Aerodynamic design optimization of rear body shapes of a sedan for drag reduction

This study proposes an aerodynamically optimized outer shape of a sedan by using an Artificial Neural Network (ANN), which focused on modifying the rear body shapes of the sedan. To determine the optimization variables, the unsteady flow field around the sedan driving at very fast speeds was analyzed by CFD simulation, and fluctuations of the drag coefficient (C _D ) and pressure around the car were calculated. After consideration of the baseline result of CFD, 6 local parts from the end of the sedan were chosen as the design variables for optimization. Moreover, an ANN approximation model was established with 64 experimental points generated by the D-optimal methodology. As a result, an aerodynamically optimized shape for the rear end of the sedan in which the aerodynamic performance is improved by about 5.64% when compared to the baseline vehicle is proposed. Finally, it is expected that within the accepted range of shape modifications for a rear body, the aerodynamic performance of a sedan can be enhanced so that the fuel efficiency of the sedan can be improved. The YF SONATA, a sedan manufactured by Hyundai Motors Corporate, played a major role in this research as the baseline vehicle.     

Study of an optimized gas strut layout to improve opening and closing quality

The most important factor in gas strut design is determining an optimized layout. If the layout is not optimized, vehicle operators will have a suboptimal experience when opening and closing the tailgate. A poor layout of the gas struts causes operators to work excessively when they open/close the tailgate, and vehicle owners will incur additional expenses due to deterioration in the body quality of the vehicle. Thus, an optimized gas strut layout is very important, even if it does not seem interesting. This paper describes the tailgate operation process and focuses on determining an optimized gas strut layout for opening/closing the tailgate easily.     

Fuel consumption of fuel cell hybrid vehicles considering battery SOC differences

Fuel cell hybrid vehicles (FCHVs) have become one of the most promising candidates for future transportation due to current energy supply problem and environmental problem. Fuel economy is an important factor in FCHVs. In order to properly evaluate the fuel economy of an FCHV, the initial battery state of charge (SOC) and the final battery SOC have to be identical so that the effect of the battery energy usage on the fuel economy is neglected. In the simulation or in the real driving, however, the final battery SOC is usually different from the initial battery SOC, and the final battery SOC often depends on the power management strategy. To consider the difference between the two battery SOC values, the concept of equivalent fuel consumption is presented by two methods. One is based on the relationship between delta SOC and delta fuel consumption, and the other is based on the optimal control theory. Two rule-based power management strategies for an FCHV are presented, and for each strategy, the fuel economy is evaluated based on the two methods. The characteristics of the two methods are discussed and compared, and the superior one is selected based on the comparison.     

FE modeling method for a headform used in pedestrian protection simulations

Finite element models of headforms are used in experimental simulations of pedestrian protection. In this study, a quick and accurate method for FE modeling of the headforms was developed. This method entailed the initial definition of the dimensional parameters for the mass, centroid, and inertial moment properties of the headform. The equations governing these properties were constructed using the dimensional parameters as design variables. The dimensional parameters meeting the requirements of the relevant regulations were obtained by solving these three equations. A design optimization model was constructed for the material parameters of the outer part of the headform. In this model, the parameters of the material used in the FE model were considered as design variables; the difference between the peak acceleration in a side-impact simulation test and the average value of the regulated acceleration range was used as the objective function; the first-order natural frequency, which was required to be greater than 5,000 Hz, was defined as one of the constraints; the peak drop acceleration, which was required to be within the regulated range of values, was defined as the second constraint. The material parameters were obtained by solving the optimization model. These material parameters meet the dynamic requirements of the regulations for headforms. Based on these three parameters, an FE model of a headform can be constructed quickly and accurately.     

Topology optimization of gas flow channel routes in an automotive fuel cell

An efficient topology optimization method for fluid-structure problems was developed in an effort to determine the optimum flow channel route in a fuel cell bipolar plate from first principles. This study describes the derivation and solution of new mathematical equations for topology optimization combining a density-based algorithm, the interpolation method of moving asymptotes (MMA), and the incompressible Navier-Stokes equation with a term representing the chemical reaction between hydrogen and the catalyst. The present method is based on the finite element method with a newly developed reaction rate equation. In this model, a topology variable of 0 represents viscous flow, whereas a value of 1 indicates porous flow. The flow velocity and pressure were obtained from the Navier-Stokes equation and constraints and element matrices for sensitivity analyses during the optimization. MMA was utilized to calculate the optimum flow routes in the design domain. The influence of the key design parameter q and the pressure drop on the optimum topology were also investigated. The channel topology became smoother with decreasing q, and the number of channels increased with increasing pressure drop.     

Comparison of the optimum designs of center pillar assembly of an auto-body between conventional steel and ahss with a simplified side impact analysis

This study compares the optimum designs of center pillar assembly with advanced high-strength steel (AHSS) to that of conventional steel for crashworthiness and weight reduction in side impacts. A simplified side impact analysis method was used to simulate the crash behavior of the center pillar assembly with efficient computing time. Thickness optimization aims to perform an S-shaped deformation of the center pillar toward the cabin to reduce the injury level of a driver in a crash test. Center pillar members were regarded as an assembly of parts that are fabricated with tailor-welded blanks, and the thickness of each part was selected as a design variable. The thickness variables of parts that have significant effects on the deformation mechanism were extracted as the main design variables for thickness optimization based on the results of a sensitivity analysis with design of experiments. The optimization condition was constructed to induce an S-shaped deformation mode and reduce the weight of the center pillar assembly. An optimum design was obtained after several iterations with response surface methodology (RSM). Optimization was first performed with conventional steel and then with AHSS with the same procedure to optimize the crashworthiness of the center pillar assembly. After thickness optimization, optimum designs were applied to the full vehicle analysis to evaluate the validity of the optimization scheme with the simplified side impact analysis method. Then, the crashworthiness of optimum designs with conventional steel and AHSS were compared using the full vehicle analysis. This comparison demonstrates that AHSS can be more effectively utilized than conventional steel to obtain a lightweight design of an auto-body with enhanced crashworthiness.     

Prediction of the dimensional deformation of the addendum and dedendum after the warm shrink fitting process using a correction coefficient

The warm shrink fitting process is generally used to assemble automobile transmission parts (shafts/gears). However, this process causes a deformation in the addendum and dedendum of the gear depending on the fitting interference and gear profile, and this deformation causes additional noise and vibration between the gears. To address these problems, the warm shrink fitting process is analyzed by considering the error in the dimensional deformation of the addendum and dedendum found when comparing the results of a theoretical analysis and finite element analysis (FEA). A correction coefficient that reduces this error is derived through an analysis of the difference in the cross-sectional area between the shapes used for the theoretical analysis and that of the actual gear, and a closed-form equation to predict the dimensional deformation of the addendum and dedendum is proposed. The FEA method is proposed to analyze the thermal-structural-thermal coupled field analysis of the warm shrink fitting process (heating-fitting-cooling process). To verify the closed-form equation using the correction coefficient, measurements are made of actual helical gears used in automobile transmissions. The results are in good agreement with those given by the closed-form equation.     

重型普通载货汽车典型用户的整车行驶循环工况研究

以重型普通载货汽车为研究对象,针对用户使用情况进行了调研和实车试验,并应用数理统计和多参数统计理论主成分分析与聚类分析方法,解析出重型普通载货汽车典型用户行驶循环工况。基于所提取的循环工况进行了综合油耗的仿真分析,结果表明,仿真计算油耗与用户实际使用油耗一致,证明了所提取重型普通载货汽车整车行驶循环工况的正确性。     

Development of Vehicle Electronic Diagnostic System Based on AUTOSAR

采用汽车开放系统架构标准AUTOSAR的开发方法和诊断架构,开发了基于Freescale HCS12X处理器的汽车电子故障诊断系统,并进行了仿真实验.结果表明,该系统能及时准确地诊断出故障,基于AUTOSAR的诊断系统是可配置的,能够适应多种情况.     

Parameters Optimization for HEV Based on Simulated Annealing Particle Swarm Algorithm

分别采用多层次参数扫描(MLPS)算法和模拟退火粒子群优化(SAPSO)算法对并联式混合动力车逻辑门限控制策略的参数进行优化.将优化后的车辆以TEST-CITY-HWY测试循环进行仿真,并将结果与优化前的车辆的仿真结果进行对比.结果表明,经MLPS算法优化后,燃油消耗和HC与NOx排放分别下降了11.98％、6.01％和4.03％,但CO排放增加了25.18％;经SAPSO算法优化后,燃油消耗和HC、CO与NOx排放分别下降了13.61％、9.57％、27.78％和18.53％,且电池荷电状态(SOC)比MLPS优化略高.说明SAPSO算法在混合动力车控制参数优化效果上明显优于MLPS算法.