Vehicle Seat Structure Optimization in Front and Rear Impact

Seat was one of the important parts for occupant safety during impact. The occupant injury characteristics is vital for better seat development. To improve occupant safety during impact, the research on the seat structure optimization in front and rear impact was conducted in this paper. Dummy-seat finite element simulation model was established and analyzed by using HyperMesh and LS-DYNA software. The model was verified with test data before further application. Then, the model was simulated to determine its performance on the head, chest and neck injury of the dummy in the frontal impact and rear impact. The simulation results showed that the original model cannot provide effective protection according to CNCAP regulation. Modification should be done. Based on previous study, seat side plate, lower bracket under cushion, and back lock member was modified by implementing orthogonal experiment design method, to determine the best option. The optimized solution A4B2C2 was gained through range analysis and integrated balance method. After simulation, chest compression reduced 17.21%, 3ms resultant acceleration reduced 21.16%, dummy neck FX decreased 15.44%, MZ value decreased 3.13%, backrest angle decreased 46.1%. It was indicated that the optimized structure can improve passenger protection. It was illustrated that the model based method combining HyperMesh and LS-DYNA was an effective way for seat development and occupant injury study.


INTRODUCTION
With the development of auto industry, the safety of the car has increasingly become an important research field for modern automobile development design [1,2].
As an important safety component, vehicle seat is a hot spot in the study of automobile safety and it provides a decisive protection for passengers [3].In 2011, Jin [4] systematically introduced the seat safety performance requirements pointing out that seat back should be strengthened, the cushion stiffness improved and headrest redesigned on low-speed crash protection.Yang [5] analyzed that insufficient stiffness of seat cushion is the cause of human body diving in rear collision which results in greater damage on the abdomen in 2012.According to the recent research status abroad, Nicolas [6] established finite element models of multi-body human body and seat to study the safety of the crew in rear crash.Masahide et al. [7] from Japan's Toyota motor crop studied occupant protection during the car crashes.Chen [8] established multi-rigid-body crash model to analyze the seat parameter effect on passenger injury during rear impact and optimize the seat structure.Wang [9] conducted research study on seat strength and stiffness in front crash and optimized seat to provide better passenger protection.
The seat safety research has laid emphasis on the strength of seat and body connection and seat features during frontal crash, and headrest safety and backrest strength in rear impact.Seat safety refers to the ability to prevent vehicle accidents effectively and to reduce the damage of occupants to a minimum at the time of the accident [10].Research on vehicle seat in a front and rear collision mechanism of injury to the occupant can provide theoretical technical support for the seat design, research and development.It can improve vehicle passive safety performance in a collision and have greater significance in traffic safety [11][12][13].
Based on a domestic car seat, a seat-occupant finite element model was established by using HyperMesh and LS-DYNA simulation software to study the passenger injury characteristics during front and rear impact.The main purpose was to analyze the performance of the seat and the improvement and optimization of the structure, thus to improve dummy injury indicators to provide effective protection for occupants and also to provide a method for modern seat design and passenger protection evaluation.

Model Establishment
The CAD geometric model was imported to HyperMesh software and meshed according to engineering experience.In this model, the sheet metal parts using two-dimensional grid method were meshed by quadrilateral element and triangular element.Three-dimensional mesh was used in headrest, backrest, cushions and other special components.Belt model was the combination of one-dimensional multi-rigid-body seat belt element and two dimensional membrane elements.Model grid size was controlled at about 10 mm.Final mesh model is shown in Fig. (1).A total of 30954 nodes and 93421 elements were included in this model.And then, on the basis of this seat model to join Hybrid , 50% male dummy completed the dummy-seat model as shown in Fig. (1).

Front and Rear Impact Setup
For front impact, the analysis was a low speed collision.In accordance with requirements of the low-speed collision, the collision speed was selected 50km/h, at X-axis negative direction.Collision time was 150ms.Acceleration curve is shown in Fig. (2), which was obtained by vehicle collision test.The chest compression, chest 3ms synthetic acceleration data were used as evaluation indexes.
For rear impact, the speed of rear collision was 50hm/h, in X direction.The remaining boundary conditions were the same as the frontal collision simulation.Seat acceleration curve is shown in Fig. (3), which was obtained by the collision test of the vehicle.protection performance of simulation was valid which indicated that the model can be used for simulation [14].

Frontal Impact
Fig. (5) shows that the maximum amount of chest compression was 63.84mm which was higher than the CNCAP requirement (50mm), and the value higher than 50mm period was about 50ms leading to a large amount of dummy chest compression resulting in more serious injuries in chest.3ms synthesis acceleration of chest was 24.67g Fig. (6), meeting CNCAP requirement.Results indicated that modification should be carried out to optimize seat for better front injury protection.

Rear Impact
As shown in Fig. (7), dummy neck X to a maximum force FX was 881N.The value was greater than zero starting at around 104ms, and was largely changed.The neck Xforce FX was over 730N (CNCAP value) and had a longer duration with more than 15ms.This showed that there was larger impact force acting on the dummy neck over a long period of time which could cause greater harm to the dummy neck.Optimization was needed to meet the requirement.
For Z-torque M Z (12.13N•m), the value was greater than 0 after 102ms which could cause some dummy neck injury as shown in Fig. (8).
With the increase in the force of the backrest, the backrest angle was increased Fig. (9).The maximum change of the backrest angle was 22.97°, which could be further optimized.

Orthogonal Design Arrangement
Based on the previous study of Chen [15], the seat structure was modified through three aspects, seat side plate, the lower bracket under cushion and back lock member, by combining the stress analysis during front and rear impact simulation, and industry engineering experience.The factors and level of each factor are shown in Table 1.And a mixed orthogonal table L 8 (4X2 2 ) is shown in Table 2.

Orthogonal Analysis
Based on the experimental arrangement, specific modification was carried out to establish the corresponding model.All the modes were simulated in front and rear impact.The simulation results are shown in Table 3. Results indicated that the occupant injury was reduced after different modification.
In order to determine the primary and secondary sequence of each factor, range analysis was applied.The simulation experiment data and range R, and the referred range R′are shown in Table 4.For chest compression, the sequence was C-A-B, with the best combination of A 2 B 1 C 1 .For 3ms synthetic acceleration, the sequence was B-C-A, with the best combination of A 4 B 2 C 2 .It is indicated that the lower portion had relatively important effect on chest compression and 3ms synthetic acceleration.
For neck F x , the influence order was A-C-B, with the best group A 2 B 1 C 1 .For Neck M Z , the order was A-B-C, with the best group A 4 B 1 C 1 .It was illustrated that the back lock member had major effect on occupant neck injury.
To determine the optimized option, integrated balance method was used.According to the simulation results, 3

Yang et al.
factors had different sequence on the selected occupant injury index.Factor A had larger effect on neck F x and M z .A 4 had the best overall performance.Factor B had larger effect on 3ms synthetic acceleration, and the best option was B 2 .Factor C had larger effect on chest compression, with the best option C 2 .Therefore, the optimized solution was A 4 B 2 C 2 .

Occupant Protection Performance of Optimized Model
On the basis of the improvement structure, the optimized seat finite element model was established.Improved model analysis was carried out again in LS-DYNA, and compared with the simulation results of the original model.Chest compression, 3ms resultant acceleration, neck X-force, Zmoment and seat back angle were compared before and after optimization, as listed in Table 5. 4, the dummy injury indicators decreased obviously.Among them, the backrest angle, chest compression and 3ms synthetic acceleration decreased 17.21% and 21.16%, respectively.This was due to the increase in optimized sides' strength, so that the seat could withstand a greater impact in front collision.Due to the modification on back lock member and center hinge, the seat can reduce force and torque on neck.The dummy neck F X decreased 15.44%, M Z value decreased 3.13%, and its angle variation also decreased.The dummy rebound decreased when impacted by an external force.Results illustrate that the optimized structure can improve occupant protection during front and rear impact.

CONCLUSION (1)
Dummy-seat model was established by HyperMesh and LS-DYNA software.Analysis was conducted to determine the performance of seat in dummy protection during front and rear impact.Results Fig. (1).FE model seat-dummy.