Method and System for Optimizing the Design of Movable Steel Guardrails Based on the Euler Angles of Trucks

Abstract

The present invention provides a method for optimizing the design of a movable steel guardrail based on the Euler angles of a freight vehicle, including: S1, constructing a finite element model of the vehicle and a finite element model of the movable steel guardrail, and conducting a collision simulation test on the vehicle and the movable steel guardrail; S2, obtaining the collision result data generated by the collision simulation experiment; S3, and using the trained RBF neural network to predict the value of the Euler angle of the vehicle; S4, taking the guardrail height H and the anchoring spacing L of the movable steel guardrail as design variables, the vehicle centroid acceleration and the vehicle guiding and exiting frame as limiting factors, and the minimum value of the predicted value of the Euler angle of the vehicle predicted by the RBF neural network as the objective, constructing a multi-objective optimization model of the movable steel guardrail; S5, using the multi-objective particle swarm algorithm to solve the multi-objective optimization model of the movable steel guardrail, drawing the boundary diagram of the Pareto optimal solution set, and determining the guardrail height H and the anchoring spacing L. The present invention effectively improves the protection efficiency of the movable steel guardrail.

Description

Technical Field

[0001] The invention relates to the technical field of highway guardrail design, and in particular to a method and system for optimizing the design of a movable steel guardrail based on Euler angles of trucks. Background Art

[0002] With the continuous advancement of highway construction and the gradual improvement of infrastructure, highway reconstruction and expansion projects play a vital role in the construction of modern transportation infrastructure. However, in this process, since the traffic organization and implementation have changed the traffic environment of existing highways, it is necessary to set up temporary movable steel guardrails between the construction control area and the traffic lane to guide and protect the passing vehicles. During the implementation of the project, the collision problem between vehicles and movable steel guardrails has received more and more attention. This type of collision not only has a negative impact on traffic fluency and safety, but also in actual engineering, it may cause vehicles (especially trucks) to collide with movable steel guardrails and cause dangerous situations such as riding, rolling over, and climbing over, posing a potential threat to road traffic safety, and may even cause serious traffic accidents and personal injuries. Therefore, how to improve the protective effect of movable steel guardrails and reduce the danger of vehicles (especially trucks) after collision has become one of the urgent problems to be solved in the engineering field.

[0003] Domestic and foreign scholars have conducted a lot of research on the problem of vehicle-guardrail collision. The current research mainly focuses on two directions: one is to improve and optimize the guardrail structure through real vehicle collision experiments to improve its impact resistance; the other is to deeply analyze the collision process through simulation and real vehicle tests to provide a scientific basis for guardrail design.

[0004] On the one hand, scholars have improved the guardrail structure by optimizing the material and structural design of the guardrail to improve its energy absorption capacity and impact resistance. For example, the use of high-strength materials and reasonable cross-section design can absorb more energy during a collision and reduce the impact force during the collision. Bank et al. and Zhang et al. focused on the material parameters of the guardrail. The results showed that the material properties responded to the energy absorption after the vehicle collided with the guardrail. Jiao Chiyu et al. proposed to use a new type of aluminum alloy anti-collision guardrail to replace the traditional steel guardrail, and analyzed the bearing capacity and safety performance of the guardrail under vehicle collision conditions. In addition, some studies have also focused on the influence of the geometric shape of the guardrail on the protection effect, and improved its performance during a collision by adjusting the height, inclination angle and other parameters of the guardrail. For example, Yin et al. designed an η-type corrugated beam guardrail, which can prevent tire hanging and improve the safety performance in vehicle collisions.

[0005] On the other hand, the research on the collision process mainly focuses on two approaches: numerical simulation and full-scale vehicle tests. Numerical simulation usually adopts the finite element analysis method. By establishing models of vehicles and guardrails, the collision process is simulated to analyze the stress, deformation, etc. during the collision, providing theoretical support for the guardrail design. Hou et al., Zhou Wei et al., Yin et al., and Lu et al. carried out finite element modeling and simulation on the vehicle-guardrail collision system, and combined methods such as radial basis function (RBF) and genetic algorithm (GA) to optimize the design of the guardrail and improve its safety protection performance. Yang et al. carried out simulation according to the NCHRP 350 standard using LS-DYNA, and systematically evaluated and optimized the vehicle acceleration, post-collision trajectory, energy absorption capacity, and deformation of the guardrail during the collision. Full-scale vehicle tests are conducted through controlled tests in the actual road environment to obtain real collision data, verify the results of numerical simulation, and visually evaluate the performance of the guardrail. Lei Zhengbao et al. comprehensively analyzed the collision safety of a new type of flexible guardrail by establishing finite element simulation models of vehicles and guardrails and combining multiple full-scale vehicle collision tests. Atahan et al. and Ozcanan et al. developed and optimized bridge guardrails, H1W4, and H2W4 type guardrails, etc. based on full-scale vehicle collision data and combined with simulation and other means.

[0006] From the above analysis of the existing technologies, it can be seen that the current research mainly uses the finite element simulation method to carry out evaluation and design parameter optimization research on corrugated beam guardrails, concrete guardrails, bridge guardrails, etc. under normal highway traffic conditions, and has achieved certain research results, providing strong guarantees for highway traffic safety.

[0007] However, it can also be seen that there is less research on the evaluation and optimization of the protection performance of movable steel guardrails during the reconstruction and expansion construction of expressways. There are mainly the following two deficiencies: First, most of the existing research uses small vehicles for finite element simulation, ignoring the response behavior of large vehicles after colliding with movable steel guardrails; second, starting from indicators such as vehicle centroid acceleration and guiding exit frames, less consideration is given to the overturning, straddling, etc. of vehicles after colliding with movable steel guardrails, resulting in the inability to meet the current traffic safety requirements for the reconstruction and expansion of expressways. Summary of the Invention

[0008] The present invention provides a method and system for optimizing the design of a movable steel guardrail based on the Euler angles of trucks to solve the technical problem that the existing design of movable steel guardrails gives less consideration to the overturning, straddling, etc. of vehicles after colliding with movable steel guardrails, resulting in the inability to meet the current traffic safety requirements for the reconstruction and expansion of expressways.

[0009] The technical solutions provided by the present invention are as follows:

[0010] An object of the present invention is to provide a method for optimizing the design of a movable steel guardrail based on the Euler angles of trucks, and the method includes the following method steps:

[0011] S1. Construct a finite element model of the vehicle and a finite element model of the movable steel guardrail. Using the guardrail height H and the anchoring spacing L of the movable steel guardrail as variables, conduct a collision simulation test on the vehicle and the movable steel guardrail;

[0012] S2. Obtain the collision result data generated from the collision simulation experiment of the vehicle and the movable steel guardrail. The collision result data includes: the vehicle's centroid acceleration after the vehicle collides with the movable steel guardrail, the vehicle's guiding exit frame parameters, and the vehicle's Euler angles;

[0013] S3. Using the guardrail height H and the anchoring spacing L of the movable steel guardrail as inputs, and the vehicle's centroid acceleration, the vehicle's guiding exit frame parameters, and the vehicle's Euler angles after the vehicle collides with the movable steel guardrail as outputs, use the RBF neural network to fit the vehicle's centroid acceleration, the vehicle's guiding exit frame parameters, and the vehicle's Euler angles, and use the trained RBF neural network to predict the value of the vehicle's Euler angles;

[0014] S4. Using the guardrail height H and the anchoring spacing L of the movable steel guardrail as design variables, the vehicle's centroid acceleration and the vehicle's guiding exit frame as limiting factors, and the minimum value of the predicted value of the vehicle's Euler angles predicted by the RBF neural network as the objective, construct a multi-objective optimization model for the movable steel guardrail;

[0015] S5. Use the multi-objective particle swarm algorithm to solve the multi-objective optimization model of the movable steel guardrail, obtain the Pareto optimal solution set of the multi-objective optimization model of the movable steel guardrail, and draw the boundary diagram of the Pareto optimal solution set;

[0016] Determine the guardrail height H and the anchoring spacing L according to the boundary diagram of the Pareto optimal solution set.

[0017] In a preferred embodiment, in step S2, the vehicle's Euler angles include: the heading angle Y, the pitch angle X, and the roll angle Z.

[0018] In a preferred embodiment, in step S3, for the guardrail height H and the anchoring spacing L of the movable steel guardrail obtained in step S2, and the vehicle's centroid acceleration, the vehicle's guiding exit frame parameters, and the vehicle's Euler angles after the movable steel guardrail collides, use the Monte Carlo sampling method to expand the data.

[0019] In a preferred embodiment, select 60% of the data from the data expanded by the Monte Carlo sampling method as the training samples;

[0020] Taking the guardrail height H and the anchoring spacing L of the movable steel guardrail as inputs, inputting them into the RBF neural network, and taking the vehicle centroid acceleration, the vehicle guiding out-of-frame parameters, and the vehicle Euler angles after the vehicle collides with the movable steel guardrail as outputs, using the RBF neural network to fit the vehicle centroid acceleration, the vehicle guiding out-of-frame parameters, and the vehicle Euler angles, and using the trained RBF neural network to predict the value of the vehicle Euler angles.

[0021] In a preferred embodiment, 20% of the data is selected as the verification sample from the data expanded by the Monte Carlo sampling method;

[0022] Taking the guardrail height H and the anchoring spacing L of the movable steel guardrail as inputs, inputting them into the RBF neural network, and taking the vehicle centroid acceleration, the vehicle guiding out-of-frame parameters, and the vehicle Euler angles after the vehicle collides with the movable steel guardrail as outputs, to verify the fitting effect of the RBF neural network.

[0023] In a preferred embodiment, 20% of the data is selected as the test sample from the data expanded by the Monte Carlo sampling method;

[0024] Taking the guardrail height H and the anchoring spacing L of the movable steel guardrail as inputs, inputting them into the RBF neural network, and taking the vehicle centroid acceleration, the vehicle guiding out-of-frame parameters, and the vehicle Euler angles after the vehicle collides with the movable steel guardrail as outputs, to test the fitting effect of the RBF neural network.

[0025] In a preferred embodiment, in step S3, the predicted values of the vehicle Euler angles include: the predicted value of the heading angle Y, the predicted value of the pitch angle X, and the predicted value of the roll angle Z.

[0026] In a preferred embodiment, in step S4, the multi-objective optimization model of the movable steel guardrail is constructed by the following method:

[0027] Multi-objective optimization model of the movable steel guardrail: min[X(H,L),Y(H,L),Z(H,L)],

[0028] Boundary conditions: 30 ≤ L ≤ 70,

[0029] 600 ≤ H ≤ 1100,

[0030] ORA x ≤ 200,

[0031] ORA y ≤ 200,

[0032] DIS ≤ 8.1,

[0033] Wherein, X(H,L) is the relationship function between the predicted value of the pitch angle X of the vehicle's Euler angle and the guardrail height H and the anchoring spacing L of the movable steel guardrail, Y(H,L) is the relationship function between the predicted value of the heading angle Y of the vehicle's Euler angle and the guardrail height H and the anchoring spacing L of the movable steel guardrail, and Z(H,L) is the relationship function between the predicted value of the roll angle Z of the vehicle's Euler angle and the guardrail height H and the anchoring spacing L of the movable steel guardrail;

[0034] H is the guardrail height of the movable steel guardrail, and L is the anchoring spacing of the movable steel guardrail; ORA x is the lateral component of the vehicle's centroid acceleration, ORA y is the longitudinal component of the vehicle's centroid acceleration; DIS is the vehicle guiding exit frame parameter.

[0035] In a preferred embodiment, in step S5, according to the following method, the guardrail height H and the anchoring spacing L are determined based on the Pareto optimal solution set boundary diagram:

[0036] When, in the Pareto optimal solution set boundary diagram, the predicted value of the pitch angle X of the vehicle's Euler angle and the predicted value of the roll angle Z of the vehicle's Euler angle are the smallest, select the guardrail height H and the anchoring spacing L corresponding to the first optimal solution as the design parameters of the guardrail height H and the anchoring spacing L;

[0037] When, in the Pareto optimal solution set boundary diagram, the predicted value of the heading angle Y of the vehicle's Euler angle is the smallest, select the guardrail height H and the anchoring spacing L corresponding to the second optimal solution as the design parameters of the guardrail height H and the anchoring spacing L;

[0038] Wherein, the first optimal solution and the second optimal solution are the two ends of the Pareto optimal solution set boundary diagram.

[0039] Another object of the present invention is to provide a system for optimizing the design of a movable steel guardrail based on the Euler angle of a freight vehicle, and the system is used to execute the method for optimizing the design of a movable steel guardrail based on the Euler angle of a freight vehicle provided by the present invention.

[0040] The above technical solution of the present invention has at least the following beneficial effects compared with the prior art:

[0041] The present invention provides a method and a system for optimizing the design of a movable steel guardrail based on the Euler angle of a freight vehicle. By using the finite element simulation method, a collision simulation test is carried out for the scenario of a large freight vehicle colliding with a movable steel guardrail. A multi-objective optimization model of the movable steel guardrail is constructed with the optimal Euler angle as the goal and the guardrail height of the movable steel guardrail and the anchoring spacing of the movable steel guardrail as influencing factors. The optimal solution of the multi-objective optimization model of the movable steel guardrail is solved by combining the multi-objective particle swarm algorithm, thereby improving the protection efficiency of the movable steel guardrail.

[0042] The present invention provides a method and a system for optimizing the design of a movable steel guardrail based on the Euler angles of a freight truck. By using the Euler angles as the objectives of the multi-objective optimization model of the movable steel guardrail and the centroid acceleration and the guiding exit frame as the control indexes, it is more suitable for optimizing the design of the movable steel guardrail for large freight trucks.

[0043] The present invention provides a method and a system for optimizing the design of a movable steel guardrail based on the Euler angles of a freight truck. The results show that by increasing the height of the movable steel guardrail and reducing the anchoring spacing, the Euler angles can be improved and the incidence of dangerous situations can be reduced. This is because as the height of the guardrail increases, the performance indexes such as the strength and stiffness of the movable steel guardrail itself will also be improved, and the movable steel guardrail will have a higher energy absorption effect when the vehicle collides with it; a higher guardrail height can also have a better blocking effect on the vehicle, especially reducing the probability of large freight trucks crossing and rolling over. Reducing the anchoring spacing and increasing the number of anchorings can improve the stability of the movable steel guardrail and can play a better blocking effect when the vehicle collides with the movable steel guardrail.

[0044] The present invention provides a method and a system for optimizing the design of a movable steel guardrail based on the Euler angles of a freight truck. It is proposed to use the Euler angles as the evaluation indexes of the protection performance of the movable steel guardrail, the guardrail height and the anchoring spacing as the optimization factors, and the pitch angle, the heading angle and the roll angle after the vehicle collision as the optimization objectives to construct a multi-objective optimization model of the movable steel guardrail, which can effectively improve the protection efficiency of the movable steel guardrail.

[0045] The present invention provides a method and a system for optimizing the design of a movable steel guardrail based on the Euler angles of a freight truck. By using the method combining Monte Carlo and RBF neural network, the simulation test data is sampled and expanded to obtain a better fitting effect. The comparative analysis shows that while reducing the computational cost of the simulation test, it also ensures the accuracy of the cost function required for the optimization design of the movable steel guardrail.

[0046] The present invention provides a method and a system for optimizing the design of a movable steel guardrail based on the Euler angles of a freight truck. The Euler angles can effectively measure the safety state when the vehicle collides with the movable steel guardrail, and the results of the collision simulation test show that the guardrail height and the anchoring spacing have a greater impact on it. By solving the multi-objective optimization model of the movable steel guardrail, the structural design parameters of the movable steel guardrail under different optimization objectives are obtained. And by synthesizing each objective and the optimization design results of the collision of a small passenger car, the recommended values of a guardrail height of 750 mm and an anchoring spacing of 40 m in engineering practice are proposed, and the safety protection efficiency of the movable steel guardrail is improved. Brief Description of the Drawings

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0048] Figure 1 It is a flowchart of a method for optimizing the design of a movable steel guardrail based on the Euler angles of a freight vehicle in the present invention.

[0049] Figure 2 It is a schematic structural diagram of a vehicle finite element model in the present invention.

[0050] Figure 3 It is a schematic diagram of the geometric dimensions of the movable steel guardrail in the present invention.

[0051] Figure 4 It is a schematic structural diagram of a movable steel guardrail finite element model in the present invention.

[0052] Figure 5 It is a schematic diagram of the curve comparison of the longitudinal component of the vehicle's centroid acceleration after the vehicle collides with the movable steel guardrail and the real vehicle collision in the present invention.

[0053] Figure 6 It is a schematic diagram of the curve comparison of the longitudinal component of the vehicle's speed after the vehicle collides with the movable steel guardrail and the real vehicle collision in the present invention.

[0054] Figure 7 It is a schematic diagram of the vehicle guiding and exiting frame in the present invention.

[0055] Figure 8 It is a schematic diagram of the vehicle's Euler angles in the present invention.

[0056] Figure 9 It is a schematic diagram of the data distribution of the vehicle's centroid acceleration after the vehicle collides with the movable steel guardrail with different guardrail heights and anchoring spacings in the present invention.

[0057] Figure 10 It is a schematic diagram of the data distribution of the parameters of the vehicle guiding and exiting frame after the vehicle collides with the movable steel guardrail with different guardrail heights and anchoring spacings in the present invention.

[0058] Figure 11 It is a schematic diagram of the data distribution of the vehicle's Euler angles after the vehicle collides with the movable steel guardrail with different guardrail heights and anchoring spacings in the present invention.

[0059] Figure 12 It is a schematic diagram of the fitting effect of fitting the course angle using the RBF neural network in the present invention.

[0060] Figure 13It is a schematic diagram of the boundary graph of the Pareto optimal solution set of the present invention.

[0061] Figure 14 It is a schematic diagram of the simulation of the collision process between a vehicle and a movable steel guardrail after optimizing the guardrail height and anchoring spacing of the movable steel guardrail of the present invention. Specific embodiments

[0062] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0063] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the art to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Similarly, the terms such as "a", "an" or "the" do not denote a limitation of quantity, but mean that there is at least one. The terms such as "comprising" or "including" mean that the elements or objects appearing before this term cover the elements or objects listed after this term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0064] It should be noted that the "upper", "lower", "left", "right", "front", "rear", etc. used in the present invention are only used to represent relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0065] As Figure 1 shown, according to an embodiment of the present invention, a method for optimizing the design of a movable steel guardrail based on the Euler angles of a freight vehicle is provided, including the following method steps:

[0066] Step S1: Conduct a collision simulation test on the vehicle and the movable steel guardrail.

[0067] Construct a vehicle finite element model and a movable steel guardrail finite element model, and use the guardrail height H and anchoring spacing L of the movable steel guardrail as variables to conduct a collision simulation test on the vehicle and the movable steel guardrail.

[0068] In this embodiment, the constructed vehicle finite element model adopts the truck finite element model from the National Crash Analysis Center (NCAC) of the United States. The original mass of the truck finite element model is 7.887t. The model has been verified by NCAC tests and has a relatively accurate simulation effect.

[0069] According to the requirements of standards and specifications such as Highway Guardrail Safety Performance Evaluation Standard (JTG B05-01-2013) and Highway Traffic Safety Facilities Design Specification (JTG D81-2017), the collision mass of a truck should be 10 tons. In order to meet the collision conditions, the parameters of the original truck finite element model were modified, and the original vehicle counterweight was adjusted to 10 tons. The constructed vehicle finite element model is shown in the figure below. Figure 2 As shown, the parameters of the constructed vehicle finite element model are shown in Table 1.

[0070] Table 1 Vehicle finite element model parameters

[0071] Total mass (t) Vehicle length (mm) Vehicle width Vehicle height Number of units (pcs) 10 8505 2343 3320 22290

[0072] At present, the temporary guardrails used in domestic highway reconstruction and expansion projects are mainly movable steel guardrails, which are usually made of Q235 steel. In order to facilitate installation and protect the road surface, the movable steel guardrail is usually placed directly on the road surface and anchored to the road surface with rivets every 50m to isolate the construction area from the maintenance area and prevent passing vehicles from rushing into the construction area. According to the materials and structural dimensions of the movable steel guardrails actually used in the engineering field, the typical geometric dimensions and finite element model of the A-level movable steel guardrail are constructed.

[0073] In one embodiment, the geometric dimensions of the movable steel guardrail are as follows: Figure 3 As shown, Figure 3 (a) is the cross-sectional geometric dimensions of the movable steel guardrail, (b) is the front geometric dimensions of the movable steel guardrail, and (c) is the pin geometric dimensions of the movable steel guardrail.

[0074] The finite element model of the movable steel guardrail is as follows Figure 4 As shown, Figure 4 (d) is the cross-sectional structure of the movable steel guardrail, (e) is the front structure of the movable steel guardrail, and (f) is the structure of the pin of the movable steel guardrail.

[0075] According to an embodiment of the present invention, a collision simulation test between a vehicle and a movable steel guardrail is performed with the guardrail height H and the anchoring spacing L of the movable steel guardrail as variables.

[0076] Collision simulation tests of a vehicle with a movable steel guardrail require generating a sufficient number of collision result data. In this embodiment, to ensure the accuracy of the multi-objective optimization model of the movable steel guardrail (the multi-objective optimization model of the movable steel guardrail will be elaborated hereinafter), among the two variables of the guardrail height H and the anchoring spacing L for the optimization of the movable steel guardrail, the guardrail height H is 700, 720, 740, and 760 mm respectively, and the anchoring spacing L is 30, 40, 50, and 60 m respectively, with a total of 16 experimental sample points for finite element simulation. Substitute the K file constructed by the hyper mesh software into the Ls-Dyna simulation software to calculate the collision result data generated by the collision of the vehicle with the movable steel guardrail with different guardrail heights H and anchoring spacings L.

[0077] In one embodiment, taking the guardrail height H and the anchoring spacing L of the movable steel guardrail as variables, a collision simulation test of the vehicle with the movable steel guardrail is carried out. Although the simulation analysis has powerful functions, before using the finite element model for collision analysis, it is necessary to verify the reliability and accuracy of the vehicle finite element model and the movable steel guardrail finite element model.

[0078] Generally, the longitudinal speed of the vehicle's center of mass and the longitudinal component curve of the vehicle's center of mass acceleration in the simulation results are compared with the real vehicle collision curve to verify the accuracy of the finite element model. "Highway Guardrail Safety Performance Evaluation Standard" (JTGB05—01—2013), "Highway Traffic Safety Facilities Design Specification" (JTG D81—2017), etc. stipulate that when conducting real vehicle collision guardrail experiments, that is, the collision conditions and performance of Class A guardrails should meet the requirements of Table 2. Therefore, the collision speed for simulation is set to 60 km / h and the collision angle is 20°.

[0079] Table 2 Collision conditions and performance of Class A guardrails

[0080]

[0081] As Figure 5 and Figure 6 shown, the comparison results show that the simulation curve and the real vehicle collision curve are basically coincident in terms of profile, change trend, and the time when the peak value appears, which proves the effectiveness of the vehicle finite element model and the movable steel guardrail finite element model constructed in step S1.

[0082] Step S2: Obtain the collision result data generated by the collision simulation experiment of the vehicle with the movable steel guardrail.

[0083] According to an embodiment of the present invention, the collision result data includes: the vehicle centroid acceleration after the vehicle collides with the movable steel guardrail, the vehicle guiding and exiting frame parameters, and the vehicle Euler angles. The vehicle centroid acceleration, the vehicle guiding and exiting frame parameters, and the vehicle Euler angles after the vehicle collides with the movable steel guardrail are used as the safety performance evaluation criteria for the movable steel guardrail.

[0084] The vehicle centroid acceleration and the guiding and exiting frame are important indicators for evaluating the safety performance of movable steel guardrails in standards and specifications such as the "Highway Guardrail Safety Performance Evaluation Standard" (JTG B05—01—2013), providing strong guidance for aspects such as the design and installation of movable steel guardrails. However, since the process of a vehicle colliding with a movable steel guardrail is extremely complex.

[0085] Ideally, the collided vehicle will drive away normally, stop, etc. However, in complex scenarios such as on highways, especially during the reconstruction and expansion of highways where movable steel guardrails are used, it is very likely that a vehicle traveling at high speed will experience dangerous situations such as penetration, straddling, and rollover after colliding with the movable steel guardrail. The current relevant evaluation indicators are relatively insufficient in characterizing and evaluating the above-mentioned dangerous behaviors. Therefore, the present invention introduces the vehicle Euler angles into the safety performance evaluation criteria for movable steel guardrails.

[0086] Vehicle centroid acceleration

[0087] The main safety evaluation criteria for movable steel guardrails are the "Manual for Assessing Safety Hardware" (MASH) promulgated by the American Association of State Highway and Transportation Officials (AASHTO). The performance of movable steel guardrails is evaluated according to indicators such as "structural adequacy of safety characteristics", "behavior of the test vehicle after collision", and "risk of injury to the occupants of the collided vehicle". However, since it does not consider the actual traffic conditions in China, it has certain reference value for evaluating the performance of movable steel guardrails in China, but it cannot be fully applied.

[0088] The basis for evaluating the safety performance of movable steel guardrails in China is the "Highway Guardrail Safety Performance Evaluation Standard" (JTGB05—01—2013), which stipulates the safety performance evaluation indicators for movable steel guardrails, namely the blocking function, the buffering function, and the guiding function.

[0089] Among them, the buffering function selects the collision acceleration of the occupants of a small passenger car to measure the degree of injury to the human body. Since there is no dummy model in the collided vehicle model, the vehicle centroid acceleration is used instead. The standard requires that the lateral component ORA x and the longitudinal component ORA y of the vehicle centroid acceleration after collision shall not be greater than 200 m / s 2 2, and the lateral component ORA x and the longitudinal component ORA yIt can be calculated separately by the following methods:

[0090]

[0091]

[0092] In the formula, t i is the i-th moment at any 10 ms scale after the collision, n is the duration from the moment when the vehicle contacts the movable steel guardrail to the moment when it finally leaves the position of the movable steel guardrail, a x and a y are the longitudinal and lateral accelerations of the vehicle respectively.

[0093] Vehicle guiding exit frame

[0094] The vehicle guiding and exiting frame is a safe area entered by the vehicle after it collides with the movable steel guardrail and exits the last contact point between the vehicle and the original position of the guardrail. As Figure 7 shown, the width of the vehicle guiding and exiting frame is A and the length is B. After the vehicle collides with the movable steel guardrail along the vehicle driving trajectory, the vehicle enters the vehicle guiding and exiting frame after exiting the last contact point between the vehicle and the original position of the guardrail.

[0095] According to the regulations in the "Highway Guardrail Safety Performance Evaluation Standard" (JTG B05-01-2013), it is required that the vehicle cannot roll over after the collision, and when entering the vehicle guiding and exiting frame after exiting the last contact point between the vehicle and the original position of the guardrail, it shall not cross the edge line F on the side away from the movable steel guardrail of the vehicle guiding and exiting frame.

[0096] At the same time, in order to ensure the redirection of the vehicle in the vehicle guiding and exiting frame, the regulations in the "Highway Guardrail Safety Performance Evaluation Standard" (JTG B05-01-2013) require that when the vehicle exits the last contact point between the vehicle and the original position of the guardrail and enters the vehicle guiding and exiting frame, the distance DIS between the outermost wheel of the vehicle and the movable steel guardrail is less than or equal to the length B of the vehicle guiding and exiting frame.

[0097] In this embodiment, when the vehicle used is a truck, the calculation of the width A and length B of the vehicle guiding and exiting frame shall comply with the regulations in the "Highway Guardrail Safety Performance Evaluation Standard" (JTG B05-01-2013), and the calculation of the width A and length B of the vehicle guiding and exiting frame is shown in Table 3.

[0098] Table 3 Calculation of the width A and length B of the vehicle guiding and exiting frame

[0099] Collision vehicle type Collision speed (km / h) Collision angle (°) A (m) B (m) Truck 60 20 <![CDATA[4.4 + V W + 0.16V L > 20

[0100] In Table 3, V W is the total width of the vehicle (m); V Lis the overall length of the vehicle (m).

[0101] Vehicle Euler angle

[0102] As Figure 8 shown, the vehicle Euler angles are the general terms for the heading angle Y, pitch angle X, and roll angle Z, which can be used to describe whether there are unsafe behaviors such as straddling, rollover, and over-turning after a truck collides with a movable steel guardrail. The vehicle Euler angles of the present invention include: heading angle Y, pitch angle X, and roll angle Z.

[0103] As Figure 8 shown, the Z-axis direction is the normal forward direction of the vehicle. The heading angle Y can describe whether the guardrail can safely redirect the collided vehicle. When the heading angle Y is 0, it means the vehicle is traveling in the normal heading, that is, the smaller the heading angle Y, the better the redirection function.

[0104] The pitch angle X can describe whether the vehicle has straddling or over-turning behaviors. The larger the pitch angle X, the more serious the pitching behavior of the vehicle. When the pitch angle X exceeds a certain threshold, it means the front of the vehicle is severely lifted, and there may be a behavior that the vehicle straddles the movable steel guardrail.

[0105] The roll angle Z can describe whether the vehicle has a rollover behavior. The larger the roll angle Z, the more serious the roll of the vehicle. When the roll angle Z exceeds a certain threshold, the vehicle may roll over.

[0106] Compared with evaluation criteria such as the guiding exit frame and the centroid acceleration, the vehicle Euler angles can more intuitively reflect the blocking effect of the movable steel guardrail on the vehicle, and can be used as an important optimization goal in the optimization design of the movable steel guardrail. Controlling the vehicle Euler angles within a certain range can ensure that the vehicle does not have straddling, rollover and other behaviors, and can redirect the vehicle.

[0107] Therefore, in order to improve the protection efficiency of the movable steel guardrail, the present invention refers to the current relevant standard specifications and research results, and takes the lateral component ORA x and the longitudinal component ORA y of the vehicle centroid acceleration, the vehicle guiding exit frame parameter DIS (that is, after the vehicle collides with the movable steel guardrail, the vehicle exits the vehicle and enters the vehicle guiding exit frame at the last contact point between the vehicle and the original position of the guardrail, and the distance DIS between the outermost wheel of the vehicle and the movable steel guardrail) and the vehicle Euler angles (heading angle Y, pitch angle X, and roll angle Z) as evaluation indicators to measure the protection effect of the movable steel guardrail.

[0108] Among them, the lateral component ORA x and the longitudinal component ORA y of the vehicle centroid acceleration and the vehicle guiding exit frame parameter DIS are used as evaluation control criteria to ensure that the various parameters of the vehicle when colliding with the guardrail do not exceed the specified values.

[0109] The Euler angles of the vehicle (heading angle Y, pitch angle X, and roll angle Z) are used as the evaluation optimization objectives, so that the heading angle Y, pitch angle X, and roll angle Z of the vehicle are as small as possible to improve the protection efficiency of the guardrail.

[0110] As Figure 9 shown, after the vehicle collides with the movable steel guardrail with different guardrail heights and anchoring spacings in step S2, the data distribution of the vehicle's centroid acceleration (the change of the vehicle's centroid acceleration).

[0111] Figure 9 In y it, divided by the middle gray plane, the upper part is the longitudinal component ORA of the vehicle's centroid acceleration x change situation, and the lower part is the transverse component ORA of the vehicle's centroid acceleration y change situation. It can be clearly seen from the figure that the longitudinal component ORA of the vehicle's centroid acceleration x is larger than the transverse component ORA of the vehicle's centroid acceleration, indicating that when the vehicle collides with the movable steel guardrail, the longitudinal speed of the vehicle changes greatly and the force is also large, while the transverse speed changes little and the force is also small. This is consistent with the actual situation where the vehicle collides with the movable steel guardrail at a collision angle of 30°.

[0112] The longitudinal component ORA of the maximum vehicle centroid acceleration of the vehicle y appears when the height H of the movable steel guardrail is 760 mm and the anchoring spacing L is 30 m. In this scenario, due to the relatively large height H of the movable steel guardrail and the relatively small anchoring spacing L, it has a relatively large tensile force and a relatively large effect on the vehicle, so the acceleration is the largest.

[0113] It can be seen from the figure that as the height H of the guardrail decreases, both the transverse component ORA of the vehicle's centroid acceleration x and the longitudinal component ORA of the vehicle's centroid acceleration y show a decreasing trend, indicating that the guardrail height H has a certain influence on the vehicle acceleration. The smaller the guardrail height H, the smaller the acceleration applied to the vehicle and the worse the effect of changing the vehicle speed. As the anchoring spacing L increases, both the transverse component ORA of the vehicle's centroid acceleration x and the longitudinal component ORA of the vehicle's centroid acceleration y show a decreasing trend, but compared with the guardrail height H, the change is smaller. This indicates that the change of the guardrail height H has a more significant effect on the change of the vehicle's centroid acceleration.

[0114] As Figure 10 shown, after the vehicle collides with the movable steel guardrail with different guardrail heights and anchoring spacings in step S2, the data distribution of the vehicle's guiding exit frame parameter DIS. From Figure 10It can be found that as the height H of the movable steel guardrail increases and the anchoring spacing L decreases, the vehicle guiding and exiting frame parameter DIS (the distance DIS between the outermost wheel of the vehicle and the movable steel guardrail when the vehicle exits the vehicle guiding and exiting frame after colliding with the movable steel guardrail and enters the vehicle guiding and exiting frame from the last contact point between the vehicle and the original position of the guardrail) shows a gradually decreasing trend, indicating that the distance between the vehicle and the adjacent lane after colliding with the movable steel guardrail gradually increases, the probability of colliding with the adjacent lane gradually decreases, and the probability of secondary collision injury after the vehicle collision also gradually decreases. The test results show that the increase in the guardrail height H and the decrease in the anchoring spacing L contribute to the improvement of traffic safety level.

[0115] As Figure 11 shown, the data distribution of the vehicle Euler angles after the vehicle collides with the movable steel guardrails with different guardrail heights and anchoring spacings obtained in step S2. As can be seen from Figure 11 it, the pitch angle X is less sensitive to the guardrail height H and the anchoring spacing L. As the guardrail height H increases and the anchoring spacing L decreases, both the roll angle Z and the heading angle Y show a decreasing trend, and the change is relatively obvious.

[0116] As the anchoring spacing L increases, the change of the heading angle Y is more sensitive, indicating that the anchoring spacing L has a more obvious influence on the heading angle Y. It can be explained that as the anchoring spacing L increases, the vehicle can have a larger lateral displacement when colliding with the movable steel guardrail, resulting in a gradual increase in the heading angle Y. The analysis of the test data shows that controlling the Euler angles (heading angle Y, pitch angle X, and roll angle Z) plays an important role in ensuring the safe driving of the vehicle. At the same time, there is a strong positive relationship between the Euler angles and the guardrail height H and the anchoring spacing L. Taking the Euler angles as the optimization control target can ensure that vehicles such as trucks reduce the safety risks when colliding with the guardrail.

[0117] By analyzing the collision result data (the vehicle centroid acceleration, vehicle guiding and exiting frame parameters, and vehicle Euler angles) generated from the collision simulation experiment of the vehicle with the movable steel guardrail obtained in step S2, it can be found that the guardrail height H and the anchoring spacing L have a greater impact on the safety protection effect of the movable steel guardrail. When designing and using the guardrail, the guardrail height H and the anchoring spacing L should be focused on. In order to maximize the protection effect of the movable steel guardrail against vehicle collisions, the guardrail height H and the anchoring spacing L should be reasonably designed.

[0118] Step S3: Use the RBF neural network (Radial Basis Function Neural Network) to fit the vehicle centroid acceleration, vehicle guiding and exiting frame parameters, and vehicle Euler angles, and use the trained RBF neural network to predict the values of the vehicle Euler angles.

[0119] According to an embodiment of the present invention, taking the guardrail height H and the anchoring spacing L of the movable steel guardrail as inputs, and taking the vehicle centroid acceleration, the vehicle guiding and exiting frame parameters, and the vehicle Euler angle after the vehicle collides with the movable steel guardrail as outputs, an RBF neural network is used to fit the vehicle centroid acceleration, the vehicle guiding and exiting frame parameters, and the vehicle Euler angle, and the trained RBF neural network is used to predict the value of the vehicle Euler angle.

[0120] In step S1, with the guardrail height H and the anchoring spacing L of the movable steel guardrail as variables, a collision simulation test of the vehicle and the movable steel guardrail is carried out. In step S2, the guardrail height H and the anchoring spacing L of the movable steel guardrail, as well as the vehicle centroid acceleration, the vehicle guiding and exiting frame parameters, and the vehicle Euler angle after the movable steel guardrail collides are obtained.

[0121] Since a single simulation calculation takes too long, it is somewhat unrealistic to construct a finite model and perform simulation calculations for each possible result. At the same time, the calculation results of Ls-Dyna are discrete, and it is difficult to perform joint calculations with continuous data. To solve the above problems, simulation experiments are usually designed according to a certain rule to obtain discrete experimental results. Then, a cost function is used to fit the event results to obtain the functional relationship between various influencing factors and evaluation criteria, thereby facilitating and quickly calculating in the multi-objective optimization model of the movable steel guardrail (the multi-objective optimization model of the movable steel guardrail will be described below), and at the same time, a certain optimization accuracy can be satisfied. Usually, the cost function can use a radial basis function neural network (Radial Basis Function Neural Network, RBF neural network).

[0122] The RBF neural network is a commonly used three-layer feedforward network, which has excellent performance in the field of function approximation. It was first introduced and used by Broomhead, Lowe, Moody, and Darken in 1998. Compared with other types of artificial neural networks, the RBF network has the characteristics of simple structure, fast learning speed, excellent approximation performance, and high generalization ability. Therefore, an RBF neural network is used to construct a cost function to approximate and simulate the relationship between factors such as the guardrail height, the anchoring spacing, the vehicle centroid acceleration, the vehicle Euler angle, and the vehicle guiding and exiting frame. Compared with cost functions such as radial basis functions, it has the advantage of high fitting accuracy.

[0123] Furthermore, for the different guardrail heights H and anchoring spacings L of the movable steel guardrail obtained in step S2, as well as the vehicle centroid acceleration, the vehicle guiding and exiting frame parameters, and the vehicle Euler angle after the movable steel guardrail collides, the method of Monte Carlo sampling is used for data augmentation.

[0124] Generally, enough data is required to train an RBF neural network that meets the accuracy requirements well. In this embodiment, the 16 sets of collision result data obtained from 16 finite element simulation test sample points usually cannot meet the training requirements of the RBF neural network. Therefore, the Monte Carlo sampling method is used to expand the 16 sets of basic data to generate 100 new data sets that conform to the original probability distribution, providing sufficient data for subsequent guardrail optimization. Part of the data of Monte Carlo sampling is shown in Table 4.

[0125] Table 4 Part of the data of Monte Carlo sampling

[0126]

[0127]

[0128] As an important data sampling and expansion method, the Monte Carlo method can effectively sample the data of large trucks colliding with movable steel guardrails, expand the 16 sets of simulation experiment data to 100 sets, enable the RBF neural network to be effectively trained, and the training fitting accuracy meets the accuracy requirements. Compared with obtaining 100 sets of data through simulation experiments, while ensuring the accuracy, it reduces the simulation time consumption and the occupation of computing resources.

[0129] Take the guardrail height H and the anchoring spacing L in Table 4 as inputs, the heading angle Y, the pitch angle X, and the roll angle Z, the lateral component ORA of the vehicle centroid acceleration x and the longitudinal component ORA of the vehicle centroid acceleration y , and the vehicle guiding exit box parameter DIS (after the vehicle collides with the movable steel guardrail, the vehicle exits the vehicle and enters the vehicle guiding exit box at the last contact point between the vehicle and the original position of the guardrail, and the distance DIS between the outermost wheel of the vehicle and the movable steel guardrail) are used as outputs respectively. Use the RBF neural network to fit the vehicle centroid acceleration, the vehicle guiding exit box parameter, and the vehicle Euler angles to train the RBF neural network, and use the trained RBF neural network to predict the values of the vehicle Euler angles.

[0130] Furthermore, select 60% of the data from the data expanded by the Monte Carlo sampling method as training samples. Take the guardrail height H and the anchoring spacing L of the movable steel guardrail as inputs and input them into the RBF neural network. Take the vehicle centroid acceleration, the vehicle guiding exit box parameter, and the vehicle Euler angles after the vehicle collides with the movable steel guardrail as outputs. Use the RBF neural network to fit the vehicle centroid acceleration, the vehicle guiding exit box parameter, and the vehicle Euler angles to train the RBF neural network, and use the trained RBF neural network to predict the values of the vehicle Euler angles.

[0131] 20% of the data were selected as validation samples from the data expanded by the Monte Carlo sampling method. The guardrail height H and anchor spacing L of the movable steel guardrail were used as input and input into the RBF neural network. The vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angles after the vehicle collided with the movable steel guardrail were used as output to verify the fitting effect of the RBF neural network.

[0132] 20% of the data were selected as test samples from the data expanded by the Monte Carlo sampling method. The guardrail height H and anchor spacing L of the movable steel guardrail were used as input and input into the RBF neural network. The vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angles after the vehicle collided with the movable steel guardrail were used as output to test the fitting effect of the RBF neural network.

[0133] like Figure 12 As shown, the fitting accuracy of the heading angle Y is given as an example. Figure 12 The horizontal axis is the original heading angle Y data in Table 4, and the vertical axis is the predicted value of the heading angle Y by the RBF neural network. When all the data points are distributed on the straight line Y=T, it means that the RBF neural network has the best fitting effect.

[0134] Figure 12 (g) is the fitting accuracy of the training sample, (h) is the fitting accuracy of the validation sample, (i) is the fitting accuracy of the test sample, and (j) is the fitting accuracy of all samples. Figure 12 It can be seen from (g), (h), (i), and (j) that the data points are distributed near Y = T, and Figure 12 The fitting accuracy of the (i) test sample is the lowest, which is 0.9194. Figure 12 It can be seen that the trained RBF neural network has high fitting accuracy and can meet the requirements of optimization design.

[0135] Figure 12 The fitting accuracy of the heading angle Y is given as an example, and it should be understood that the predicted values ​​of the vehicle Euler angles include: the predicted value of the heading angle Y, the predicted value of the pitch angle X, and the predicted value of the roll angle Z. Those skilled in the art should understand that the same method as the heading angle Y is used to verify that the fitting accuracy of the predicted value of the pitch angle X and the predicted value of the roll angle Z predicted by the RBF neural network meets the optimization design requirements.

[0136] Step S4: construct a multi-objective optimization model for movable steel guardrail.

[0137] A multi-objective optimization model for movable steel guardrail is constructed with the guardrail height H and anchor spacing L of the movable steel guardrail as design variables, vehicle center of mass acceleration and vehicle guide exit box as limiting factors, and the minimum value of the predicted value of the vehicle Euler angle predicted by RBF neural network as the goal.

[0138] Specifically, the multi-objective optimization model of the movable steel guardrail is constructed by the following method:

[0139] Multi-objective optimization model of the movable steel guardrail: min[X(H,L), Y(H,L), Z(H,L)],

[0140] Boundary conditions: 30 ≤ L ≤ 70,

[0141] 600 ≤ H ≤ 1100,

[0142] ORA x ≤ 200,

[0143] ORA y ≤ 200,

[0144] DIS ≤ 8.1,

[0145] Among them, X(H,L) is the relationship function between the predicted value of the pitch angle X of the vehicle's Euler angle and the guardrail height H and the anchoring distance L of the movable steel guardrail, Y(H,L) is the relationship function between the predicted value of the heading angle Y of the vehicle's Euler angle and the guardrail height H and the anchoring distance L of the movable steel guardrail, and Z(H,L) is the relationship function between the predicted value of the roll angle Z of the vehicle's Euler angle and the guardrail height H and the anchoring distance L of the movable steel guardrail;

[0146] H is the guardrail height of the movable steel guardrail, and L is the anchoring distance of the movable steel guardrail; ORA x is the lateral component of the vehicle's centroid acceleration, ORA y is the longitudinal component of the vehicle's centroid acceleration; DIS is the vehicle guiding exit box parameter, that is, after the vehicle collides with the movable steel guardrail, the vehicle exits the vehicle and enters the vehicle guiding exit box at the last contact point of the original position of the guardrail, and the distance DIS between the outermost wheel of the vehicle and the movable steel guardrail.

[0147] The protection effect of the movable steel guardrail is affected by many factors such as the steel plate thickness, guardrail height, and anchoring distance. It is difficult to consider all factors during the optimization design. According to engineering practical experience and theoretical research results, combined with the design optimization objectives, usually one or several factors with more prominent influence are optimized. The present invention selects two most critical influencing factors, namely the guardrail height H and the anchoring distance L, and takes the minimum value of the predicted value of the pitch angle X of the vehicle's Euler angle, the minimum value of the predicted value of the heading angle Y of the vehicle's Euler angle, and the minimum value of the predicted value of the roll angle Z of the vehicle's Euler angle as the objectives to construct a multi-objective optimization model of the movable guardrail.

[0148] The boundary conditions of the multi-objective optimization model of the movable guardrail are:

[0149] The value range of the anchoring spacing L of the movable steel guardrail is from 30 m to 70 m. Currently, an anchoring spacing of 50 m is mostly adopted in the reconstruction and expansion projects. An overly long anchoring spacing will reduce the protective performance of the guardrail, while an overly short anchoring spacing will aggravate the degree of pavement damage.

[0150] For the maximum and minimum values of the guardrail height H of the movable steel guardrail, since there is no relevant design standard for the movable steel guardrail, referring to the main types of guardrails in the current market, the maximum height of the guardrail is set to 1100 mm and the minimum height is set to 600 mm.

[0151] The lateral component ORA of the vehicle's centroid acceleration x and the longitudinal component ORA y shall not be greater than the requirements in the current standard specifications, that is, the lateral component ORA of the vehicle's centroid acceleration x and the longitudinal component ORA y shall not be greater than 200 m / s 2 .

[0152] If the vehicle guiding and exiting frame parameter DIS (after the vehicle collides with the movable steel guardrail, the vehicle exits the last contact point between the vehicle and the original position of the guardrail and enters the vehicle guiding and exiting frame, and the distance DIS between the outermost wheel of the vehicle and the movable steel guardrail) is less than or equal to 8.1 m, it can be determined that the vehicle is safely redirected.

[0153] Step S5: Use the multi-objective particle swarm optimization algorithm to solve the multi-objective optimization model of the movable steel guardrail, draw the boundary graph of the Pareto optimal solution set, and determine the guardrail height H and the anchoring spacing L.

[0154] According to the embodiments of the present invention, use the multi-objective particle swarm optimization algorithm to solve the multi-objective optimization model of the movable steel guardrail, obtain the Pareto optimal solution set of the multi-objective optimization model of the movable steel guardrail, and draw the boundary graph of the Pareto optimal solution set. According to the boundary graph of the Pareto optimal solution set, determine the guardrail height H and the anchoring spacing L.

[0155] The multi-objective optimization model of the movable steel guardrail constructed in step S4 is usually very difficult to obtain an exact solution by solving the canonical form. Heuristic algorithms such as the multi-objective particle swarm optimization algorithm (Multiple Objectives With Particle Swarm Optimization, MOPSO) can effectively and more efficiently solve multi-objective optimization problems. This algorithm was proposed by Carlos A Coello et al. in 2004, aiming to apply the particle swarm optimization algorithm (Particle Swarm Optimization, PSO), which was originally only applicable to single objectives, to multi-objectives. The present invention uses the multi-objective particle swarm optimization algorithm to solve the multi-objective optimization model of the movable steel guardrail constructed in step S4.

[0156] In a specific embodiment, first, the values of the key parameters of the multi-objective particle swarm optimization algorithm are determined. The population size is set to 90, the archive size is set to 450, the number of iterations is set to 500, the inertia weight coefficient is set to 0.5, the inertia weight decay rate is set to 0.99, the global learning factor is set to 2, and the mutation rate is set to 0.2. Through multiple operations, a relatively stable Pareto optimal solution set of the multi-objective optimization model of the movable steel guardrail is obtained, and each solution in the solution set can meet the requirements of the vehicle's centroid acceleration and the vehicle's guiding exit box.

[0157] According to the Pareto optimal solution set of the multi-objective optimization model of the movable steel guardrail, a Pareto optimal solution set boundary graph is drawn. The Pareto optimal solution set boundary graph is as Figure 13 shown.

[0158] According to the Pareto optimal solution set boundary graph, the guardrail height H and the anchoring spacing L are determined. As Figure 13 shown, the two ends of the Pareto optimal solution set boundary graph are the first optimal solution ( Figure 13 point C in Figure 13 ) and the second optimal solution (

[0159] point D in

[0160] shown). Figure 13 Specifically, according to the following method, the guardrail height H and the anchoring spacing L are determined based on the Pareto optimal solution set boundary graph: Figure 13 As

[0161] shown, when the predicted values of the pitch angle X of the vehicle's Euler angle and the predicted value of the roll angle Z of the vehicle's Euler angle in the Pareto optimal solution set boundary graph are the smallest, the guardrail height H and the anchoring spacing L corresponding to the first optimal solution ( Figure 13 point C in

[0162] ) are selected as the design parameters of the guardrail height H and the anchoring spacing L. Figure 13 In this embodiment, the guardrail height H corresponding to the first optimal solution ( Figure 13 point C in

[0163] ) is 758 mm and the anchoring spacing L is 47 m. Figure 13 As

[0164] Figure 13 shown, when the predicted value of the heading angle Y of the vehicle's Euler angle in the Pareto optimal solution set boundary graph is the smallest, the guardrail height H and the anchoring spacing L corresponding to the second optimal solution ( Figure 13 point D in

[0163] ) are selected as the design parameters of the guardrail height H and the anchoring spacing L.

[0163] In this embodiment, the guardrail height H corresponding to the second optimal solution ( Figure 13 point D in

[0164] ) is 721 mm and the anchoring spacing L is 36 m.

[0164] When comprehensively measuring the magnitudes of the predicted values of the pitch angle X, the heading angle Y, and the roll angle Z of the vehicle's Euler angles, the guardrail height H is 729 mm, and the anchoring spacing L is 40 m. The optimized design can provide a reference for practical engineering applications.

[0165] It is recommended that in practical engineering applications, in order to improve the protection efficiency of the movable steel guardrail, the anchoring spacing L can be selected to be about 40 m, and at the same time, the guardrail height H can be increased to 750 mm or higher.

[0166] When the anchoring spacing L of the movable steel guardrail is set to 40 m and the guardrail height H is set to 750 mm, it can ensure that when the most types of vehicles such as small passenger cars and large trucks collide with the movable steel guardrail on the highway, while meeting the requirements of the design specifications for the vehicle's centroid acceleration and the vehicle's guiding exit frame, it can also effectively reduce the occurrence rate of dangerous situations such as large trucks straddling, rolling over, and climbing over. At the same time, compared with the 50 m anchoring spacing L used in actual engineering, for every 1 km distance, the number of anchorings will increase by 5. Although it will increase some construction workload, it can effectively improve the safety efficiency of the guardrail itself.

[0167] Using the optimized finite element model of the movable steel guardrail, conduct a collision simulation test again. The collision process of the vehicle is as Figure 14 shown. The pitch angle X, the heading angle Y, and the roll angle Z of the vehicle have decreased, and the straddling, rolling over, and climbing over situations of the truck have all been alleviated.

[0168] It should be noted that when designing the movable steel guardrail, appropriate design objectives should be selected according to different vehicle types. For example, small cars travel at a relatively high speed on the highway. When they collide with the movable steel guardrail, it is easier to cause harm to the drivers and passengers in the vehicle and drive into the adjacent lane. Therefore, it is more appropriate to use the vehicle's centroid acceleration and guiding exit frame as the optimization objectives of the optimization model. Since large trucks have a relatively high body height, when colliding with the movable steel guardrail, they are more likely to have dangerous situations such as straddling, rolling over, and climbing over. Therefore, in this invention, it is more appropriate to use the vehicle's centroid acceleration, the vehicle's guiding exit frame as control indicators, and the vehicle's Euler angles as the optimization design objectives. Through the collision simulation analysis and optimization design of large trucks and small passenger cars, it is found that the anchoring spacing L has an important impact on the safety of the movable steel guardrail, and in practical engineering, the research and analysis of the anchoring spacing L should be strengthened.

[0169] According to an embodiment of the present invention, there is provided a system for optimizing the design of a movable steel guardrail based on the Euler angles of a truck, which is used to execute a method for optimizing the design of a movable steel guardrail based on the Euler angles of a truck according to the present invention.

[0170] A method and system for optimizing the design of a movable steel guardrail based on vehicle Euler angles provided by the present invention. In order to reduce the incidence of dangerous behaviors such as riding over, rolling over, and climbing over of trucks after colliding with the movable steel guardrail during the reconstruction and expansion project of expressways, the vehicle Euler angle is introduced as an evaluation index for the safety performance of the guardrail. A multi-objective optimization model is constructed with the guardrail height H and the anchoring spacing L as design variables, the vehicle centroid acceleration and the vehicle guiding exit frame as control indexes, and the minimum vehicle Euler angle (pitch angle, roll angle, yaw angle) as the objective. First, a finite element model of the vehicle and the movable steel guardrail that is consistent with the real vehicle collision test is constructed, and 16 simulation tests are designed. Secondly, the collision simulation data is sampled and expanded through Monte Carlo simulation, solving the problems of time-consuming finite element simulation and insufficient simulation data volume. Then, the RBF neural network is used to fit the vehicle collision situation. Finally, the multi-objective particle swarm algorithm is used to solve the multi-objective optimization model of the movable steel guardrail. The results show that the Monte Carlo and RBF neural network methods have high accuracy in fitting the vehicle collision situation. The guardrail height H and the anchoring spacing L have an impact on the vehicle Euler angle, the vehicle centroid acceleration, and the vehicle guiding exit frame, etc. Under the requirements of meeting the vehicle centroid acceleration and the vehicle guiding exit frame, by reducing the anchoring spacing L and increasing the guardrail height H, the vehicle Euler angle is reduced, realizing the improvement of the protection effect of the movable steel guardrail, providing a certain reference for engineering applications.

[0171] The following points need to be explained:

[0172] (1) The attached drawings of the embodiments of the present invention only relate to the structures involved in the embodiments of the present invention, and other structures can refer to the general design.

[0173] (2) For clarity, in the attached drawings used to describe the embodiments of the present invention, the thickness of the layer or region is enlarged or reduced, that is, these drawings are not drawn according to the actual ratio. It can be understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element can be "directly" on or under the other element or there can be an intermediate element.

[0174] (3) Without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0175] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for optimizing the design of a movable steel guardrail based on Euler angles for trucks, characterized in that: The method comprises the following steps: S1. Constructing a vehicle finite element model and a movable steel guardrail finite element model, taking the guardrail height H and anchor spacing L of the movable steel guardrail as variables, performing a collision simulation test between the vehicle and the movable steel guardrail; S2, obtaining collision result data generated by a collision simulation experiment between a vehicle and a movable steel guardrail, wherein the collision result data includes: vehicle center of mass acceleration, vehicle guide exit frame parameters, and vehicle Euler angles after the vehicle collides with the movable steel guardrail; S3, taking the guardrail height H and anchor spacing L of the movable steel guardrail as input, and taking the vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angle after the vehicle collides with the movable steel guardrail as output, using RBF neural network to fit the vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angle, and using the trained RBF neural network to predict the values ​​of vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angle; S4, taking the guardrail height H and anchor spacing L of the movable steel guardrail as design variables, the vehicle center of mass acceleration and the vehicle guide exit box parameters as limiting factors, and the minimum value of the predicted value of the vehicle Euler angle predicted by the RBF neural network as the goal, a multi-objective optimization model for the movable steel guardrail is constructed; S5. Using a multi-objective particle swarm algorithm, solve the multi-objective optimization model of the movable steel guardrail, obtain the Pareto optimal solution set of the multi-objective optimization model of the movable steel guardrail, and draw a Pareto optimal solution set boundary diagram; According to the Pareto optimal solution set boundary diagram, the guardrail height H and anchor spacing L are determined.

2. The method according to claim 1, characterized in that In step S2, the vehicle Euler angles include: heading angle Y, pitch angle X and roll angle Z.

3. The method according to claim 1, characterized in that In step S3, the guardrail height H and anchor spacing L of the movable steel guardrail obtained in step S2, as well as the vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angle after the movable steel guardrail collides, are data expanded using the Monte Carlo sampling method.

4. The method according to claim 3, characterized in that From the data expanded by the Monte Carlo sampling method, 60% of the data are selected as training samples; The guardrail height H and anchor spacing L of the movable steel guardrail are taken as input and input into the RBF neural network. The vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angle after the vehicle collides with the movable steel guardrail are taken as output. The vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angle are fit by RBF neural network, and the trained RBF neural network is used to predict the values ​​of vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angle.

5. The method according to claim 3, characterized in that: From the data expanded by the Monte Carlo sampling method, 20% of the data are selected as validation samples; The guardrail height H and anchor spacing L of the movable steel guardrail are taken as input into the RBF neural network, and the vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angles after the vehicle collides with the movable steel guardrail are taken as output to verify the fitting effect of the RBF neural network.

6. The method according to claim 3, characterized in that From the data expanded by the Monte Carlo sampling method, 20% of the data are selected as test samples; The guardrail height H and anchor spacing L of the movable steel guardrail are taken as input and input into the RBF neural network. The vehicle center of mass acceleration, vehicle guide exit box parameters and vehicle Euler angles after the vehicle collides with the movable steel guardrail are taken as output to test the fitting effect of the RBF neural network.

7. The method according to claim 1, characterized in that In step S3, the predicted values ​​of the vehicle's Euler angles include: a predicted value of the heading angle Y, a predicted value of the pitch angle X, and a predicted value of the roll angle Z.

8. The method according to claim 1, characterized in that In step S4, the multi-objective optimization model of the movable steel guardrail is constructed by the following method: Multi-objective optimization model of movable steel guardrail: min[X(H,L),Y(H,L),Z(H,L)], Boundary conditions: 30≤L≤70, 600≤H≤1100， ORA x ≤200, ORA y ≤200, DIS≤8.1, Among them, X(H,L) is the relationship function between the predicted value of the pitch angle X of the vehicle's Euler angle and the guardrail height H and anchor spacing L of the movable steel guardrail, Y(H,L) is the relationship function between the predicted value of the heading angle Y of the vehicle's Euler angle and the guardrail height H and anchor spacing L of the movable steel guardrail, and Z(H,L) is the relationship function between the predicted value of the roll angle Z of the vehicle's Euler angle and the guardrail height H and anchor spacing L of the movable steel guardrail; H is the guardrail height of the movable steel guardrail, L is the anchorage spacing of the movable steel guardrail; ORA x is the lateral component of the vehicle center of mass acceleration, ORA y is the longitudinal component of the vehicle's center of mass acceleration; DIS is the vehicle's guided exit frame parameter.

9. The method according to claim 1, characterized in that: In step S5, the guardrail height H and the anchor spacing L are determined according to the Pareto optimal solution set boundary graph in the following manner: When the predicted values ​​of the pitch angle X of the vehicle's Euler angle and the predicted values ​​of the roll angle Z of the vehicle's Euler angle are the smallest in the Pareto optimal solution set boundary graph, the guardrail height H and anchor spacing L corresponding to the first optimal solution are selected as the design parameters of the guardrail height H and anchor spacing L; When the predicted value of the heading angle Y of the vehicle Euler angle in the Pareto optimal solution set boundary graph is the smallest, the guardrail height H and anchor spacing L corresponding to the second optimal solution are selected as the design parameters of the guardrail height H and anchor spacing L; The first optimal solution and the second optimal solution are two ends of the Pareto optimal solution set boundary graph.

10. A system for optimizing the design of movable steel guardrails for trucks based on Euler angles, characterized in that: The system is used to perform the method according to any one of claims 1 to 9.

Images