A safety design method for a new energy vehicle battery pack and a battery pack thereof

Abstract

The present disclosure relates to a safety design method for a new energy vehicle battery pack and the battery pack thereof. The method comprises the following steps: establishing an initial finite element model of the battery pack; determining the design variables, optimization objectives, and constraints of the battery pack, wherein the optimization objectives are maximizing the first-order natural frequency, minimizing the mass, and minimizing the extrusion deformation, thereby obtaining a multi-objective optimization design function; obtaining sample points for the experimental design of the design variables, and fitting the sample points and multiple target parameters using the response surface methodology to obtain an approximate model; obtaining Pareto solutions for multiple structural objectives through multi-objective optimization design, obtaining weight values ​​for the multiple target parameters based on the best-worst method-entropy weight method-game theory, and then obtaining an optimal solution for the optimized design of the battery pack structure; inputting the optimal solution into the initial finite element model to obtain an optimized finite element model of the battery pack, thereby ultimately achieving a lightweight design of the new energy vehicle battery pack. The optimized model obtained in the present disclosure achieves weight reduction without compromising performance.

Description

Technical Field

[0001] The present disclosure relates to the field of new energy vehicle structures, and in particular to a safety design method for a new energy vehicle battery pack and a battery pack thereof. Background Art

[0002] With environmental pollution and energy shortages becoming increasingly severe, the research and development of new energy vehicles has become a key focus of the automotive industry. Power batteries, as one of its key technologies, directly impact a vehicle's range and driving safety. The battery pack, which houses the power battery modules, is a critical safety component of electric vehicles. Therefore, research on the safety performance of power battery packs is a top priority for the passive safety of new energy vehicles. Their mechanical properties simultaneously impact the safety and reliability of electrical, thermal, and mechanical systems. During driving, impacts and excitations from the road and other external forces are transmitted to the battery pack via the vehicle body and floor system, simultaneously bearing the risk of external impact and extrusion deformation. Furthermore, existing technologies lack a multi-objective design optimization method for new energy vehicle battery packs. Summary of the Invention

[0003] The present disclosure provides a safety design method for a new energy vehicle battery pack and a battery pack thereof, which can solve the problem of the lack of multi-objective design optimization of new energy vehicle battery packs in the prior art and achieve the most satisfactory optimization design results. To achieve the above-mentioned purpose, the present disclosure provides a safety design method for a new energy vehicle battery pack, comprising the following steps:

[0004] S1. Establish an initial finite element model of the battery pack;

[0005] S2. Simulating the stress conditions of the battery pack under sudden braking on a bumpy road, sharp turning on a bumpy road, and reverse braking on a bumpy road based on the initial finite element model, as well as the dynamic performance of the first six modes under constraint conditions;

[0006] S3. Perform a simulated extrusion test on the battery pack defined by the initial finite element model. Based on the simulated extrusion test results, partially replace the materials of the battery pack, wherein the upper top plate is made of carbon fiber composite material, the middle enclosure is made of high-strength steel material, and the lower bottom plate is made of aluminum alloy material;

[0007] S4. Using the upper top plate thickness, middle panel thickness, lower bottom plate thickness, upper top plate material, middle panel material, and lower bottom plate material in the battery pack structural dimension parameters as parameter design variables to be optimized, and setting initial values ​​for the above parameter design variables to be optimized respectively;

[0008] S5. Determine the optimization objectives and constraints for the battery pack. The optimization objectives are maximizing the first-order natural frequency, minimizing the mass, and minimizing the extrusion deformation. The constraints are that the maximum deformation under sudden braking, sharp turns, and reverse braking on bumpy roads is no more than 2 mm. Obtain a multi-objective optimization design function.

[0009] S6. Calculating multiple performance target parameters in the multi-objective optimization design function based on the initial finite element model data with boundary conditions set in S4 and S5, obtaining initial model test design sample points through the initial design variables, and then fitting the sample points and target parameters using the response surface method to obtain an approximate model;

[0010] S7. Use the NSGA-II algorithm in HyperStudy to obtain the multi-objective Pareto solution set, and then obtain the weight values ​​of the multiple objective parameters based on the best-worst solution-entropy weight method-game theory, and then obtain the optimal solution after the optimized design of the battery pack structure. Input the optimal solution into the initial finite element model to obtain the optimized battery pack finite element model.

[0011] Preferably, the initial values ​​of the design variables of the parameters to be optimized in step S4 are: upper top plate thickness 2.2 mm, middle enclosure plate thickness 2.4 mm, lower bottom plate thickness 3 mm, and the upper top plate material, middle enclosure plate material and lower bottom plate material are all aluminum alloy materials.

[0012] Preferably, the multi-objective optimization design function in step S5 is as follows:

[0013]

[0014] Where: T1, T2, T3, M1, M2, M3 are design variables, M is the mass of the battery pack, f is the first-order natural frequency; D e is the extrusion deformation; D b 、D t and D r They are the deformation of the battery pack during bumpy braking, the deformation of the battery pack during sharp turns in a bumpy state, and the deformation of the battery pack during reverse braking.

[0015] Preferably, the response surface method is used to fit the sample points and multiple target parameters to obtain an approximate model, including: based on multiple target parameters of the initial finite element model, and then using Nastran and Ls-dyna software imported into the initial finite element model data to calculate and obtain sample points, using the response surface method to construct an approximate model of the sample points and performance responses, and then performing multi-objective optimization design based on the approximate model, and obtaining the final lightweight optimization result through the NSGA-II algorithm.

[0016] Preferably, the lightweight optimization result is a Pareto solution set.

[0017] Preferably, obtaining the weight values ​​of the multiple target parameters includes the following steps:

[0018] The best-worst method and the entropy weight method are used to obtain different weight values ​​respectively, and then the different weight values ​​are synthesized using game theory to obtain the weight value.

[0019] Preferably, the best-worst method includes:

[0020] Through the best-worst method, in the criterion set {c1,c2,…,c n}, select the best criterion and the worst criterion, score them according to the scale, and construct the comparison vector A B =(a B1 ,a B2 ,…,a Bn ), where a Bi is the optimal criterion and the criterion set {c1,c2,…,c n c in} i The preference degree of all other criteria compared to the worst criterion is determined, where i takes the value of 1, 2, ..., n; the preference degree of all other criteria compared to the worst criterion is constructed, and the comparison vector A is constructed. W =(a 1w ,a 2w ,…,a nw ) T ;Convert the nonlinear BWM model into a linear model as follows:

[0021] minmax{|w B -a Bi w i |,|w i -a iw w W |}

[0022]

[0023] The linear model is transformed into:

[0024] minρ

[0025]

[0026] Solve the linear model to obtain the subjective weight value and ρ, where ρ is an indicator to measure the degree of consistency;

[0027] And / or, solve the objective weight value based on the entropy weight method:

[0028] First, establish a matrix for the evaluation indicators. If there are p evaluation indicators, the indicator values ​​of n evaluation objects constitute the evaluation matrix P:

[0029]

[0030] Where p ij For the evaluation value of the i-th evaluation object under the j-th evaluation index, find the proportion of the index value of the i-th evaluation object under the j-th evaluation index: Then find the entropy value of the j-th evaluation index:

[0031]

[0032] According to Shannon's information theory, when R ij =0,R ij ln ij =0, and finally find the difference coefficient ε of the jth evaluation index j and weight coefficient w j As an objective weight value:

[0033]

[0034] The weight coefficient reflects the amount of information in the indicator. The same evaluation indicator has different objective weights for different objects.

[0035] Preferably, the subjective weight value and the objective weight value are calculated based on game theory, including the following steps: establishing an initial weight vector set consisting of the subjective weight value and the objective weight value:

[0036] v l ={v l1 ,v l2 ,…,v lm}(l=1,2,…,η),

[0037] Define the linear combination between different vectors as:

[0038]

[0039] Where: v is the optimal weight vector in the weight vector set, β1 is the linear combination coefficient, which is always positive and sums to 1. By optimizing η combination coefficients β1 through game theory, v and v are promoted. l The spacing reaches the minimum deviation, as shown in the following formula:

[0040]

[0041] The expression of the optimized first-order derivative linear equation system corresponding to this formula is:

[0042]

[0043] Find the optimal vector (β1, β2, ..., β η )T After normalization, the most satisfactory game theory comprehensive weight vector v' is obtained as follows:

[0044]

[0045] Preferably, after obtaining the weight value, the following steps are also included: obtaining the optimal solution after the optimized design of the battery pack structure, inputting the obtained optimal design variables after the optimized design into the finite element model respectively, and obtaining the optimized battery pack finite element model.

[0046] The present disclosure also provides a battery pack, including the battery pack designed by the above method.

[0047] Compared with the prior art, the beneficial effects of the present disclosure are as follows: adopting the optimization objectives and constraints of multi-objective optimization, and then using the response surface method to fit the approximate model of the sample points and performance responses after calculating each performance, and realizing the multi-objective optimization design of the battery pack structure through the NSGA-II algorithm; after obtaining the Pareto solution set after the multi-objective optimization design, the subjective weight is obtained by the best-worst method, and the objective weight is obtained by the entropy weight method, and the game theory method is introduced to organically combine the subjective weight and the objective weight to avoid the defects of the single weighting method; the most satisfactory optimization design variables are input into the finite element model to obtain the optimized model for calculation and analysis of multiple working conditions, and the obtained optimization model achieves weight reduction without reducing performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is the initial finite element model of the battery pack;

[0049] Figure 2 is the Pareto solution set;

[0050] Figure 3 is the Pareto optimal solution;

[0051] Figure 4 To optimize the finite element model. DETAILED DESCRIPTION

[0052] The following will be combined with the accompanying drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the embodiments described are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present disclosure.

[0053] Example 1

[0054] This embodiment provides a safety design method for a new energy vehicle battery pack, including the following steps:

[0055] S1. Establish an initial finite element model of the battery pack. Based on the initial aluminum alloy battery pack structure at the bottom of the pure electric vehicle cab, establish a 3D solid model of the battery compartment in CATIA. Then, establish a finite element model in HyperMesh using the finite element modeling principles. Then, simplify the model, clean up the geometry, and mesh it according to a 10mm size. Finally, perform a quality check on the established finite element model to obtain the finite element model of the battery pack, such as Figure 1 The material used in the battery compartment structure is 6082-T6 aluminum alloy, and its elastic modulus is 7.2×10 4 MPa, Poisson's ratio is 0.32, and density is 2.7×10 -9 t / mm 3 , yield strength of 281 MPa, and tensile strength of 355 MPa. Furthermore, hexahedral elements were used to simulate the battery module elements. For example, the elastic modulus was set to 0.5 GPa, the density was set to 2000 kg / m³, and the Poisson's ratio was set to 0.01. Ultimately, 142,846 nodes and 141,626 elements were obtained, and the mesh quality met the requirements.

[0056] S2. Simulate the stress conditions of the battery pack under conditions of sudden braking on a bumpy road, sharp turning on a bumpy road, and reverse braking on a bumpy road based on the initial finite element model, as well as the dynamic performance of the first six modes under constraint conditions; wherein, referring to the stress conditions of the battery pack under actual conditions encountered during use, Nastran can be used to calculate the stress conditions of the battery pack under conditions such as braking, turning, and emergency braking on a road surface; the details are as follows:

[0057] 1) Sudden braking on bumpy roads

[0058] All translational and rotational degrees of freedom were constrained about the center of the bolt hole between the battery pack and the frame bracket, with vertical and longitudinal accelerations set to 2g and 1g, respectively. The force exerted by the battery assembly (300kg mass) on the battery pack was then applied as a uniformly distributed load at the corresponding locations. Finally, Nastran analysis revealed a deformation of 0.88mm.

[0059] 2) Sharp turns on bumpy roads

[0060] All translational and rotational degrees of freedom were constrained between the battery pack and the center of the frame bracket bolt hole, with vertical and lateral accelerations set to 2g and 1g, respectively. The force exerted by the battery assembly (300kg mass) on the battery pack was then applied as a uniformly distributed load at the corresponding locations. Finally, Nastran was used to calculate and analyze the deformation, which was 1.2mm.

[0061] 3) Reverse braking on bumpy roads

[0062] All translational and rotational degrees of freedom were constrained between the battery pack and the center of the frame bracket bolt hole, with vertical and lateral accelerations set to 2g and 0.8g, respectively. The force exerted by the battery assembly (300kg mass) on the battery pack was then applied as a uniformly distributed load at the corresponding locations. Finally, Nastran was used to calculate and analyze the deformation, which was 0.65mm.

[0063] 4) Modes under constraints

[0064] A full-degree-of-freedom constraint simulation was performed on the bolt holes connecting the battery pack to the frame. Then, the modal analysis method was set to EIGRL in HyperMesh, and the finite element model of the battery pack was imported into Nastran for solution. The first-order frequency of the initial model was obtained to be 25.6 Hz.

[0065] S3. Perform a simulated extrusion test on the battery pack defined by the initial finite element model. Based on the simulated extrusion test results, partially replace the materials of the battery pack, wherein the upper top plate is made of carbon fiber composite material, the middle enclosure is made of high-strength steel material, and the lower bottom plate is made of aluminum alloy material;

[0066] In this embodiment, according to the requirements of GB / 31467.3-2005, a 90cm long semi-cylinder (8cm radius) was used to analyze the extrusion deformation of the battery pack structure. The extrusion direction was the X and Y directions (the vehicle's travel direction was the X axis). At the same time, a dedicated battery pack extruder, model CX-5067-AP, was used as a reference. An extrusion plate was installed on the non-extrusion side of the extrusion direction. The three translational and three rotational degrees of freedom of the extrusion plate were constrained and simulated using the rigid material MATL20. All degrees of freedom of the extrusion head of the extrusion column, except for the translational degree of freedom in the extrusion direction, were constrained. A constant velocity was applied in the extrusion direction to extrude the battery pack. The extrusion ends when the extrusion pressure reaches 100KN (Amendment No. 1 of GB / 31467.3-2005) or the extrusion deformation in the extrusion direction reaches 30% of the overall size of the battery pack; it is maintained for 10 minutes, and it is required that the battery pack does not catch fire or explode; then, based on the performance results, the aluminum alloy material of the battery pack is partially replaced, and the upper top plate is initially selected to use carbon fiber composite material, the middle enclosure is selected to use high-strength steel material, and the lower bottom plate is still selected to use traditional aluminum alloy material.

[0067] S4. The top plate thickness, middle panel thickness, lower panel thickness, top plate material, middle panel material, and lower panel material of the battery pack structural parameters were selected as design variables for optimization, and initial values ​​were set for each of these design variables. The top plate thickness, middle panel thickness, lower panel thickness, top plate material, middle panel material, and lower panel material were described by T1, T2, T3, M1, M2, and M3, respectively. The initial value of the dimension variable and a 20% fluctuation were used as the three levels, and carbon fiber composite material, aluminum alloy, and high-strength steel were defined as the three different material levels. A factor level table for the battery pack structural parameter design of experiment simulation analysis was established based on the Latin hypercube, as shown in Table 1.

[0068] Table 1 Structural parameter factor level table

[0069] Run T1 T2 T3 M1 M2 M3 1 1.9 2 3.1 1 2 2 2 1.9 2.5 2.8 2 3 1 3 2 2.2 2.6 3 1 3 4 2.4 2.1 3 1 3 1 5 1.8 2.8 3.4 2 3 3 6 1.8 2.4 2.6 3 3 2 7 2.4 2.8 3.4 1 2 2 8 2.6 2 3.5 2 3 3 9 2.1 2 3.1 3 2 2 10 1.9 2.2 3.5 1 1 2 11 2.6 2.5 2.7 2 1 2 12 2.2 2 2.5 3 3 1 13 2.5 2.3 2.7 1 2 3 14 2.3 2.7 3.2 2 2 3 15 2 2.2 2.5 3 2 1 16 2.2 2.1 2.7 1 3 1 17 1.9 2.4 3.3 2 1 3 18 2.4 2.1 2.8 3 1 1 19 2.1 2.3 3 1 2 3 20 1.8 2.1 2.9 2 1 2 21 2.3 2.3 2.6 3 1 2 22 2 2.7 3.2 1 3 1 23 2.5 2.8 2.9 2 1 2 24 2 2.5 3.2 3 1 2 25 2.2 2.2 2.9 1 3 1 26 2.5 2.6 3.3 2 2 1 27 2.1 2.7 3.1 3 3 3 28 1.8 2.6 2.5 1 2 3 29 2.6 2.6 3 2 2 1 30 2.3 2.4 2.8 3 1 3

[0070] The six design variables of size and material, including the upper top plate thickness, middle panel thickness, lower bottom plate thickness, upper top plate material, middle panel material and lower bottom plate material, in the defined battery pack structural size parameters are used as parameter design variables to be optimized, with their initial values ​​being 2.2mm, 2.4mm, 3mm, aluminum alloy material (5052), aluminum alloy material (5052), aluminum alloy material (5052); the upper and lower limits of their sizes are [1.8, 2.6], [2, 2.8], [2.5, 3.5]; the level values ​​of the material variables (1, 2, 3) correspond to DP450, carbon fiber composite material and aluminum alloy 5052, respectively.

[0071] S5. Determine the optimization objectives and constraints for the battery pack. The optimization objectives are maximizing the first-order natural frequency, minimizing the mass, and minimizing the extrusion deformation. The constraints are that the maximum deformation under sudden braking, sharp turns, and reverse braking on bumpy roads is no more than 2 mm. Obtain a multi-objective optimization design function.

[0072] In this embodiment, the mathematical model of multi-objective optimization can be expressed as:

[0073]

[0074] Where: T1, T2, T3, M1, M2, M3 are design variables, M is the mass of the battery pack, f is the first-order natural frequency; D e is the extrusion deformation; D b 、D t and D r They are the deformation of the battery pack during bumpy braking, the deformation of the battery pack during sharp turns in a bumpy state, and the deformation of the battery pack during reverse braking.

[0075] S6. Calculating multiple performance target parameters in the multi-objective optimization design function based on the initial finite element model data with boundary conditions set in S4 and S5, obtaining initial model test design sample points through the initial design variables, and then fitting the sample points and target parameters using the response surface method to obtain an approximate model;

[0076] In this embodiment, the response surface method is used to fit the sample points and multiple target parameters to obtain an approximate model, including: based on multiple target parameters of the initial finite element model, and then using Nastran and Ls-dyna software to import the initial finite element model data to calculate and obtain sample points, the response surface method is used to construct an approximate model of the sample points and performance responses, and then based on the approximate model, an optimization design under a multi-objective optimization design function is performed to obtain the final lightweight optimization result. The fitting accuracy of the approximate model is analyzed. It can be seen from Table 2 that the R 2 Value (R 2 The value is generally between [0-1], the closer to 1, the better the fit) are all above 0.89, meeting the R requirements in engineering design 2 The value is greater than 0.85, so this approximate model can achieve multi-objective optimization design of battery pack structure. Among them, based on the performance information of the initial model of the battery pack, and then after obtaining sample points based on the experimental design, the response surface method is used to build an approximate model of sample point information and performance response based on a large number of sample points. Then, the NSGA-II algorithm is used to optimize the design based on the approximate model to obtain the final lightweight optimization result, such as Figure 2 As shown in Figure 2, the lightweight optimization result is a Pareto solution set.

[0077] Table 2 R of each response of the battery pack approximate model 2 value

[0078] response M f <![CDATA[D e ]]> <![CDATA[D b ]]> <![CDATA[D t ]]> <![CDATA[D r ]]> <![CDATA[R 2 ]]> 0.9965 0.9921 0.9134 0.9546 0.9512 0.9471

[0079] S7. Use the NSGA-II algorithm in HyperStudy to obtain the multi-objective Pareto solution set, and then obtain the weight values ​​of the multiple objective parameters based on the best-worst solution-entropy weight method-game theory, and then obtain the optimal solution after the optimized design of the battery pack structure. Input the optimal solution into the initial finite element model to obtain the optimized battery pack finite element model.

[0080] In this embodiment, obtaining the weight values ​​of the multiple target parameters includes the following steps:

[0081] The best-worst method and the entropy weight method are used to obtain different weight values ​​respectively, and then the different weight values ​​are synthesized using game theory to obtain the weight value.

[0082] In some preferred embodiments, the best-worst method is used to select the best-worst criterion set {c1, c2, ..., c n}, select the best criterion and the worst criterion, score them according to the scale, and construct the comparison vector A B =(a B1 ,a B2 ,…,a Bn ), where a Bi is the optimal criterion and the criterion set {c1,c2,…,c n c in} i The preference degree of all other criteria compared to the worst criterion is determined, where i takes the value of 1, 2, ..., n; the preference degree of all other criteria compared to the worst criterion is constructed, and the comparison vector A is constructed. W =(a 1w ,a 2w ,…,a nw ) T ;Convert the nonlinear BWM model into a linear model as follows:

[0083] minmax{|w B -a Bi w i |,|w i -a iw w W |}

[0084]

[0085] The linear model is transformed into:

[0086] minρ

[0087]

[0088] Solve the linear model to obtain the subjective weight value And ρ, where ρ is an indicator for measuring the degree of consistency; this value can be considered as an indicator for measuring the consistency relationship. The closer the value is to 0, the higher the degree of consistency.

[0089] In some preferred embodiments, the entropy weight method has simple calculation steps, effectively utilizes the index data, and eliminates the influence of subjective factors. In this evaluation index method, the various evaluation indicators have great differences in content, dimension, and value quality. Generally, it is necessary to perform dimensionless processing of the data. The standardization processing system includes two types of indicators: positive indicators and negative indicators. At the same time, these two types of indicators are standardized. When the index data is larger, it is better, that is, it is a positive indicator, and its standardization formula is When the indicator value is as small as possible, that is, it is an inverse indicator, its standardized formula is: The data indicators studied in this embodiment include positive indicators and negative indicators.

[0090] Solving the objective weight value based on the entropy weight method includes the following steps:

[0091] First, establish a matrix for the evaluation indicators. If there are p evaluation indicators, the indicator values ​​of n evaluation objects constitute the evaluation matrix P:

[0092]

[0093] Where p ij For the evaluation value of the i-th evaluation object under the j-th evaluation index, find the proportion of the index value of the i-th evaluation object under the j-th evaluation index: Then find the entropy value of the j-th evaluation index:

[0094]

[0095] According to Shannon's information theory, when R ij =0,R ij ln ij =0, and finally find the difference coefficient ε of the jth evaluation index j and weight coefficient w j As an objective weight value:

[0096]

[0097] The weight coefficient reflects the amount of information in the indicator. The same evaluation indicator has different objective weights for different objects.

[0098] Preferably, the subjective weight value and the objective weight value are calculated based on game theory, including the following steps: establishing an initial weight vector set consisting of the subjective weight value and the objective weight value:

[0099] v l ={v l1 ,v l2 ,…,v lm}(l=1,2,…,η),

[0100] Define the linear combination between different vectors as:

[0101]

[0102] Where: v is the optimal weight vector in the weight vector set, β1 is the linear combination coefficient, which is always positive and sums to 1. By optimizing η combination coefficients β1 through game theory, v and v are promoted. l The spacing reaches the minimum deviation, as shown in the following formula:

[0103]

[0104] The expression of the optimized first-order derivative linear equation system corresponding to this formula is:

[0105]

[0106] Find the optimal vector (β1, β2, ..., β η ) T After normalization, the most satisfactory game theory comprehensive weight vector v' is obtained as follows:

[0107]

[0108] Final performance M, f, D e The weight values ​​obtained through game theory are v'=(0.4009, 0.2867, 0.3124).

[0109] As a preferred embodiment, after obtaining the weight value, the following steps are also included: obtaining the optimal solution of the battery pack structure after optimization design, inputting the obtained optimal design variables after optimization design into the finite element model respectively, and obtaining the optimized battery pack finite element model.

[0110] After the calculation of the above weight values, the Pareto optimal solution after the optimized design of the battery pack structure is finally obtained. Figure 3 Shown (five-pointed star).

[0111] S8. By performing multi-objective optimization design on the battery pack, the optimized design variables obtained above are input into the finite element model respectively to obtain an optimized finite element model of the battery pack.

[0112] The changes of design variables before and after optimization are shown in Table 3.

[0113] Table 3 Battery pack design variables before and after optimization

[0114] T1 / mm T2 / mm T3 / mm M1 M2 M3 initial 2.2 2.4 3 Aluminum alloy 5052 Aluminum alloy 5052 Aluminum alloy 5052 optimization 1.8 2.2 2.6 DP450 carbon fiber Aluminum alloy 5052

[0115] The optimized model was analyzed for the aforementioned operating conditions to obtain performance responses, and the performance responses before and after optimization are summarized in Table 4. The results show that the model achieved a 22.3% weight reduction without sacrificing performance.

[0116] Table 4 Performance response before and after optimization

[0117] M / kg f / Hz <![CDATA[D e / mm]]> <![CDATA[D b / mm]]> <![CDATA[D t / mm]]> <![CDATA[D r / mm]]> initial 260 25.6 0.92 0.88 1.2 0.65 optimization 202.02 29.4 1.11 1.07 1.36 0.97 Relative rate of change -22.3％ 14.84％ 20.65％ 21.59％ 13.33％ 49.23％

[0118] The optimized finite element model is as follows Figure 4As shown. This embodiment adopts the optimization objectives and constraint functions of multi-objective optimization. After calculating each performance, the response surface method is used to fit an approximate model to the sample points and performance responses, achieving multi-objective optimization design of the battery pack structure. After obtaining the Pareto solution set after the multi-objective optimization design, the best-worst method is used to obtain subjective weights, and the entropy weight method is used to obtain objective weights. Game theory methods are introduced to organically combine subjective and objective weights to avoid the shortcomings of a single weighting method. The most satisfactory optimization design variables are input into the finite element model to obtain an optimized model for calculation and analysis of multiple operating conditions. The obtained model achieves a weight reduction of 22.3% without compromising performance.

[0119] Compared with the prior art, the beneficial effects of this embodiment are as follows: adopting the optimization objectives and constraints of multi-objective optimization, and then using the response surface method to fit the approximate model of the sample points and performance responses after calculating each performance, so as to realize the multi-objective optimization design of the battery pack structure; after obtaining the Pareto solution set after the multi-objective optimization design, the subjective weight is obtained by the best-worst method, and the objective weight is obtained by the entropy weight method, and the game theory method is introduced to organically combine the subjective weight and the objective weight to avoid the defects of the single weighting method; the most satisfactory optimization design variables are input into the finite element model to obtain the optimized model for calculation and analysis of multiple working conditions, and the obtained model achieves weight reduction without reducing performance.

[0120] Example 2

[0121] This embodiment also provides a battery pack designed using the design method in Example 1.

[0122] Although the embodiments of the present disclosure have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations may be made to the embodiments without departing from the principles and spirit of the present disclosure, and the scope of the present disclosure is defined by the appended claims and their equivalents.

Claims

1. A safety design method for a new energy vehicle battery pack, characterized in that: The steps include: S1. Establish an initial finite element model of the battery pack; S2. Simulating the stress conditions of the battery pack under sudden braking on a bumpy road, sharp turning on a bumpy road, and reverse braking on a bumpy road based on the initial finite element model, as well as the dynamic performance of the first six modes under constraint conditions; S3. Perform a simulated extrusion test on the battery pack defined by the initial finite element model. Based on the simulated extrusion test results, partially replace the materials of the battery pack, wherein the upper top plate is made of carbon fiber composite material, the middle enclosure is made of high-strength steel material, and the lower bottom plate is made of aluminum alloy material; S4. Using the upper top plate thickness, middle panel thickness, lower bottom plate thickness, upper top plate material, middle panel material, and lower bottom plate material in the battery pack structural dimension parameters as parameter design variables to be optimized, and setting initial values ​​for the above parameter design variables to be optimized respectively; S5. Determine the optimization objectives and constraints for the battery pack. The optimization objectives are maximizing the first-order natural frequency, minimizing the mass, and minimizing the extrusion deformation. The constraints are that the maximum deformation under sudden braking, sharp turns, and reverse braking on bumpy roads is no more than 2 mm. Obtain a multi-objective optimization design function. S6. Calculating multiple performance target parameters in the multi-objective optimization design function based on the initial finite element model data with boundary conditions set in S4 and S5, obtaining initial model test design sample points through the initial design variables, and then fitting the sample points and target parameters using the response surface method to obtain an approximate model; S7. Use the NSGA-II algorithm in HyperStudy to obtain the multi-objective Pareto solution set, and then obtain the weight values ​​of the multiple objective parameters based on the best-worst solution-entropy weight method-game theory, and then obtain the optimal solution after the optimized design of the battery pack structure. Input the optimal solution into the initial finite element model to obtain the optimized battery pack finite element model.

2. The safety design method for a new energy vehicle battery pack according to claim 1, characterized in that: The initial values ​​of the design variables of the parameters to be optimized in step S4 are: upper top plate thickness 2.2 mm, middle enclosure plate thickness 2.4 mm, lower bottom plate thickness 3 mm, and the upper top plate material, middle enclosure plate material and lower bottom plate material are all aluminum alloy materials.

3. The safety design method for a new energy vehicle battery pack according to claim 1, wherein: The multi-objective optimization design function in step S5 is as follows: Where: T1, T2, T3, M1, M2, M3 are design variables, M is the mass of the battery pack, f is the first-order natural frequency; D e is the extrusion deformation; D b 、D t and D r They are the deformation of the battery pack during bumpy braking, the deformation of the battery pack during sharp turns in a bumpy state, and the deformation of the battery pack during reverse braking.

4. The safety design method for a new energy vehicle battery pack according to claim 3, characterized in that: The response surface methodology is used to fit the sample points and multiple target parameters to obtain an approximate model, including: obtaining the initial model test design sample points through the Latin hypercube method, importing each sample point into the initial finite element model, and then calculating them separately through Nastran and Ls-dyna software, using the response surface methodology to construct an approximate model of the sample points and performance responses, and then performing multi-objective optimization design based on the approximate model to obtain the final lightweight optimization result.

5. The safety design method for a new energy vehicle battery pack according to claim 4, characterized in that: The lightweight optimization result of the NSGA-II algorithm is a Pareto solution set.

6. The safety design method for a new energy vehicle battery pack according to any one of claims 1 to 5, characterized in that: Obtaining weight values ​​of the multiple target parameters includes the following steps: The best-worst method is used to obtain the subjective weight value, the entropy weight method is used to obtain the objective weight value, and then game theory is used to integrate the subjective weight value and the objective weight value to obtain the comprehensive weight value.

7. The safety design method for a new energy vehicle battery pack according to claim 6, characterized in that: The best-worst method includes: using the best-worst method to select the best-worst method in the criterion set {c1, c2, ..., c n }, select the best criterion and the worst criterion, score them according to the scale, and construct the comparison vector A B =(a B1 ,a B2 ,…,a Bn ), where a Bi is the optimal criterion and the criterion set {c1,c2,…,c n c in} i The preference degree of all other criteria compared to the worst criterion is determined, where i takes the value of 1, 2, ..., n; the preference degree of all other criteria compared to the worst criterion is constructed, and the comparison vector A is constructed. W =(a 1w ,a 2w ,…,a nw ) T ;Convert the nonlinear BWM model into a linear model as follows: min max{|w B -a Bi w i |,|w i -a iw w W |} The linear model is transformed into: minρ Solve the linear model to obtain the subjective weight value and ρ, where ρ is an indicator to measure the degree of consistency; And / or, solve the objective weight value based on the entropy weight method: Establish a matrix for the evaluation indicators. If there are p evaluation indicators, the indicator values ​​of n evaluation objects constitute the evaluation matrix P: Where p ij For the evaluation value of the i-th evaluation object under the j-th evaluation index, find the proportion of the index value of the i-th evaluation object under the j-th evaluation index: Then find the entropy value of the j-th evaluation index: According to Shannon's information theory, when R ij =0,R ij ln R ij =0, and finally find the difference coefficient ε of the jth evaluation index j and weight coefficient w j As an objective weight value: The weight coefficient reflects the amount of information in the indicator. The same evaluation indicator has different objective weights for different objects.

8. The safety design method for a new energy vehicle battery pack according to claim 7, characterized in that: Based on game theory, the subjective weight values ​​and objective weight values ​​calculated above are comprehensively calculated to obtain a comprehensive weight value, which includes the following steps: establishing an initial weight vector set consisting of the sovereign weight value and the objective weight value: v l ={v l1 ,v l2 ,…,v lm }(l=1,2,…,η), Define the linear combination between different vectors as: Where: v is the optimal weight vector in the weight vector set, β1 is the linear combination coefficient, which is always positive and sums to 1. By optimizing η combination coefficients β1 through game theory, v and v are promoted. l The spacing reaches the minimum deviation, as shown in the following formula: The expression of the optimized first-order derivative linear equation system corresponding to this formula is: Find the optimal vector (β1, β2, ..., β η ) T After normalization, the most satisfactory game theory comprehensive weight vector v' is obtained as follows:

9. The safety design method for a new energy vehicle battery pack according to claim 8, characterized in that: After obtaining the weight value, the following steps are also included: obtaining the optimal solution after the optimized design of the battery pack structure, inputting the obtained optimal design variables after the optimized design into the finite element model respectively, and obtaining the optimized battery pack finite element model.

10. A battery pack, characterized in that: A battery pack designed and obtained by the method according to any one of claims 1 to 9.

Images