Guardrail collision performance virtual impact simulation test method and system
By combining multi-scale finite element models and adaptive solution algorithms with cloud computing, we can achieve efficient and accurate evaluation of guardrail collision performance, solving the problems of high cost and large error in traditional methods, and providing digital guardrail design support.
Patent Information
- Application Number
- CN202511448446.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-11
- Publication Date
- 2026-01-23
AI Technical Summary
Traditional guardrail collision performance assessment relies on real vehicle collision tests, which are costly and prone to large errors. Existing simulation methods have simplified modeling, resulting in high errors and low solution efficiency. The single safety evaluation method leads to a high misjudgment rate, making it difficult to meet the high standards of modern intelligent manufacturing.
A multi-scale finite element model combined with explicit dynamic collision control equations is adopted. The solution process is controlled by an energy-deformation dual-threshold adaptive algorithm, and a multi-dimensional safety evaluation system is integrated, including displacement, acceleration and energy data. Parallel computing and optimization design are performed using a cloud computing platform.
It significantly reduces modeling errors, improves solution efficiency and accuracy, reduces costs, provides multi-dimensional safety evaluation, and supports digital certification and optimization of guardrail design.
Smart Images
Figure CN121389596A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of computer-aided engineering technology, and in particular to a method and system for virtual impact simulation testing of guardrail collision performance. Background Technology
[0002] Traditional guardrail collision performance assessment mainly relies on real-vehicle crash tests, with a single test costing over 500,000 yuan. It also faces three major technical bottlenecks: First, at the model construction level, existing simulation schemes excessively simplify connection structures to reduce computational complexity, reducing bolted nodes to rigid connections, resulting in force response errors as high as 25%. Second, at the solution efficiency level, fixed-time-step explicit dynamic algorithms struggle to adapt to the nonlinear abrupt changes in the collision process; when the structure enters the plastic deformation stage, the risk of solution divergence increases by 40% due to excessively large time steps. Third, at the safety evaluation level, current industry standards only consider the maximum displacement. For the criterion, the risk of secondary injury to occupants caused by acceleration impact is ignored. The simplified modeling of bolted connections by the LS-DYNA explicit solver leads to an anchoring force prediction error of up to 18.2%, and the misjudgment rate of safety levels using traditional single-index evaluation systems (relying solely on displacement) exceeds 30%. This invention specifically establishes a technical chain for evaluating material strain rate effects and multi-index fusion, compressing the error to within 5.8%. However, traditional methods generally suffer from low modeling efficiency, high computational resource requirements, and inability to directly guide design. Furthermore, traditional methods lack reliability assessment of design parameter fluctuations, making it difficult to meet the high standards of modern intelligent manufacturing. Summary of the Invention
[0003] In order to overcome the above-mentioned defects of the prior art and to achieve the above objectives, the present invention proposes the following technical solution: A virtual impact simulation testing method and system for guardrail collision performance, including: S1: Construct a multi-scale finite element model that includes a macroscopic framework and microscopic connectors; S2: Based on the multi-scale finite element model, an explicit dynamic collision control equation is established, and then an adaptive algorithm based on energy-deformation dual thresholds is used to dynamically control the solution process. S3: Based on the results output by the solution process, collect displacement, acceleration and energy data during the collision process, and then output the protection level through a multi-dimensional safety evaluation system.
[0004] The formula for calculating the size of the macroscopic frame unit in S1 is as follows: ; in, Based on the standard dimensions, For the material's yield strength, For elastic modulus, It is the geometric curvature.
[0005] The force transmission at the micro-connector nodes in S1 satisfies: ; For the transmission coefficient, For the nodal displacement, the micro-connector sub-model is coupled to the macro-mesh boundary through displacement interpolation, satisfying: , The shape function and the transfer coefficient K are calibrated using the ASTM E8 standard quasi-static tensile test: three sets of standard specimens are prepared and loaded with a displacement of... Record force value , fitting Take the average value; the error should be controlled within ±5%.
[0006] The time step adjustment logic in S2 is as follows: ; Thresholds of 0.3 and 0.9 were determined based on sensitivity analysis of 50 collision cases, where the plastic deformation error exceeded 5%. ,in This represents the yield strain threshold of the material. For Q235 steel, For concrete (such as C40). The value is determined by the inflection point of the plastic segment of the stress-strain curve. The 50 case studies cover steel and concrete materials. The error analysis report shows that thresholds of 0.3 and 0.9 make the plastic deformation error <5%.
[0007] The formula for calculating the safety index in S3 is as follows: ; in, The displacement penetration index. The acceleration damage index, Energy absorption rate, weighting coefficient Determined using the entropy weight method: a) Construct an indicator matrix for 100 case studies ; b) Standardization process: ; c) Calculate information entropy: ; d) Determine the weights: ,have to , For standardized values, An index of 1.5 corresponds to the occupant injury risk curve.
[0008] The security levels are classified as follows: Level 1 High Protection; Level 2, Medium Protection; Level 3 basic protection; Level 4 is not up to standard.
[0009] The system includes: Finite element model module: Configure material constitutive relations and mesh generation parameters; Solver engine module: performs dual-threshold adaptive explicit dynamics calculations, parallel distribution algorithm: input task queue, MPI protocol to divide the mesh domain, master node monitors CPU utilization, and dynamically adjusts node resources when >85%; Evaluation output module: Generates collision process animation and safety level report.
[0010] The model building module also includes: Parametric modeling unit: Used to receive user input parameters such as guardrail cross-section type, post spacing, and material grade, and automatically generate corresponding macroscopic and microscopic finite element models; Material database unit: Pre-stores constitutive model parameters for common guardrail materials such as Q235 steel and C40 concrete. The strain rate effect coefficient is based on CEB-FIP Model Code 2010, and the formula is as follows: ,in It supports one-click access.
[0011] The solver engine module is deployed on a cloud simulation platform, supporting parallel task distribution and elastic scheduling of computing resources. Parallel computing is implemented based on the MPI protocol, and the elastic scheduling is triggered when the CPU utilization rate is >85%. It also has a built-in standardized collision scenario library that conforms to international standards such as EN 1317 and NCHRP 350.
[0012] The evaluation output module also includes: Optimized design unit: based on safety index Using the guardrail size and material as design variables, and employing a second-order polynomial response surface model for automatic optimization iteration, the response surface model construction process is as follows: 1. Design variable: Guardrail thickness Column spacing Material grade ; 2. Latin hypercube sampling: 10k sample points (spatial uniformity >0.85); 3. Second-order polynomial response surface: ; 4. Optimization Algorithm: BFGS Quasi-Newton Iterative Method, Convergence Condition Repeat the process three times, then output the optimal design solution. Reliability Analysis Unit: Performs Monte Carlo sampling on key input parameters, conducts uncertainty analysis, and outputs reliability indicators for the protection level, such as... The probability, sampling variables include , Thickness tolerance ±0.2mm, sample size 1000.
[0013] The present invention has the following beneficial effects: 1. Through the collaborative construction of macroscopic shell and microscopic connecting component sub-models, combined with nonlinear force transmission formulas... This reduced the bolt connection response error from 22.7% using traditional methods to 6.3%. In the case of a corrugated beam guardrail collision, the deviation between the predicted anchorage failure force of 68.2 kN and the physical test value of 72.4 kN was controlled within 5.8%.
[0014] 2. An adaptive algorithm with dual thresholds for energy and deformation is used to dynamically control the solution process. When the deformation energy ratio is detected... (e.g., during the concrete crushing stage) the step size is automatically reduced by 20% to avoid a 75% risk of solution divergence; in the low energy change stage... Increasing the step size improves speed. By combining parallel computing with a cloud computing platform, the time for a 10-second collision simulation was reduced from 4.2 hours to 2.6 hours, an efficiency improvement of 37%.
[0015] 3. Fusion displacement penetration index Acceleration damage index Energy absorption rate The three indicators, through a weighted safety index Dynamically determine the protection level. In bridge concrete guardrail testing, the system accurately identifies... The high acceleration damage risk was corrected, and the traditional result, which was misjudged as level two, was revised to level three, which is 100% consistent with the conclusion of the real vehicle test.
[0016] 4. By integrating parametric modeling with a cloud platform, the model preparation time, which originally required several days, is reduced to hours. The optimized design function can automatically find the best guardrail design scheme, reducing material costs by up to 15% while ensuring safety. The uncertainty analysis module elevates simulation from deterministic analysis to probabilistic design, outputting the safety and reliability of the product, providing a core tool for the digital certification and lifecycle management of guardrails. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the method flow proposed in this invention. Detailed Implementation
[0018] The following will refer to the accompanying drawings of the embodiments of the present invention. Figure 1 The three-stage architecture shown clearly and completely describes the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0019] Example 1: Passenger Vehicle Collision Simulation of Corrugated Beam Guardrail on Highway 1. Test Scenario and Initial Condition Settings According to Article 5.2.3 of JTG D81-2017 "Design Specification for Highway Traffic Safety Facilities" regarding guardrail collision protection level requirements, a Q235 steel corrugated beam guardrail system meeting Class A protection standards was selected as the test object. The guardrail structural parameters are: beam height 750mm, corrugation thickness 4.5mm, post spacing 4m, and burial depth 1.4m. A medium-sized passenger car (total mass 1500kg) was selected as the collision vehicle and, according to the specification, was to collide perpendicularly with the center of the guardrail at a speed of 60km / h (16.7m / s). The initial kinetic energy was calculated as follows: ; 2. Construction of Multi-Scale Finite Element Model A multi-scale modeling approach combining macroscopic and microscopic perspectives is employed. (1) Macroscopic framework modeling: Based on the material constitutive relation and geometric characteristics, S4R shell elements are used for discretization. The element size is dynamically determined according to the curvature adaptive formula: A total of 218,540 finite element elements were generated, including 85,420 beam and slab elements, 73,650 column elements, and 59,470 foundation elements.
[0020] (2) Microscopic connection modeling: A refined sub-model is established for the bolted connection structure, using a nonlinear force transmission model: ; When the nodal displacement δ = 0.8 mm, the calculated force transmission value is: A total of 35,680 connector sub-model units were established to realize the mechanical transmission at both the macroscopic and microscopic scales.
[0021] Taking Q235 steel bolts as an example, the force transmission value was measured by the ASTM E8 standard quasi-static tensile test when the displacement δ=0.5mm. Calculated The typical range is recommended to be 8–12 kN / mm0.7.
[0022] 3. Adaptive explicit dynamics solution process Establish the collision control equations: Where M is the mass matrix, C is the damping matrix, and K is the stiffness matrix, which is solved using the explicit central difference method.
[0023] Employing an energy-deformation dual threshold control algorithm: The initial time step was set to 1.2 μs to satisfy the CFL stability condition. The deformation energy ratio was detected to exceed the limit at t=32.7ms: ; Calculated according to claim 4, Q235 steel ; Triggering step size adjustment mechanism: ; The total computation time was 2.6 hours, a 37% reduction compared to the fixed step size method (4.2 hours).
[0024] 4. Multi-dimensional security evaluation and verification Extracting collision response data: (1) Displacement response: maximum dynamic displacement Less than the safety threshold ; Displacement penetration index: ; (2) Acceleration response: Collect collision acceleration time history data and calculate the acceleration damage index: ; (3) Energy response: Total absorbed energy 154.8 kJ, energy absorption rate: ; Comprehensive safety index calculation: ;
[0025] It is classified as Level 2 (medium protection) safety level.
[0026] Table 1 Comparison of simulation and real vehicle test results.
[0027]
[0028] As shown in Table 1, the method of this invention (the traditional method refers to the simplified beam element method built into LS-DYNA R11.0, with MPC rigid constraints used for bolt connections) demonstrates a high degree of consistency with the actual vehicle test in key evaluation indicators: the predicted maximum displacement D_max is 312mm (actual vehicle 305mm, relative error only 2.3%), the anchoring force is 68.2kN (actual vehicle 72.4kN, error 5.8%), the peak acceleration is 28.5g (actual vehicle 29.2g, error 2.4%), and the safety level is accurately determined as Level II medium protection, fully verifying the accuracy advantages of multi-scale modeling and adaptive solution algorithms. In contrast, the traditional method suffers from significant deviation in anchoring force prediction (85.6kN, error up to 18.2%) due to the simplification of connecting parts, and misjudges the safety level as Level I, highlighting its insufficient reliability; at the same time, this invention reduces the computational cost to 48,000 yuan (90% lower than the actual vehicle cost), significantly better than the 62,000 yuan of the traditional method, efficiently supporting engineering decision-making. In summary, this solution achieves a breakthrough in cost and efficiency optimization while ensuring an assessment accuracy rate of over 94%, providing core digital support for guardrail safety certification.
[0029] Example 2: Simulation of Bus Collision with Bridge Concrete Guardrail 1. Test Scenario Setup Based on Article 4.3.2 of JTG D60-2015 "Specifications for Design of Highway Bridges and Culverts", an F-type concrete guardrail structure was adopted, with concrete grade C40 (standard compressive strength 40MPa). The colliding vehicle was a medium-sized passenger bus (gross mass 12,000kg), which collided at a speed of 80km / h (22.2m / s) at a 15° angle. Initial kinetic energy: ; 2. Dynamic Material Modeling Dynamic constitutive relation of concrete considering strain rate effect: ; Unit size calculation: ; 3. Critical Event Response Analysis The critical event of concrete crushing was detected at t=41.6ms: Deformation energy ratio: ;
[0030] Trigger step size adjustment: ; 4. Multi-dimensional security evaluation (1) Displacement response: ; (2) Acceleration damage index: ; (3) Energy absorption rate: ; Security level calculation: ;
[0031] It is classified as Level 3 (basic protection).
[0032] Table 2 Comparison of Security Level Determination
[0033] As shown in Table 2, the method of the present invention demonstrates significant evaluation advantages in the simulation of bus collisions with concrete guardrails on bridges: the maximum displacement D_max predicted by the multi-index method of the present invention is 285mm (close to the actual vehicle's 280mm, with a deviation of only 1.8%), the acceleration damage index I_a is 152.4 (actual vehicle 155.2, error 1.8%), and the energy absorption rate E_r is 82.1% (actual vehicle 83.5%, error 1.7%). It also accurately determines the safety level as Level III basic protection, which is 100% consistent with the conclusions of the actual vehicle test, verifying the comprehensiveness and reliability of the multi-dimensional safety evaluation system (integrating displacement, acceleration, and energy indicators). In contrast, the traditional single-index method relies solely on displacement data (D_max=262mm), ignoring the risk of acceleration damage, resulting in a misclassification of the safety level as Level II (with a "no" consistency), with a deviation rate as high as 33%, highlighting the limitations of the traditional method. This solution, through a scientifically weighted safety index Ps (calculated value 0.760), achieves a high degree of consistency with the actual vehicle under complex collision scenarios, providing accurate digital decision-making basis for guardrail safety design, and further strengthening the core value of this method in engineering certification.
[0034] Example 3: Collision Simulation of a Novel Glass Fiber Reinforced Polymer (GFRP) Guardrail 1. Test scenario and initial condition settings (extended material properties and collision parameters) According to the ASTM D790-17 composite material testing standard, the following scenario is set up: Guardrail structure: Triple-wave GFRP beam slab (density 1.8g / cm³, fiber content 60%), post spacing 3.5m. Material constitutive model: ; in (Damage softening coefficient); Vehicle involved in the collision: SUV (gross weight 2000kg); Collision conditions: Speed: 90km / h (25m / s), Angle: 25° oblique collision; Initial kinetic energy: ; Acceptance criteria: Refer to the TL3 protection level requirements in EN 1317-2010 (maximum acceleration <40g).
[0035] 2. Multi-scale model construction (1) Special processing of macro-grids Unit size formula adjustment: ; in For GFRP board thickness, As the reference thickness; Mesh properties: SC8R continuous shell element (for anisotropic materials) was used, generating a total of 185,420 elements; (2) Innovation in micro-connector modeling Force transmission model of GFRP bolt connection: ;
[0036] Temperature compensation coefficient α = 1.2 × 10−4 / ℃ (measured value) K-value calibration process (supplementary details of ASTM D3039 standard): Sample preparation: 5 sets of GFRP bolt connectors (16mm diameter) Environmental conditions: 23±2℃ / 50±5%RH (72h humidification) Quasi-static testing: Loading rate: 2mm / min Data collection points: δ=0.2, 0.5, 0.8mm Curve fitting: The least squares method was used to fit the data, and K was found to be 15.2 kN / mm. 0 · 7 (R²=0.983).
[0037] 3. Key processes for dynamic solution (1) Strain rate effect modeling GFRP dynamic strength model: ; Peak strain rate of impact .
[0038] (2) Critical event response Dual threshold regulation is triggered at t=38.2ms: ; Step size dynamic adjustment: (Dual over-limit enhanced control); 4. Multi-dimensional safety assessment (in-depth biomechanical analysis) (1) Key Response Data Acquisition
[0039] (2) Specific verification of weighting coefficients Added 50 sets of GFRP collision cases for entropy weight method review:
[0040] (3) Comprehensive determination of security level ; Determined as Level 2 (medium protection), compliant with EN 1317 TL3 requirements. 5. Verification and Comparative Analysis (Extended Experimental Details) (1) Vehicle test configuration Test site: CATARC Proving Ground Instrument setup: Guardrail displacement: 12 LVDT sensors (±0.1mm accuracy) Vehicle acceleration: 3-axis ICL-3A sensor (1000Hz sampling) Temperature monitoring: 8 infrared thermal imagers (FLIR A655sc) (2) Result Comparison and Error Analysis
[0041] (3) Biomechanical verification Occupant injury risk analysis (compliant with ISO 13232 standard): Head injury index HIC15: Simulated value 412 vs. experimental value 428 (<1000 safety limit) Chest compression: Simulated value 32.7mm vs. experimental value 34.1mm (<42mm limit) Leg stress: Simulated value 5.2kN vs. experimental value 5.4kN (<10kN limit) 6. Quantification of technical benefits 1. Computational efficiency: Simulation time: 3.8 hours (42% longer than steel guardrail) Cloud computing resource consumption: 128 core hours (cost $156) 2. Economic benefits:
[0042] 3. Reliability verification: Monte Carlo analysis (sample size 5000) P(Ps≥0.8)=96.7% (material parameter fluctuation ±15%) Environmental adaptability: Passed verification in the temperature range of -20℃ to 60℃.
[0043] This solution achieves a revolutionary breakthrough in virtual impact simulation testing of guardrail collision performance through a three-stage innovative architecture: In the multi-scale modeling stage, it adopts collaborative modeling of the macroscopic shell and microscopic connectors (e.g., unit size calculation...). Among them, the force transmission model of micro-connectors coefficient Certified by ASTM standard tests (e.g., GFRP materials) To ensure modeling accuracy error is less than 6%, in the adaptive solution phase, the energy-deformation dual threshold algorithm is used. The control logic dynamically adjusts the time step, reducing computation time by 37%, and improves efficiency by combining cloud platform parallel computing (MPI protocol); in the multi-dimensional evaluation stage, the displacement penetration index is integrated. Acceleration damage index and energy absorption rate Safety Index (weight) Verified using the entropy weight method), the output dynamic protection level (such as the GFRP guardrail in Example 3) is shown. (It is determined to be a level 2 protection system), and the verification shows that the safety level accuracy is >96%. In terms of economy, it replaces the actual vehicle test, reducing costs by 90%, and provides a new digital certification standard for intelligent transportation infrastructure, supporting the optimization design of the entire life cycle.
[0044] In the application, several formulas are calculated by removing dimensions and taking their numerical values. The formulas are established by collecting a large amount of data and simulating the most recent real situation. Some coefficients or weights in the formulas are set by those skilled in the art according to the actual situation, so they will not be elaborated here.
[0045] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.
[0046] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A virtual impact simulation test method for the collision performance of guardrails, characterized in that, include: S1: Construct a multi-scale finite element model that includes a macroscopic framework and microscopic connectors; S2: Based on the multi-scale finite element model, an explicit dynamic collision control equation is established, and then an adaptive algorithm based on energy-deformation dual thresholds is used to dynamically control the solution process. S3: Based on the results output by the solution process, collect displacement, acceleration and energy data during the collision process, and then output the protection level through a multi-dimensional safety evaluation system.
2. The virtual impact simulation test method for the collision performance of guardrails according to claim 1, characterized in that, The formula for calculating the size of the macroscopic frame unit in S1 is as follows: ； in, Based on the standard dimensions, For the material's yield strength, For elastic modulus, It is the geometric curvature.
3. The virtual impact simulation test method for the collision performance of guardrails according to claim 1, characterized in that, The force transmission at the micro-connector nodes in S1 satisfies: ； For the transmission coefficient, For the nodal displacement, the micro-connector sub-model is coupled to the macro-mesh boundary through displacement interpolation, satisfying: , The shape function and the transfer coefficient K are calibrated using the ASTM E8 standard quasi-static tensile test: three sets of standard specimens are prepared and loaded with a displacement of... Record force value , fitting Take the average value; the error should be controlled within ±5%.
4. The virtual impact simulation test method for the collision performance of guardrails according to claim 1, characterized in that, The time step adjustment logic in S2 is as follows: ； Thresholds of 0.3 and 0.9 were determined based on sensitivity analysis of 50 collision cases. At that time, the plastic deformation error exceeded 5%; in For Q235 steel, For concrete (such as C40). The value is determined by the inflection point of the plastic segment of the stress-strain curve. The 50 case studies cover steel and concrete materials. The error analysis report shows that thresholds of 0.3 and 0.9 make the plastic deformation error <5%.
5. The virtual impact simulation test method for the collision performance of guardrails according to claim 1, characterized in that, The formula for calculating the safety index in S3 is as follows: ； in, The displacement penetration index. The acceleration damage index, Energy absorption rate, weighting coefficient Determined using the entropy weight method: a) Construct an indicator matrix for 100 case studies ; b) Standardization process: ; c) Calculate information entropy: ; d) Determine the weights: ,have to , For standardized values, An index of 1.5 corresponds to the occupant injury risk curve, with weights... It can be adjusted based on the case type, such as oblique collision. Slightly higher.
6. The virtual impact simulation test method for the collision performance of guardrails according to claim 5, characterized in that, The security levels are classified as follows: Level 1 High Protection; Level 2, Medium Protection; Level 3 basic protection; Level 4 is not up to standard.
7. A virtual impact simulation testing system for guardrail collision performance, implemented according to any one of claims 1 to 6, characterized in that, include: Finite element model module: Configure material constitutive relations and mesh generation parameters; Solver engine module: performs dual-threshold adaptive explicit dynamics calculations, parallel distribution algorithm: input task queue, MPI protocol to divide the mesh domain, master node monitors CPU utilization, and dynamically adjusts node resources when >85%; Evaluation output module: Generates collision process animation and safety level report.
8. The virtual impact simulation testing system for guardrail collision performance according to claim 7, characterized in that, The model building module also includes: Parametric modeling unit: Used to receive user input parameters such as guardrail cross-section type, post spacing, and material grade, and automatically generate corresponding macroscopic and microscopic finite element models; Material database unit: Pre-stores constitutive model parameters for common guardrail materials such as Q235 steel and C40 concrete. The strain rate effect coefficient is based on CEB-FIP Model Code 2010, and the formula is as follows: ,in It supports one-click access.
9. The virtual impact simulation testing system for guardrail collision performance according to claim 7, characterized in that, The solver engine module is deployed on a cloud simulation platform, supporting parallel task distribution and elastic scheduling of computing resources. Parallel computing is implemented based on the MPI protocol, and the elastic scheduling is triggered when the CPU utilization rate is >85%. It also has a built-in standardized collision scenario library that conforms to international standards such as EN 1317 and NCHRP 350.
10. A virtual impact simulation testing system for the collision performance of a guardrail according to claim 7, characterized in that, The evaluation output module also includes: Optimized design unit: based on safety index Using the guardrail size and material as design variables, and employing a second-order polynomial response surface model for automatic optimization iteration, the response surface model construction process is as follows:
1. Design variable: Guardrail thickness Column spacing Material grade ; 2. Latin hypercube sampling: 10k sample points (spatial uniformity >0.85); 3. Second-order polynomial response surface: ; 4. Optimization Algorithm: BFGS Quasi-Newton Iterative Method, Convergence Condition Repeat the process three times, then output the optimal design solution. Reliability Analysis Unit: Performs Monte Carlo sampling on key input parameters, conducts uncertainty analysis, and outputs reliability indicators for the protection level, such as... The probability, sampling variables include , Thickness tolerance ±0.2mm, sample size 1000.