Research on crashworthiness and lightweight optimi

Research on the crashworthiness and lightweight optimization design of car B-pillar

Abstract: in order to improve the crashworthiness of car B-pillar in side impact and reduce the mass of B-pillar to realize the lightweight of car body, the safety performance of a car in side impact is analyzed by finite element method. Aiming at the problems of excessive intrusion and intrusion speed at the waist line of B-pillar in side impact, large number of stiffeners and overweight in B-pillar structure, tailor welded plate structure is used to improve the design of B-pillar outer plate. On the premise of considering crashworthiness and lightweight, orthogonal experimental design and multi-objective genetic algorithm are used to optimize the position of tailor welded seam and the thickness of each part. By optimizing the design, the deformation mode of B-pillar in side impact is improved. The maximum intrusion volume of B-pillar is reduced by 10%, the intrusion volume and intrusion speed at the waist line are reduced by 18% and 12% respectively, and the mass is reduced by 18%. The results show that the requirements of crashworthiness and lightweight can be effectively balanced by using tailor welded blanks on the B-pillar and optimizing the design

key words: crashworthiness; Lightweight; B-pillar; Optimization design

introduction

due to the continuous improvement of safety regulations and market requirements for vehicle crash safety, the quality of traditional vehicle bodies will likely become larger and larger, but at the same time, vehicle lightweight is an important measure to achieve vehicle fuel economy [1]. Therefore, when designing and improving vehicle bodies, taking into account the conflicting requirements of crashworthiness and lightweight has become a hot issue in the automotive industry. Using tailor welded blanks on key components that affect vehicle crash safety is one of the effective ways to meet these two requirements [2]

tailor welded blanks are light-weight sheets with ideal strength and stiffness obtained by welding two or more steel plates with different mechanical properties, coatings and thicknesses. Min et al. [3] obtained from the material tensile test that the tensile strength of tailor welded steel plate of the same material is almost the same as that of a single steel plate, that is, the stress-strain characteristics of tailor welded steel plate with good welding are basically not affected by the welding process, so it can be considered that the collision performance of tailor welded steel plate is not affected by the welding process

tailor welded technology has been widely concerned and applied in the automotive industry, but its design mainly relies on expert experience or refers to the existing tailor welded plate structure, and only a few scholars have done some quantitative research. Shin et al. [4], Lee et al. [5], Zhu et al. [2], song et al. [6] used tailor welded blanks in door design and carried out a series of optimization respectively. Yang Yuze et al. [7] used tailor welded blanks to improve the design of the front longitudinal beam of a vehicle, and optimized the material grade and thickness of each different thickness steel plate by orthogonal test. Shi Yuliang et al. [8] studied the design method of lightweight improvement of the front longitudinal beam using tailor welded blanks

in car side impact, the intrusion volume, intrusion speed and intrusion form of side wall structure are the main factors that directly affect the safety of passengers [9]. The side wall structure mainly includes B-pillar, door inner and outer panels, anti-collision bar, sill and other components, and B-pillar is the main force bearing component in side impact. Therefore, the deformation mode of B-pillar is very important in the whole collision process. After comparing the local and global approximation methods, marklund et al. [1] optimized the B-pillar with the global approximation in the form of linear and quadratic response surface, reducing the total mass of the B-pillar by 25%. You Guozhong et al. [10] optimized and improved the inner plate of B-pillar by establishing a simplified model of B-pillar and combining topology and shape optimization in Altair Optistruct software, reducing the intrusion speed at the waist line of B-pillar

in this paper, the ideal B-pillar deformation model - "pendulum deformation mode" is obtained by using tailor welded blanks on the B-pillar [11]. At the same time, the relevant parameters of tailor welded blanks (weld position, plate thickness) are analyzed by combining orthogonal experimental design and multi-objective genetic algorithm The optimization design is carried out to improve the crashworthiness of B-pillar and reduce the total mass of B-pillar

1 preparation for multi-objective optimization

1.1 establishment and verification of finite element model

this paper takes a mass-produced car as the research object. According to the requirements of China's regulations on occupant protection in side impact of vehicles, a side impact finite element model of the moving deformable barrier colliding vertically with the vehicle model at the speed of 50km/h is established, as shown in Figure 1. The collision simulation time is set to 0.1s. The validity of the finite element model of the whole vehicle has been verified by experiments. Figure 2 shows the comparison between the simulation and experiment of the acceleration curve of the lower part of the B-pillar on the non impact side of the whole vehicle in the side impact of the car. It can be seen from Figure 2 that the change trend of the acceleration curve of the experiment and simulation is basically the same, and the peak time is relatively consistent. There is a certain difference between the acceleration peak of the experiment curve and the simulation curve, but the error is less than 5%

Figure 1 finite element model of side impact

Figure 2 acceleration curve of car body in simulation and experiment

1.2 side impact safety analysis

use LS-DYNA software to simulate the side impact of the target car. The simulation results show that the side impact safety performance of the body structure is not ideal. As shown in Figure 3 and Figure 4, the side wall of the car body is seriously sunken due to the large deformation of the roof beam and the floor, and the intrusion volume and intrusion speed of the side wall are too large

Figure 3 side wall deformation of the whole vehicle

Figure 4 intrusion speed time curve at the waist line of the inner plate of the B-pillar

the purpose of this paper is to optimize the products of our company specializing in the production of B-pillar: universal testing machine, tensile testing machine, pressure testing machine, impact testing machine, and so on. Therefore, it is decided to properly adjust the materials and thickness of the door sill, roof beam, rear floor and door anti-collision bar in the vehicle model, so as to improve the deformation of the vehicle body structure locally, but there are still two obvious problems in the B-pillar

(1) the deformation mode of B pillar in side impact is not ideal, the deformation at the waist line is serious, and the intrusion speed is too high. The damage of human chest (rib deformation index RDC) is directly proportional to the speed of impacting the dummy at the waist line of B pillar [12], so there is a great risk of human chest injury in the side impact of this vehicle

(2) there are many reinforcing plates in the B-pillar structure, as shown in Figure 5. Using too many stiffeners is not only not conducive to the lightweight of the body, but also increases the complexity of body design and vehicle assembly

Figure 5 basic composition of B-pillar

in order to solve the above two problems, this paper improves and optimizes the B-pillar of the target car

1.3 structural design scheme of B-pillar tailor welded blanks

in side impact, the damage of internal organs in human abdominal cavity is less threatening to life than that of chest. At the same time, the impact of invasion speed on the deformation of human shoulder and chest is more significant than that of abdomen [12]. Therefore, from the perspective of occupant injury protection, we should first reduce the deformation and intrusion speed of the side wall of the car body at the chest position under the normal sitting posture of the human body. This study attempts to meet the design requirements by obtaining an ideal B-pillar deformation mode, which is a pendulum deformation mode [11] with the connection between the upper end of the B-pillar and the roof beam as the center of the circle and the lower end of the B-pillar rotating inward, as shown in Figure 6. This deformation mode requires that the stiffness of B-pillar structure obey the distribution form of high up and low down. In order to obtain this structural stiffness, plates with different thicknesses and materials can be used for butt welding connection, which can not only save materials, but also improve the flexibility of design [13]. Therefore, the original B-pillar model is replaced by tailor welded plate structure. The B-pillar outer plate can be divided into upper and lower parts for butt welding, as shown in Figure 7. The weld is simplified as a simple boundary of plates with different thickness, and the nodes can be merged directly in the finite element model

Figure 6 schematic diagram of pendulum deformation mode of B-pillar Figure 7 tailor welded structure of B-pillar outer plate

another advantage of applying tailor welded plate structure is that it can reduce the number of components, and can use high-strength materials or increase the thickness in the key areas that must bear high stress to replace the reinforcing plate of the original model in this area [2]. In this study, the original B-pillar model includes outer plate, inner plate and stiffener 1 ~ stiffener 4, as shown in Figure 5. After the tailor welded plate structure is used to replace the structure of the original model, the B-pillar reinforcing plate 1 ~ reinforcing plate 4 can be removed. The B-pillar structure will only consist of the B-pillar inner plate and the B-pillar outer plate using tailor welded blanks, as shown in Figure 8

figure 8. Description of multi-objective optimization problem using B-pillar after tailor welded blanks

2.1 design objectives and variables

in side impact safety analysis, the safety performance of vehicle body structure in side impact is usually evaluated by indexes such as side intrusion volume, intrusion speed and intrusion form. In this study, the maximum intrusion um of the B-pillar, the intrusion LW at the waist line of the inner plate of the B-pillar and the intrusion speed VW were selected as the design objectives to evaluate the crashworthiness. At the same time, the total mass m of B-pillar is selected as the lightweight index. According to the evaluation method of the American Highway Safety Insurance Association on the structural deformation of the B-pillar, combined with the structural size of the vehicle, under the condition of ensuring that the structural deformation of the B-pillar is at the "excellent" level, the inner plate of the B-pillar still has 180mm deformable space, so um ≤ 180mm is taken. In side impact, the acceptable range of intrusion speed of side structure is generally between 7~10m/s. However, relevant literature shows that controlling the intrusion speed of the side structure below 8m/s can better meet the requirements of passenger safety performance in side impact [9]. Therefore, the optimization target of the intrusion speed VW at the waist line of the B-pillar inner plate is set between 7~8m/s

as shown in Figure 7, this paper selects three factors that have a significant impact on crashworthiness and lightweight: the sheet thickness t1 and T2 of the two parts of the tailor welded blanks and the height h of the tailor welded seam (the distance between the seam and the lower edge of the threshold). The value range of T1 and T2 is selected according to the general thickness of automobile steel plate, that is, the value range is 0.8~2.5mm. For the protection of human chest, the height h of the weld is selected at the corresponding B-pillar part below the chest of the dummy in the normal sitting position and above the threshold, with a value range of 136~455mm

2.2 multi-objective optimization problem

a typical multi-objective optimization problem can be defined as

from the above, the mathematical model of the multi-objective optimization problem in this paper can be defined as

3 multi-objective optimization process

the multi-objective optimization program in this study includes: experimental design, mathematical model regression and multi-objective genetic algorithm, so the whole multi-objective optimization process can be divided into three steps: first, Obtain sufficient sample points through the experimental design; Then, based on these sample points, the mathematical approximate model is obtained, and the fitting accuracy of the mathematical model is evaluated; Finally, multi-objective genetic algorithm is used to optimize the mathematical model

3.1 orthogonal experimental design

experimental design is to design how to select a limited number of sample points in the whole design space to make it reflect the characteristics of the design space as much as possible [14]. Orthogonal experimental design is a scientific method to arrange multi factor experiments by using a set of existing specifications, which will leave fatal safety hazards -- orthogonal tables, and to statistically analyze the experimental results to find out a better experimental scheme

in order to obtain sufficient design samples to establish mathematical models, L9 (34) orthogonal experimental design table was selected in this study. In the design space, three levels of design variables h, T1 and T2 are selected respectively, and nine simulations are carried out using LS-DYNA. The corresponding simulation values of design objectives VW, m, LW and um are calculated, as shown in Table 1. All sample values will be used in the coefficient calculation of the mathematical model in the next step

3.2 mathematical model regression

in response surface method, in order to obtain the design variables and objectives expressed