domain data. Calculate the average value of the logarithmic domain data of all old car models at each frequency point to obtain the average value sequence; Subtract the average value of the corresponding frequency point from the logarithmic field data of each old model at each frequency point, and construct a centered matrix based on the obtained difference; Singular value decomposition is performed on the centered matrix to obtain the decomposition result; The column vectors of the right singular matrix in the decomposition result are identified as multiple vibration modes.

[0009] In one possible implementation, calculating the predicted PSD curve of the new vehicle model based on the predicted score, the vibration mode, and the average value sequence includes: Based on the predicted score, the vibration mode, and the average value sequence, the predicted PSD curve of the new model in the logarithmic domain is calculated. An exponential transformation is performed on the predicted PSD curve in the logarithmic domain to obtain the final predicted PSD curve of the new vehicle model.

[0010] In one possible implementation, calculating the predicted PSD curve of the new vehicle model in the logarithmic domain based on the predicted score, the vibration mode, and the average value sequence includes: For each frequency point, the predicted score of the new model under each vibration mode is multiplied by the value of the corresponding vibration mode at that frequency point and then summed to obtain a weighted sum; The weighted sum is added to the average value of the corresponding frequency point in the average value sequence to obtain the predicted PSD value of the frequency point in the logarithmic domain. The predicted PSD values ​​of all frequency points in the logarithmic domain are used to construct the predicted PSD curve of the new model in the logarithmic domain.

[0011] In one possible implementation, the frame stiffness parameters include Z-direction bending stiffness, Y-direction bending stiffness, and torsional stiffness. Obtaining the frame stiffness parameters of the new vehicle model includes: The Z-direction bending stiffness is calculated based on the concentrated Z-direction force and the maximum Z-direction bending deformation applied to the frame of the new vehicle model. The bending stiffness in the Y direction is calculated based on the concentrated force and maximum bending deformation in the Y direction applied to the frame of the new vehicle model. The torsional stiffness is calculated based on the concentrated Z-axis force and torsional deformation applied to the frame of the new vehicle model.

[0012] According to a second aspect of this application, a PSD curve calculation device for a battery pack is provided, the device comprising: The parameter acquisition module is used to obtain the frame stiffness parameters of the new vehicle model; The model acquisition module is used to acquire multiple pre-built regression models. The regression models are obtained by fitting the frame stiffness parameters of multiple old models and the mode scores of multiple vibration modes. The vibration modes are feature vectors obtained by discretizing and logarithmically transforming the actual PSD curves of multiple old models and then performing singular value decomposition on the obtained logarithmic domain data. The mode scores are extracted from the decomposition results of the singular value decomposition. The score prediction module is used to input the frame stiffness parameters of the new vehicle model into multiple regression models to obtain the predicted scores of the new vehicle model under multiple vibration modes. The sequence acquisition module is used to obtain the average value sequence of logarithmic field data for multiple old car models; The curve calculation module is used to calculate the predicted PSD curve of the new vehicle model based on the predicted score, the vibration mode, and the average value sequence.

[0013] According to a third aspect of this application, a computer-readable storage medium is provided, wherein at least one instruction is stored therein, the at least one instruction being loaded and executed by a processor to implement the PSD curve calculation method for a battery pack as described above.

[0014] According to a fourth aspect of this application, an electronic device is provided, the electronic device including a PSD curve calculation device for the aforementioned battery pack.

[0015] The beneficial effects of the technical solution provided in this application include at least the following: By acquiring actual PSD curves and frame stiffness parameters of multiple older vehicle models, the actual PSD curves were discretized, logarithmically transformed, and decomposed using singular value decomposition to extract multiple vibration modes that characterize the vibration energy distribution. A regression model was then established between the mode scores of these vibration modes and the frame stiffness parameters. This modeling approach considers the influence of frame stiffness on vibration transmission characteristics, making the prediction results more closely match the actual vibration environment of the new vehicle models. This provides more accurate input for subsequent random vibration simulation analysis, vibration fatigue analysis, and vibration bench testing.

[0016] When it is necessary to predict the PSD curve of a new vehicle model, the frame stiffness parameters of the new model can be input into multiple regression models to obtain prediction scores under multiple vibration modes. These scores are then combined with the average sequence of logarithmic domain data from multiple older models to calculate the predicted PSD curve of the new model. This allows for obtaining a relatively accurate PSD curve as simulation input during the design phase of the new model, thereby improving the accuracy of simulation analysis results and bench test results. Attached Figure Description

[0017] 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.

[0018] Figure 1 This is a flowchart of a method for calculating the PSD curve of a battery pack according to an embodiment of this application; Figure 2 This is a flowchart of a method for calculating the PSD curve of a battery pack according to an embodiment of this application; Figure 3 This is a schematic diagram of a concentrated force F in the Z direction applied at the center position of the front and rear axle centers of a vehicle frame, according to an embodiment of this application. Figure 4 This is a schematic diagram of a concentrated force F in the Y direction applied at the center position of the front and rear axle centers of a vehicle frame, according to an embodiment of this application. Figure 5 This is a schematic diagram of applying a concentrated force F in the Z direction at the left front axle and the right rear axle of the vehicle frame according to an embodiment of this application; Figure 6 This is a structural block diagram of a battery pack PSD curve calculation device provided in one embodiment of this application; Figure 7 This is a structural block diagram of an electronic device provided in one embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0020] like Figure 1 The diagram illustrates a flowchart of a method for calculating the PSD curve of a battery pack according to an embodiment of this application. This method can be applied to electronic devices. The method for calculating the PSD curve of a battery pack may include: Step 101: Obtain the frame stiffness parameters of the new model.

[0021] The new model is a tractor vehicle that is still in the design phase, and its battery pack PSD curve needs to be predicted.

[0022] The frame stiffness directly affects the transmission characteristics of road vibrations from the frame to the battery pack. Therefore, it is necessary to calculate the frame stiffness parameters of the new vehicle model. Frame stiffness parameters are indicators of the frame's ability to resist deformation, including Z-direction bending stiffness, Y-direction bending stiffness, and torsional stiffness. In this embodiment, the X-direction is defined as the direction of motion of the tractor, the Y-direction as the left-right direction of the tractor, and the Z-direction as the vertical or gravity direction. Thus, Z-direction bending stiffness represents the vertical bending resistance, Y-direction bending stiffness represents the lateral bending resistance, and torsional stiffness represents the torsional resistance.

[0023] Specifically, during the chassis design phase, the chassis stiffness parameters of a new vehicle model can be calculated using the finite element method.

[0024] Step 102: Obtain multiple pre-built regression models. The regression models are obtained by fitting the frame stiffness parameters of multiple old models and the mode scores of multiple vibration modes. The vibration modes are the feature vectors obtained by discretizing and logarithmically transforming the actual PSD curves of multiple old models and then performing singular value decomposition on the obtained logarithmic domain data. The mode scores are extracted from the decomposition results of singular value decomposition.

[0025] The regression model describes the mapping relationship between frame stiffness parameters and vibration mode scores. Each vibration mode corresponds to a regression model. The vibration mode is an feature vector extracted from the singular value decomposition of actual PSD curves of multiple older vehicle models, representing the distribution shape of vibration energy along the frequency axis. The older vehicle models are tractor models with completed designs and available measured data.

[0026] Actual PSD curves are used to describe the distribution of vibration energy at different frequencies. Actual PSD curves are obtained by conducting road tests on older tractor models on various road surfaces, including twisted roads, Belgian roads, pothole roads, and cobblestone roads, and then converting the acceleration data for each surface type. Each surface type corresponds to a specific actual PSD curve.

[0027] The mode score indicates the level of engagement of an older vehicle model under a given vibration mode.

[0028] The regression model is pre-built, and its construction process is as follows: First, the actual PSD curves of multiple old models are obtained. Each actual PSD curve is discretized into multiple frequency points and then logarithmically transformed to obtain logarithmic domain data. Then, singular value decomposition is performed on the logarithmic domain data, and multiple vibration modes and mode scores of each old model under each vibration mode are extracted from the decomposition results. Finally, for each vibration mode, the frame stiffness parameters of the old model are used as input parameters and the mode scores are used as output parameters to establish a regression model through fitting.

[0029] Step 103: Input the frame stiffness parameters of the new model into multiple regression models to obtain the predicted scores of the new model under multiple vibration modes.

[0030] For each vibration mode, the frame stiffness parameters of the new model are input into the regression model corresponding to that vibration mode, and the output of the regression model is used as the prediction score of the new model under that vibration mode.

[0031] Step 104: Obtain the average value sequence of logarithmic field data for multiple old car models.

[0032] At each frequency point, the arithmetic mean of the logarithmic domain data for all older vehicle models is calculated, resulting in a series of average values. This series of average values ​​represents the average vibration level of multiple older vehicle models on a specific characteristic road surface.

[0033] Step 105: Calculate the predicted PSD curve of the new model based on the predicted score, vibration mode, and average value sequence.

[0034] In summary, the battery pack PSD curve calculation method provided in this application obtains actual PSD curves and frame stiffness parameters from multiple older vehicle models. It then discretizes, logarithmically transforms, and performs singular value decomposition on the actual PSD curves to extract multiple vibration modes that characterize the vibration energy distribution. This allows for the establishment of a regression model between the mode scores of these vibration modes and the frame stiffness parameters. This modeling approach considers the influence of frame stiffness on vibration transmission characteristics, making the prediction results more closely match the actual vibration environment of newer vehicle models. This provides more accurate input for subsequent random vibration simulation analysis, vibration fatigue analysis, and vibration bench testing.

[0035] When it is necessary to predict the PSD curve of a new vehicle model, the frame stiffness parameters of the new model can be input into multiple regression models to obtain prediction scores under multiple vibration modes. These scores are then combined with the average sequence of logarithmic domain data from multiple older models to calculate the predicted PSD curve of the new model. This allows for obtaining a relatively accurate PSD curve as simulation input during the design phase of the new model, thereby improving the accuracy of simulation analysis results and bench test results.

[0036] like Figure 2 The diagram illustrates a flowchart of a method for calculating the PSD curve of a battery pack according to an embodiment of this application. This method can be applied to electronic devices. The method for calculating the PSD curve of the battery pack may include: Step 201: Obtain the frame stiffness parameters of the new model.

[0037] In this embodiment, the frame stiffness parameters include Z-direction bending stiffness, Y-direction bending stiffness, and torsional stiffness. Obtaining the frame stiffness parameters of a new vehicle model can include the following three steps: (1) Calculate the Z-direction bending stiffness based on the Z-direction concentrated force and the maximum Z-direction bending deformation applied to the frame of the new model.

[0038] When simulating the bending stiffness in the Z direction, the fixed frame is located at the center of the front and rear axles. The specific constraints are: left side constraint 123 (constraining translational degrees of freedom in the X, Y, and Z directions), right side constraint 13 (constraining translational degrees of freedom in the X and Z directions), left side constraint 23 (constraining translational degrees of freedom in the Y and Z directions), right side constraint 3 (constraining translational degrees of freedom in the Z direction). A concentrated force F in the Z direction is applied at the center of the front and rear axle axes of the frame. Figure 3 As shown. After simulation, the maximum bending deformation x of the frame in the Z direction can be obtained. Z Define the Z-direction bending stiffness K of the frame. Z =F / x Z .

[0039] (2) Calculate the bending stiffness in the Y direction based on the concentrated force and maximum bending deformation in the Y direction applied to the frame of the new model.

[0040] When simulating the bending stiffness in the Y direction, the fixed frame is located at the center of the front axle and the axial direction of the rear axle. The specific constraints are: left side constraint 123 (constraining translational degrees of freedom in the X, Y, and Z directions), right side constraint 13 (constraining translational degrees of freedom in the X and Z directions), left side constraint 23 (constraining translational degrees of freedom in the Y and Z directions), right side constraint 3 (constraining translational degree of freedom in the Z direction). A concentrated force F in the Y direction is applied at the center of the front and rear axle axes of the frame. Figure 4 As shown. After simulation, the maximum bending deformation x of the frame in the Y direction can be obtained. Y Define the Y-direction bending stiffness K of the frame. Y =F / x Y .

[0041] (3) Calculate the torsional stiffness based on the concentrated Z-direction force and torsional deformation applied to the frame of the new model.

[0042] When simulating the torsional stiffness, the fixed frame is located at the front axle axis and the rear axle axis. The specific constraints are: constraint 123456 on the right side of the front axle (i.e., constraining the translational degrees of freedom in the X, Y, and Z directions and the rotational degrees of freedom about the X, Y, and Z directions); constraint 123456 on the left side of the rear axle (i.e., constraining the translational degrees of freedom in the X, Y, and Z directions and the rotational degrees of freedom about the X, Y, and Z directions); and a concentrated force F in the Z direction is applied at the left front axle axis and the right rear axle axis of the frame. Figure 5 As shown. After simulation, the Z-axis displacement x at the loading position can be obtained. 前Tor and x 后Tor Define the torsional stiffness K of the frame. Tor =2F / (x 前Tor +x后Tor ).

[0043] Step 202: Obtain multiple pre-built regression models. The regression models are obtained by fitting the frame stiffness parameters of multiple old models and the mode scores of multiple vibration modes. The vibration modes are the feature vectors obtained by discretizing and logarithmically transforming the actual PSD curves of multiple old models and then performing singular value decomposition on the obtained logarithmic domain data. The mode scores are extracted from the decomposition results of singular value decomposition.

[0044] The process of building a regression model can include the following steps: (1) Obtain the actual PSD curves of multiple old car models.

[0045] The actual PSD curve is obtained by conducting road tests on older tractor models on roads, collecting acceleration data on characteristic road surfaces such as twisted roads, Belgian roads, pothole roads, and cobblestone roads, and then converting the acceleration for each type of road surface. Each type of road surface corresponds to one actual PSD curve.

[0046] To ensure consistency in the location across different vehicle models, the acceleration sampling point can be defined as a fixed location on the battery pack, such as 200mm from the battery pack mounting point.

[0047] Specifically, for a vehicle i of an older model, its acceleration on a certain characteristic road surface is collected, and this acceleration is converted into an actual PSD curve G. i (f).

[0048] (2) After discretizing and logarithmically transforming multiple actual PSD curves, singular value decomposition is performed on the obtained logarithmic domain data, and multiple vibration modes are determined based on the decomposition results.

[0049] Specifically, after discretizing and logarithmically transforming multiple actual PSD curves, singular value decomposition is performed on the obtained logarithmic domain data. Based on the decomposition results, multiple vibration modes are determined, which may include the following five steps: ① Discretize each actual PSD curve into multiple frequency points, and perform a logarithmic transformation on the power spectral density values ​​at each frequency point to obtain logarithmic domain data.

[0050] In this embodiment, an actual PSD curve can be discretized into M frequency points. The value of M can be set according to actual business needs, for example, 200, but is not limited here.

[0051] The actual PSD curve G of the i-th car i (f) Discretized as [G i (f1), G i (f2), ..., G i(f) M )].

[0052] Since the power spectral density values ​​can differ by several orders of magnitude at different frequencies, it is necessary to take the natural logarithm for each discrete point: Y ij =ln(G i (f) j )+ε). (1) Where i = 1, 2, ..., N, j = 1, 2, ..., M, N is the number of vehicles, and M is the number of frequency points. ε is a very small positive number, used to avoid the logarithm being meaningless.

[0053] ② Calculate the average value of the logarithmic domain data of all old models at each frequency point to obtain the average value sequence.

[0054] For each frequency point j, calculate the average value of the logarithmic domain data of N vehicles to obtain the average value sequence. .

[0055] ③ Subtract the average value of the corresponding frequency point from the logarithmic field data of each old model at each frequency point, and construct a centered matrix based on the obtained difference.

[0056] Centralized matrix (2) Among them, the centralized matrix Z ij It is an N x M matrix.

[0057] ④ Perform singular value decomposition on the centered matrix to obtain the decomposition results.

[0058] Decomposition result Z ij =USV T (3) Where, U∈R N×N Let S be a left singular matrix, S∈R M×M V is a singular value matrix. T ∈R N×M It is a right singular matrix.

[0059] ⑤ The multiple column vectors of the right singular matrix in the decomposition result are identified as multiple vibration modes.

[0060] In the decomposition result, each column of the right singular matrix is ​​a column vector of length M, and each element in the vector corresponds to a discrete frequency point. In this embodiment, the first few column vectors of the right singular matrix are determined as multiple vibration modes, and each column vector represents a typical distribution shape of vibration energy along the frequency axis.

[0061] (3) For each vibration mode, generate the mode score of each old model under the vibration mode based on the decomposition results, and fit the mode score and the frame stiffness parameters of multiple old models to obtain the corresponding regression model.

[0062] Specifically, for each vibration mode, a mode score for each old vehicle model under the vibration mode is generated based on the decomposition results. The mode scores are then fitted with the frame stiffness parameters of multiple old vehicle models to obtain the corresponding regression model, which may include the following three steps: ① For each vibration mode, multiply the left singular matrix and singular value matrix in the decomposition result to obtain the mode score of each old model in each vibration mode.

[0063] For each vibration mode p, its mode score W i,p = (US) i,p (4) ② Create a regression model corresponding to the vibration mode, which includes unknown fitting coefficients.

[0064] Regression model W i,p = (US) i,p =a p +b p ×K Y (i) + c p ×K Z (i) + d p ×K Tor (i). (5) Among them, a p b p c p and d p This represents the unknown fitting coefficient.

[0065] ③ Using the pattern score as the output parameter of the regression model and the frame stiffness parameters of multiple old models as the input parameters of the regression model, calculate the fitting coefficient of the regression model.

[0066] In one example, the least squares method can be used to calculate the fitting coefficient α. p b p c p and d p Then, the known fitting coefficients a p b p c p and d p By substituting the values ​​into the regression model, you can obtain the constructed regression model.

[0067] Step 203: Input the frame stiffness parameters of the new model into multiple regression models to obtain the predicted scores of the new model under multiple vibration modes.

[0068] After completing the chassis design of the new model, the Z-axis bending stiffness K of the new model can be calculated through step 201. Z * , Y-direction bending stiffness K Y * and torsional stiffness K Tor * .

[0069] The predicted score W for the new model p * =a p +b p ×lnK Y * +c p ×lnK Z * + d p ×lnK Tor * (6) Step 204: Obtain the average value sequence of logarithmic field data for multiple old car models.

[0070] The obtained value is the average value sequence. .

[0071] Step 205: Calculate the predicted PSD curve of the new model in the logarithmic domain based on the predicted score, vibration mode, and average value sequence.

[0072] Specifically, based on the predicted score, vibration mode, and average value sequence, the predicted PSD curve of the new model in the logarithmic domain is calculated. This can include: for each frequency point, multiplying the predicted score of the new model in each vibration mode by the corresponding vibration mode value at the frequency point and summing the results to obtain a weighted sum; adding the weighted sum to the average value of the corresponding frequency point in the average value sequence to obtain the predicted PSD value of the frequency point in the logarithmic domain; and assembling the predicted PSD values ​​of all frequency points in the logarithmic domain to form the predicted PSD curve of the new model in the logarithmic domain.

[0073] Predicted PSD curve in the logarithmic field (7) Step 206: Perform an exponential transformation on the predicted PSD curve in the logarithmic field to obtain the final predicted PSD curve for the new model.

[0074] The final predicted PSD curve G(f) i ) = exp(Y j * -ε). (8) In summary, the battery pack PSD curve calculation method provided in this application obtains actual PSD curves and frame stiffness parameters from multiple older vehicle models. It then discretizes, logarithmically transforms, and performs singular value decomposition on the actual PSD curves to extract multiple vibration modes that characterize the vibration energy distribution. This allows for the establishment of a regression model between the mode scores of these vibration modes and the frame stiffness parameters. This modeling approach considers the influence of frame stiffness on vibration transmission characteristics, making the prediction results more closely match the actual vibration environment of newer vehicle models. This provides more accurate input for subsequent random vibration simulation analysis, vibration fatigue analysis, and vibration bench testing.

[0075] When it is necessary to predict the PSD curve of a new vehicle model, the frame stiffness parameters of the new model can be input into multiple regression models to obtain prediction scores under multiple vibration modes. These scores are then combined with the average sequence of logarithmic domain data from multiple older models to calculate the predicted PSD curve of the new model. This allows for obtaining a relatively accurate PSD curve as simulation input during the design phase of the new model, thereby improving the accuracy of simulation analysis results and bench test results.

[0076] like Figure 6 The diagram illustrates a structural block diagram of a PSD curve calculation device for a battery pack according to an embodiment of this application. This PSD curve calculation device can be applied to electronic devices. The device includes: The parameter acquisition module 610 is used to acquire the frame stiffness parameters of the new vehicle model; The model acquisition module 620 is used to acquire multiple pre-built regression models. The regression models are obtained by fitting the frame stiffness parameters of multiple old models and the mode scores of multiple vibration modes. The vibration modes are the feature vectors obtained by discretizing and logarithmically transforming the actual PSD curves of multiple old models and then performing singular value decomposition on the obtained logarithmic domain data. The mode scores are extracted from the decomposition results of singular value decomposition. The score prediction module 630 is used to input the frame stiffness parameters of the new model into multiple regression models to obtain the predicted scores of the new model under multiple vibration modes. The sequence acquisition module 640 is used to acquire the average value sequence of logarithmic field data of multiple old car models; The curve calculation module 650 is used to calculate the predicted PSD curve of the new vehicle model based on the prediction score, vibration mode and average value sequence.

[0077] In an optional embodiment, the device further includes: The curve acquisition module is used to acquire the actual PSD curves of multiple old car models; The mode acquisition module is used to discretize and logarithmically transform multiple actual PSD curves, perform singular value decomposition on the obtained logarithmic domain data, and determine multiple vibration modes based on the decomposition results. The model fitting module is used to generate a mode score for each old model under each vibration mode based on the decomposition results, and to fit the mode score with the frame stiffness parameters of multiple old models to obtain the corresponding regression model.

[0078] In an optional embodiment, the model fitting module is further configured to: For each vibration mode, the left singular matrix and singular value matrix in the decomposition result are multiplied to obtain the mode score of each old model in each vibration mode; Create a regression model corresponding to the vibration mode, which includes unknown fit coefficients; Using pattern scores as the output parameters of the regression model and frame stiffness parameters of multiple older vehicle models as the input parameters, the fitting coefficients of the regression model are calculated.

[0079] In an optional embodiment, the pattern acquisition module is further configured to: Each actual PSD curve is discretized into multiple frequency points, and the power spectral density value at each frequency point is logarithmically transformed to obtain logarithmic domain data. Calculate the average value of the logarithmic domain data of all old car models at each frequency point to obtain the average value sequence; Subtract the average value of the corresponding frequency point from the logarithmic field data of each old model at each frequency point, and construct a centered matrix based on the obtained difference; Singular value decomposition is performed on the centered matrix to obtain the decomposition results; Multiple column vectors of the right singular matrix in the decomposition result are identified as multiple vibration modes.

[0080] In an optional embodiment, the curve calculation module 650 is further configured to: Based on the predicted score, vibration mode, and average value sequence, the predicted PSD curve of the new model in the logarithmic domain is calculated. An exponential transformation is performed on the predicted PSD curve in the logarithmic field to obtain the final predicted PSD curve for the new model.

[0081] In an optional embodiment, the curve calculation module 650 is further configured to: For each frequency point, the predicted score of the new model under each vibration mode is multiplied by the corresponding vibration mode value at the frequency point and then summed to obtain a weighted sum; The weighted sum is added to the average value of the corresponding frequency points in the average value sequence to obtain the predicted PSD value of the frequency points in the logarithmic domain. The predicted PSD values ​​of all frequency points in the logarithmic domain are used to construct the predicted PSD curve of the new model in the logarithmic domain.

[0082] In an optional embodiment, the frame stiffness parameters include Z-direction bending stiffness, Y-direction bending stiffness, and torsional stiffness. The parameter acquisition module 610 is further used for: The Z-direction bending stiffness is calculated based on the concentrated Z-direction force and the maximum Z-direction bending deformation applied to the frame of the new model. The bending stiffness in the Y direction is calculated based on the concentrated force and maximum bending deformation in the Y direction applied to the frame of the new model. The torsional stiffness is calculated based on the concentrated Z-axis force and torsional deformation applied to the frame of the new vehicle model.

[0083] In summary, the battery pack PSD curve calculation device provided in this application obtains actual PSD curves and frame stiffness parameters from multiple older vehicle models. It then discretizes, performs logarithmic transformation, and singular value decomposition on the actual PSD curves to extract multiple vibration modes that characterize the vibration energy distribution. This allows for the establishment of a regression model between the mode scores of these vibration modes and the frame stiffness parameters. This modeling approach considers the influence of frame stiffness on vibration transmission characteristics, making the prediction results more closely match the actual vibration environment of newer vehicle models. This provides more accurate input for subsequent random vibration simulation analysis, vibration fatigue analysis, and vibration bench testing.

[0084] When it is necessary to predict the PSD curve of a new vehicle model, the frame stiffness parameters of the new model can be input into multiple regression models to obtain prediction scores under multiple vibration modes. These scores are then combined with the average sequence of logarithmic domain data from multiple older models to calculate the predicted PSD curve of the new model. This allows for obtaining a relatively accurate PSD curve as simulation input during the design phase of the new model, thereby improving the accuracy of simulation analysis results and bench test results.

[0085] One embodiment of this application provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the PSD curve calculation method for a battery pack as described above.

[0086] Figure 7A schematic block diagram of an example electronic device 700 that can be used to implement embodiments of this application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the application described and / or claimed herein.

[0087] like Figure 7 As shown, the electronic device 700 includes a computing unit 701, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. The RAM 703 may also store various programs and data required for the operation of the electronic device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via a bus 707. An input / output (I / O) interface 705 is also connected to the bus 707.

[0088] Multiple components in electronic device 700 are connected to I / O interface 705, including: input unit 706, such as keyboard, mouse, etc.; output unit 707, such as various types of displays, speakers, etc.; storage unit 708, such as disk, optical disk, etc.; and communication unit 709, such as network card, modem, wireless transceiver, etc. Communication unit 709 allows electronic device 700 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0089] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as data processing methods. For example, in some embodiments, the data processing methods may be implemented as computer software programs tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of the data processing methods described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform data processing methods by any other suitable means (e.g., by means of firmware).

[0090] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0091] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0092] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0093] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0094] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0095] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0096] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0097] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for calculating the PSD curve of a battery pack, characterized in that, The method includes: Obtain the frame stiffness parameters of the new vehicle model; Multiple pre-constructed regression models are obtained. The regression models are obtained by fitting the frame stiffness parameters of multiple old models and the mode scores of multiple vibration modes. The vibration modes are obtained by discretizing and logarithmically transforming the actual PSD curves of multiple old models, and then performing singular value decomposition on the obtained logarithmic domain data to obtain feature vectors. The mode scores are extracted from the decomposition results of the singular value decomposition. The frame stiffness parameters of the new vehicle model are input into multiple regression models to obtain the predicted scores of the new vehicle model under multiple vibration modes. Obtain the average sequence of logarithmic field data for multiple old car models; The predicted PSD curve of the new vehicle model is calculated based on the predicted score, the vibration mode, and the average value sequence.

2. The method for calculating the PSD curve of a battery pack according to claim 1, characterized in that, The method further includes: Obtain the actual PSD curves of multiple older car models; After discretizing and logarithmically transforming multiple actual PSD curves, singular value decomposition is performed on the obtained logarithmic domain data, and multiple vibration modes are determined based on the decomposition results. For each vibration mode, a mode score for each old model under the vibration mode is generated based on the decomposition results. The mode scores are then fitted with the frame stiffness parameters of multiple old models to obtain the corresponding regression model.

3. The method for calculating the PSD curve of a battery pack according to claim 2, characterized in that, For each vibration mode, a mode score for each old vehicle model under that vibration mode is generated based on the decomposition results. The mode scores are then fitted with the frame stiffness parameters of multiple old vehicle models to obtain a corresponding regression model, including: For each vibration mode, the left singular matrix and singular value matrix in the decomposition result are multiplied to obtain the mode score of each old model in each vibration mode; Create a regression model corresponding to the vibration mode, wherein the regression model contains unknown fitting coefficients; The regression model is calculated by using the pattern score as the output parameter and the frame stiffness parameters of multiple old models as the input parameter.

4. The method for calculating the PSD curve of a battery pack according to claim 2, characterized in that, After discretizing and logarithmically transforming multiple actual PSD curves, singular value decomposition is performed on the obtained logarithmic domain data. Based on the decomposition results, multiple vibration modes are determined, including: Each actual PSD curve is discretized into multiple frequency points, and the power spectral density value at each frequency point is logarithmically transformed to obtain logarithmic domain data. Calculate the average value of the logarithmic domain data of all old car models at each frequency point to obtain the average value sequence; Subtract the average value of the corresponding frequency point from the logarithmic field data of each old model at each frequency point, and construct a centered matrix based on the obtained difference; Singular value decomposition is performed on the centered matrix to obtain the decomposition result; The column vectors of the right singular matrix in the decomposition result are identified as multiple vibration modes.

5. The method for calculating the PSD curve of a battery pack according to claim 1, characterized in that, The step of calculating the predicted PSD curve of the new vehicle model based on the predicted score, the vibration mode, and the average value sequence includes: Based on the predicted score, the vibration mode, and the average value sequence, the predicted PSD curve of the new model in the logarithmic domain is calculated. An exponential transformation is performed on the predicted PSD curve in the logarithmic domain to obtain the final predicted PSD curve of the new vehicle model.

6. The method for calculating the PSD curve of a battery pack according to claim 5, characterized in that, The step of calculating the predicted PSD curve of the new vehicle model in the logarithmic domain based on the predicted score, the vibration mode, and the average value sequence includes: For each frequency point, the predicted score of the new model under each vibration mode is multiplied by the value of the corresponding vibration mode at that frequency point and then summed to obtain a weighted sum; The weighted sum is added to the average value of the corresponding frequency point in the average value sequence to obtain the predicted PSD value of the frequency point in the logarithmic domain. The predicted PSD values ​​of all frequency points in the logarithmic domain are used to construct the predicted PSD curve of the new model in the logarithmic domain.

7. The method for calculating the PSD curve of a battery pack according to any one of claims 1 to 6, characterized in that, The frame stiffness parameters include Z-direction bending stiffness, Y-direction bending stiffness, and torsional stiffness. Obtaining the frame stiffness parameters of the new vehicle model includes: The Z-direction bending stiffness is calculated based on the concentrated Z-direction force and the maximum Z-direction bending deformation applied to the frame of the new vehicle model. The bending stiffness in the Y direction is calculated based on the concentrated force and maximum bending deformation in the Y direction applied to the frame of the new vehicle model. The torsional stiffness is calculated based on the concentrated Z-axis force and torsional deformation applied to the frame of the new vehicle model.

8. A device for calculating the PSD curve of a battery pack, characterized in that, The device includes: The parameter acquisition module is used to obtain the frame stiffness parameters of the new vehicle model; The model acquisition module is used to acquire multiple pre-built regression models. The regression models are obtained by fitting the frame stiffness parameters of multiple old models and the mode scores of multiple vibration modes. The vibration modes are feature vectors obtained by discretizing and logarithmically transforming the actual PSD curves of multiple old models and then performing singular value decomposition on the obtained logarithmic domain data. The mode scores are extracted from the decomposition results of the singular value decomposition. The score prediction module is used to input the frame stiffness parameters of the new vehicle model into multiple regression models to obtain the predicted scores of the new vehicle model under multiple vibration modes. The sequence acquisition module is used to obtain the average value sequence of logarithmic field data for multiple old car models; The curve calculation module is used to calculate the predicted PSD curve of the new vehicle model based on the predicted score, the vibration mode, and the average value sequence.

9. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, which is loaded and executed by a processor to implement the PSD curve calculation method for the battery pack as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, The electronic device includes: a PSD curve calculation device for the battery pack as described in claim 8.

Images