Method and system for optimizing crashworthiness of energy absorption structure of railway vehicle

Abstract

The invention relates to the technical field of structure optimization, and discloses a collision resistance optimization method and system for a rail vehicle energy absorption structure, and the method comprises the steps: determining corresponding simulation parameters influencing energy absorption and a multi-dimensional collision mechanical response data set of the energy absorption structure; constructing a multi-dimensional collision response prediction model; defining a mode stability constraint based on a decoding result of the image feature extraction network, defining a load stability constraint based on a decoding result of the curve feature extraction network, and defining an index matching constraint based on an output result of the multi-dimensional collision response prediction model; based on the mode stability constraint, the load stability constraint and the index matching constraint, a multi-target crashworthiness optimization model of the energy absorption structure is constructed, and solving is carried out to obtain a Pareto solution set; solving an optimal feasible solution of the Pareto solution set, and taking the optimal feasible solution as a collision resistance optimization result of the target energy absorption structure; the method solves the problem that an existing crashworthiness performance optimization method is low in optimization result reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of structural optimization, and in particular to a method and system for optimizing the crashworthiness of an energy-absorbing structure of a rail vehicle. Background Art

[0002] Energy-absorbing structures for rail vehicles can absorb impact kinetic energy through their own plastic deformation when a train collides, thereby protecting the safety of drivers and passengers. However, due to the different forms of energy-absorbing structures, numerous structural parameters, significant coupling interactions, and strong nonlinear collision response, the optimization of crashworthiness has always been a core problem in the design of energy-absorbing structures. At present, the optimization method driven by the agent model is mainly used, such as the "Multi-objective Optimization Method for Crashworthiness of Square Cone Energy-Absorbing Structures Based on Machine Learning" (CN202211151523.7), which optimizes by constructing an agent model (including response surface, Kriging, radial basis function, etc., in this case, the response surface model) of crashworthiness indicators (such as energy absorption, peak force, etc.), and cannot consider constraints such as deformation mode and load fluctuation characteristics. Although the optimization results can ensure that the average load, energy absorption, etc. meet the design expectations, non-ideal situations such as unstable deformation mode and drastic load fluctuation may occur, which seriously affect the design iteration of the scheme and the efficiency of new product development. It can be seen that the existing crashworthiness optimization methods have the problem of low reliability of optimization results. Summary of the invention

[0003] The present invention provides a method and system for optimizing the crashworthiness performance of an energy-absorbing structure of a rail vehicle, so as to solve the problem of low reliability of optimization results in existing crashworthiness optimization methods.

[0004] In order to achieve the above object, the present invention is implemented by the following technical solutions: In a first aspect, the present invention provides a method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure, comprising: Determine the simulation parameters that affect the energy absorption according to the type of the target energy absorption structure, and perform numerical simulation based on the simulation parameters to obtain a multi-dimensional collision mechanical response data set of the energy absorption structure; A multi-dimensional collision response prediction model is constructed based on a deep learning framework and the multi-dimensional collision mechanical response data set of the energy-absorbing structure, wherein the multi-dimensional collision response prediction model includes: an image feature extraction network, a curve feature extraction network and a feature vector prediction network; Defining a pattern stability constraint based on a decoding result of the image feature extraction network, defining a load stability constraint based on a decoding result of the curve feature extraction network, and defining an index matching constraint based on an output result of the multi-dimensional collision response prediction model; A multi-objective crashworthiness optimization model of an energy-absorbing structure is constructed based on the mode stability constraint, the load stability constraint, and the index matching constraint, and a non-dominated sorting genetic algorithm is used to solve the model to obtain a Pareto solution set; The optimal feasible solution of the Pareto solution set is solved based on the entropy-weighted VIKOR multi-attribute decision-making method, and the optimal feasible solution is used as the optimization result of the crashworthiness performance of the target energy-absorbing structure.

[0005] Optionally, the multi-dimensional collision mechanical response data set of the energy-absorbing structure obtained by performing numerical simulation based on the simulation parameters includes: Perform impact tests and corresponding numerical simulations on the target energy absorbing structure according to the type and simulation parameters of the target energy absorbing structure, and calibrate the numerical simulation model; The experimental design is carried out based on the numerical simulation model after experimental benchmarking, and a multi-dimensional collision mechanical response data set of the energy-absorbing structure is generated by integrating experiment and simulation.

[0006] Optionally, the multi-dimensional collision response prediction model is constructed based on the deep learning framework and the multi-dimensional collision mechanical response dataset of the energy absorbing structure, including: The multi-dimensional collision mechanical response data set of the energy-absorbing structure is randomly divided into a training set, a validation set and a test set in a ratio of 8:1:1; A deep learning model is constructed through a deep learning framework, and the deep learning model is trained using training sets, validation sets, and test sets to obtain a multi-dimensional collision response prediction model.

[0007] Optionally, the image feature extraction network is used to construct an autoencoder network with stacked convolutional layers as the main architecture, and extract the deformation mode image feature vector, wherein the deformation mode image feature vector is: ; The curve feature extraction network is used to construct an autoencoder network with stacked long and short-term memory layers as the main structure, and extract the load characteristic curve feature vector, wherein the load characteristic curve feature vector is: ; The feature vector prediction network is used to construct a vector prediction network with a fully connected layer as the main body, and fuses the deformation model image feature vector, the load characteristic curve feature vector and the crashworthiness index feature vector directly extracted from the multi-dimensional collision mechanics response data set of the energy absorbing structure into an output vector.

[0008] Optionally, the defining a pattern stability constraint based on a decoding result of the image feature extraction network includes: After obtaining the prediction result of the output vector through the multi-dimensional collision response prediction model, the feature vector part of the deformation model image is input into the image feature extraction network for decoding to obtain the pattern image prediction result; The mode stability constraint is defined based on the mode image prediction result. The stability judgment basis is different according to the type of target energy-absorbing structure. When the mode is stable, the constraint value is returned to 0, and when the mode is unstable, the constraint value is returned to 1.

[0009] Optionally, the defining a load stability constraint based on a decoding result of the curve feature extraction network includes: After obtaining the prediction result of the output vector through the multi-dimensional collision response prediction model, the characteristic vector part of the load characteristic curve is input into the curve feature extraction network for decoding to obtain the load curve prediction result; The load stability constraint is defined based on the load curve prediction result, where the load stability constraint satisfies the following relationship: ; In the formula, S 2 represents the load stability constraint value, θ is the preset constraint value of this indicator, N represents the number of sampling points of the load curve, Fi represents the load value at any sampling time, Represents the average value of the load curve. , the return constraint value is 0, otherwise the return constraint value is 1.

[0010] Optionally, the defining an indicator matching constraint based on the output result of the multi-dimensional collision response prediction model includes: The crashworthiness index prediction result is obtained by directly outputting the crashworthiness index vector part, and the index matching constraint is defined based on the crashworthiness index prediction result. According to the difference in the target energy absorption structure type, the matching judgment basis is different. When the matching meets the requirements, the constraint value is returned to 0, and when the mode is unstable, the constraint value is returned to 1. Optionally, the multi-objective crashworthiness optimization model of the energy absorbing structure satisfies the following relationship: ; Where, Obj i represents the optimization goal, V stands for Structural parameters, VL and VU represent the upper and lower boundaries of the structural parameters respectively, P, PL and PU represent the crashworthiness index and its upper and lower boundaries respectively, represents the mode stability constraint, Represents the load fluctuation constraint.

[0011] Optionally, the entropy-weighted VIKOR multi-attribute decision-making method is used to solve the optimal feasible solution of the Pareto solution set, including: S1. Construct a standardized solution set, which satisfies the following relationship: ; In the formula, xij represents the original feasible solution, z ij represents the standardized feasible solution; S2. Determine the positive ideal solution and the negative ideal solution in the standardized solution set. The determination method satisfies the following relationship:

[0012] ; In the formula, Z + represents the set of positive ideal solutions, Z i + Representative i The positive ideal solution of the target, Z - represents the set of negative ideal solutions, Z i - Representative i Negative ideal solution of a target; S3. Calculate the weight of each optimization target based on the entropy value, and the calculation includes: The contribution of the calculation target in the feasible solution satisfies the following relationship: ; In the formula, w ij represents the contribution in the feasible solution, x ij represents the original feasible solution; Calculate the entropy value of the target, and its calculation satisfies the following relationship: ; In the formula, e j represents the entropy value, w ij represents the contribution in the feasible solution; Calculate the weight of the target, and its calculation satisfies the following relationship: ; In the formula, ω j represents the weight value, e j represents the entropy value; S4. Calculate the group contribution value and individual regret value of each feasible solution. The calculation of the group contribution value satisfies the following relationship: ; In the formula, S i represents the group contribution value, ω j Represents the weight value, Z j + Representativej The positive ideal solution of the target, Z j - Representative j The negative ideal solution of the target, z ij represents the standardized feasible solution; The calculation of its individual regret value satisfies the following relationship: ; In the formula, R i represents the individual regret value, ω j Represents the weight value, Z j + Representative j The positive ideal solution of the target, Z j - Representative j The negative ideal solution of the target, z ij represents the standardized feasible solution; S5. Calculate the contribution rate of each feasible solution, and the calculation satisfies the following relationship: ; In the formula, Q i represents the contribution rate, S + , S - Represent the group contribution values ​​of positive ideal solution and negative ideal solution respectively, R + , R - Represent the individual regret values ​​of positive ideal solutions and negative ideal solutions respectively.

[0013] S6. Sort the contribution rates of feasible solutions, and obtain the optimal feasible solution in the Pareto solution set based on the sorting results.

[0014] In a second aspect, an embodiment of the present application provides a system for optimizing the crashworthiness performance of an energy-absorbing structure of a rail vehicle, including a processor and a memory; Memory, used to store computer programs; The processor is used to implement any method step described in the first aspect when executing a program stored in the memory.

[0015] Beneficial effects: The method for optimizing the crashworthiness of the energy-absorbing structure of a rail vehicle provided by the present invention can ensure that the indicators of the energy-absorbing structure meet the requirements by applying the proposed optimization method, and at the same time, the deformation mode is stable and the load fluctuation is smooth, thereby greatly improving the reliability of the optimization result; and in the final decision-making link, a multi-attribute decision-making method based on entropy value weighted VIKOR is proposed, and weighting is performed by calculating the entropy value of each optimization target, thereby effectively avoiding the interference of human factors in the multi-target decision-making process, and the obtained decision result is more objective. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of a method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure according to a preferred embodiment of the present invention; Figure 2 A schematic diagram of a network framework of a multi-dimensional collision response prediction model according to a preferred embodiment of the present invention; Figure 3 A schematic diagram of an image feature extraction network architecture according to a preferred embodiment of the present invention; Figure 4 A schematic diagram of a curve feature extraction network architecture according to a preferred embodiment of the present invention; Figure 5 This is a schematic diagram of the optimization results of the thin-walled circular tube energy absorption structure in the preferred embodiment 1 of the present invention; Figure 6 It is a schematic diagram of the constraint optimization result of the expansion type energy absorption structure in the preferred embodiment 2 of the present invention; Figure 7 This is a schematic diagram of the optimization results of the honeycomb-filled evil energy structure in the preferred embodiment 3 of the present invention. DETAILED DESCRIPTION

[0017] The technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the usual meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "one" do not indicate quantity restrictions, but indicate the existence of at least one. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0019] See also Figure 1-7 The present application provides a method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure, including: Determine the simulation parameters that affect the energy absorption according to the type of the target energy absorption structure, and perform numerical simulation based on the simulation parameters to obtain a multi-dimensional collision mechanical response data set of the energy absorption structure; A multi-dimensional collision response prediction model is constructed based on a deep learning framework and the multi-dimensional collision mechanical response data set of the energy-absorbing structure, wherein the multi-dimensional collision response prediction model includes: an image feature extraction network, a curve feature extraction network and a feature vector prediction network; Defining a pattern stability constraint based on a decoding result of the image feature extraction network, defining a load stability constraint based on a decoding result of the curve feature extraction network, and defining an index matching constraint based on an output result of the multi-dimensional collision response prediction model; A multi-objective crashworthiness optimization model of an energy-absorbing structure is constructed based on the mode stability constraint, the load stability constraint, and the index matching constraint, and a non-dominated sorting genetic algorithm is used to solve the model to obtain a Pareto solution set; The optimal feasible solution of the Pareto solution set is solved based on the entropy-weighted VIKOR multi-attribute decision-making method, and the optimal feasible solution is used as the optimization result of the crashworthiness performance of the target energy-absorbing structure.

[0020] In response to the above content, the following three examples, namely a thin-walled circular tube energy absorption structure, a bulging energy absorption structure, a reduced diameter energy absorption structure and a honeycomb filled energy absorption structure, are used to illustrate the effectiveness of the method proposed in this scheme.

[0021] Example 1, Optimization of thin-walled circular tube energy absorption structure In the thin-walled circular tube structure optimization problem, the key structural parameters include wall thickness t, radius r, material property parameters A, B, and n; the optimization goal is to maximize specific energy absorption (SEA) and minimize peak force (PCF); the boundary ranges and constraints of each structural parameter are shown in the following formula.

[0022]

[0023] The NSGA algorithm is used to solve the above optimization problem, where the number of iterations is set to 200 and the population size is 50. After optimization, the optimal feasible solution is calculated according to the multi-attribute decision-making method proposed in the scheme of the present invention. In order to further verify the reliability of the optimization results obtained by the proposed multi-dimensional collision mechanics response prediction model, a finite element simulation model of the thin-walled circular tube structure with the optimal feasible solution is established, and its collision mechanics response is calculated and compared with the optimization results after prediction. The results are as follows Figure 5 shown.

[0024] The results show that the deep learning prediction framework for the multi-dimensional collision mechanical response of the energy-absorbing structure proposed in the present invention can accurately predict the various collision responses of the thin-walled circular tube structure. After applying it to the fusion constraint optimization, the deformation mode of the obtained results is stable, the load fluctuation is controllable, and the collision resistance index meets the requirements, which fully demonstrates the effectiveness of this scheme.

[0025] Example 2: Optimization of the inflatable energy absorption structure In the optimization problem of bulging structure, the key structural parameters include wall thickness t, inner diameter r e , outer diameter r m , bulging cone angle α , friction coefficient u ; The optimization goal is to maximize specific energy absorption (SEA) and minimize peak crushing force (PCF); the boundary range and constraints of each structural parameter are shown in the following formula.

[0026]

[0027] The NSGA algorithm is used to solve the above optimization problem, where the number of iterations is set to 200 and the population size is 50. After optimization, the optimal feasible solution is calculated according to the multi-attribute decision-making method proposed in the scheme of the present invention. The finite element simulation model of the optimal feasible solution bulging energy absorption structure is established, and its collision mechanical response is calculated and compared with the predicted optimization results. The results are as follows Figure 6 shown.

[0028] The results show that the deformation mode of the optimized bulging energy-absorbing structure is stable, the load fluctuation is controllable, and the crashworthiness index meets the requirements, which fully demonstrates the effectiveness of this scheme.

[0029] Example 3: Optimization of honeycomb-filled energy-absorbing structure In the optimization problem of honeycomb filling structure, the key structural parameters include the strength of the filling honeycomb block P 1. P 2. P 3. Coating wall thickness T , Covering height L , fill the gap S ; The optimization goal is to maximize specific energy absorption (SEA) and minimize peak crushing force (PCF); the boundary range and constraints of each structural parameter are shown in the following formula.

[0030]

[0031] The NSGA algorithm is used to solve the above optimization problem, where the number of iterations is set to 200 and the population size is 50. After optimization, the optimal feasible solution is calculated according to the multi-attribute decision-making method proposed in the scheme of the present invention. A finite element simulation model of the optimal feasible solution honeycomb-filled energy-absorbing structure is established, and its collision mechanical response is calculated and compared with the predicted optimization results. The results are as follows Figure 6 Shown The embodiment of the present application also provides a system for optimizing the crashworthiness performance of a rail vehicle energy absorption structure, including a processor and a memory; Memory, used to store computer programs; The processor is used to implement any method step in the method for optimizing the crashworthiness performance of the energy-absorbing structure of a rail vehicle when executing the program stored in the memory.

[0032] The above-mentioned system for optimizing the crashworthiness performance of the energy-absorbing structure of a rail vehicle can implement various embodiments of the above-mentioned method for optimizing the crashworthiness performance of the energy-absorbing structure of a rail vehicle, and can achieve the same beneficial effects, which will not be described in detail here.

[0033] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure, characterized in that: include: Determine the simulation parameters that affect the energy absorption according to the type of the target energy absorption structure, and perform numerical simulation based on the simulation parameters to obtain a multi-dimensional collision mechanical response data set of the energy absorption structure; A multi-dimensional collision response prediction model is constructed based on a deep learning framework and the multi-dimensional collision mechanical response data set of the energy-absorbing structure, wherein the multi-dimensional collision response prediction model includes: an image feature extraction network, a curve feature extraction network and a feature vector prediction network; Defining a pattern stability constraint based on a decoding result of the image feature extraction network, defining a load stability constraint based on a decoding result of the curve feature extraction network, and defining an index matching constraint based on an output result of the multi-dimensional collision response prediction model; A multi-objective crashworthiness optimization model of an energy-absorbing structure is constructed based on the mode stability constraint, the load stability constraint, and the index matching constraint, and a non-dominated sorting genetic algorithm is used to solve the model to obtain a Pareto solution set; The optimal feasible solution of the Pareto solution set is solved based on the entropy-weighted VIKOR multi-attribute decision-making method, and the optimal feasible solution is used as the optimization result of the crashworthiness performance of the target energy-absorbing structure.

2. The method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure according to claim 1, characterized in that: The multi-dimensional collision mechanical response data set of the energy-absorbing structure obtained by numerical simulation based on the simulation parameters includes: Perform impact tests and corresponding numerical simulations on the target energy absorbing structure according to the type and simulation parameters of the target energy absorbing structure, and calibrate the numerical simulation model; The experimental design is carried out based on the numerical simulation model after experimental benchmarking, and a multi-dimensional collision mechanical response data set of the energy-absorbing structure is generated by integrating experiment and simulation.

3. The method for optimizing the crashworthiness of a rail vehicle energy absorption structure according to claim 1, characterized in that: The multi-dimensional collision response prediction model is constructed based on the deep learning framework and the multi-dimensional collision mechanical response data set of the energy absorbing structure, including: The multi-dimensional collision mechanical response data set of the energy-absorbing structure is randomly divided into a training set, a validation set and a test set in a ratio of 8:1:1; A deep learning model is constructed through a deep learning framework, and the deep learning model is trained using training sets, validation sets, and test sets to obtain a multi-dimensional collision response prediction model.

4. The method for optimizing the crashworthiness of a rail vehicle energy absorption structure according to claim 1, characterized in that: The image feature extraction network is used to construct an autoencoder network with stacked convolutional layers as the main architecture, and extract the deformation mode image feature vector, wherein the deformation mode image feature vector is: ; The curve feature extraction network is used to construct an autoencoder network with stacked long and short-term memory layers as the main structure, and extract the load characteristic curve feature vector, wherein the load characteristic curve feature vector is: ; The feature vector prediction network is used to construct a vector prediction network with a fully connected layer as the main body, and fuses the deformation model image feature vector, the load characteristic curve feature vector and the crashworthiness index feature vector directly extracted from the multi-dimensional collision mechanics response data set of the energy absorbing structure into an output vector.

5. The method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure according to claim 1, characterized in that: The decoding result based on the image feature extraction network defines a mode stability constraint, comprising: After obtaining the prediction result of the output vector through the multi-dimensional collision response prediction model, the feature vector part of the deformation model image is input into the image feature extraction network for decoding to obtain the pattern image prediction result; The mode stability constraint is defined based on the mode image prediction result. The stability judgment basis is different according to the type of target energy-absorbing structure. When the mode is stable, the constraint value is returned to 0, and when the mode is unstable, the constraint value is returned to 1.

6. The method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure according to claim 1, characterized in that: The method of defining a load stability constraint based on a decoding result of the curve feature extraction network includes: After obtaining the prediction result of the output vector through the multi-dimensional collision response prediction model, the characteristic vector part of the load characteristic curve is input into the curve feature extraction network for decoding to obtain the load curve prediction result; The load stability constraint is defined based on the load curve prediction result, where the load stability constraint satisfies the following relationship: ； In the formula, S 2 represents the load stability constraint value, θ is the preset constraint value of this indicator, N represents the number of sampling points of the load curve, Fi represents the load value at any sampling time, Represents the average value of the load curve. , the return constraint value is 0, otherwise the return constraint value is 1.

7. The method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure according to claim 1, characterized in that: The defining of index matching constraints based on the output results of the multi-dimensional collision response prediction model includes: The crashworthiness index prediction result is obtained according to the direct output of the crashworthiness index vector part, and the index matching constraint is defined based on the crashworthiness index prediction result. According to the difference in the target energy absorption structure type, the matching judgment basis is different. When the matching meets the requirements, the constraint value returned is 0, and when the mode is unstable, the constraint value returned is 1.

8. The method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure according to claim 1, characterized in that: The multi-objective crashworthiness optimization model of the energy absorption structure satisfies the following relationship: ； Where, Obj i represents the optimization goal, V stands for Structural parameters, VL and VU represent the upper and lower boundaries of the structural parameters respectively, P, PL and PU represent the crashworthiness index and its upper and lower boundaries respectively, represents the mode stability constraint, Represents the load fluctuation constraint.

9. The method for optimizing the crashworthiness of a rail vehicle energy-absorbing structure according to claim 1, characterized in that: The entropy-weighted VIKOR multi-attribute decision-making method is used to solve the optimal feasible solution of the Pareto solution set, including: S1. Construct a standardized solution set, which satisfies the following relationship: ； In the formula, x ij represents the original feasible solution, z ij represents the standardized feasible solution; S2. Determine the positive ideal solution and the negative ideal solution in the standardized solution set. The determination method satisfies the following relationship: ； In the formula, Z + represents the set of positive ideal solutions, Z i + Representative i The positive ideal solution of the target, Z - represents the set of negative ideal solutions, Z i - Representative i Negative ideal solution of a target; S3. Calculate the weight of each optimization target based on the entropy value, and the calculation includes: The contribution of the calculation target in the feasible solution satisfies the following relationship: ； In the formula, w ij represents the contribution in the feasible solution, x ij represents the original feasible solution; Calculate the entropy value of the target, and its calculation satisfies the following relationship: ； In the formula, e j represents the entropy value, w ij represents the contribution in the feasible solution; Calculate the weight of the target, and its calculation satisfies the following relationship: ； In the formula, ω j represents the weight value, e j represents the entropy value; S4. Calculate the group contribution value and individual regret value of each feasible solution. The calculation of the group contribution value satisfies the following relationship: ； In the formula, S i represents the group contribution value, ω j Represents the weight value, Z j + Representative j The positive ideal solution of the target, Z j - Representative j The negative ideal solution of the target, z ij represents the standardized feasible solution; The calculation of its individual regret value satisfies the following relationship: ； In the formula, R i represents the individual regret value, ω j Represents the weight value, Z j + Representative j The positive ideal solution of the target, Z j - Representative j The negative ideal solution of the target, z ij represents the standardized feasible solution; S5. Calculate the contribution rate of each feasible solution, and the calculation satisfies the following relationship: ； In the formula, Q i represents the contribution rate, S + , S - Represent the group contribution values ​​of positive ideal solution and negative ideal solution respectively, R + , R - Represent the individual regret values ​​of positive ideal solutions and negative ideal solutions respectively. S6. Sort the contribution rates of feasible solutions, and obtain the optimal feasible solution in the Pareto solution set based on the sorting results.

10. A system for optimizing the crashworthiness of a rail vehicle energy-absorbing structure, characterized in that: Including processor and memory; Memory, used to store computer programs; A processor, for implementing the method steps described in any one of claims 1 to 9 when executing a program stored in a memory.

Images