Detection of Cognitive and Visual Distraction Using Radial Basis Probabilistic Neural Networks

This paper suggests a real-time method for detecting a driver’s cognitive and visual distraction using lateral driving performance measures. The algorithm adopts radial basis probabilistic neural networks (RBPNNs) to construct classification models. In this study, combinations of two driving performance data measures, including the standard deviation of lane position (SDLP) and steering wheel reversal rate (SRR), were considered as measures of distraction. Data for training and testing the RBPNN models were collected under simulated conditions in which fifteen participants drove on a highway. While driving, they were asked to complete auditory recall tasks or arrow search tasks to create cognitively or visually distracted driving periods. As a result, the best performing model could detect distraction with an average accuracy of 78.0 %, which is a relatively high accuracy in the human factors domain. The results demonstrated that the RBPNN model using SDLP and SRR could be an effective distraction detector with easy-to-obtain and inexpensive inputs.     

Feature-Based Cad System Buildup for Passenger Airbag Design

The passenger airbag (PAB) requires a large volume and fast deployment because of the large distance between the dashboard and the passenger. And various shapes and sizes of the PAB are required depending on the type of vehicle. However, since the PAB modeling process for each design change is complicated and time consuming, the design parameters of the PAB could not be well investigated. In this study, a unique feature-based CAD system has been proposed that easily constructs PAB CAD model and then generates PAB FE model for collision analysis. Main keypoints and widths of PAB that determine the shape and size have been extracted by analyzing the geometric-feature of airbag. The PAB CAD model can be easily constructed by inputting keypoints and widths information. Then, from the constructed PAB CAD model, the PAB FE model is automatically generated. Finally, the generated PAB FE model can be directly employed for collision analysis, thereby reducing the modeling time of the PAB and enabling efficient parametric studies on design.     

Effects of Residual Ash on Dpf Capture and Regeneration

To study the effects of residual ash on the capture and regeneration of a diesel particulate filter (DPF), repeated capture and complete regeneration experiments were conducted. An engine exhaust particulate sizer was used to measure the particle size distribution of diesel in the front and back of DPF. Discrepancies in the size distribution of the particulate matter in repeated trapping tests were analyzed. To achieve complete DPF regeneration, a DPF regeneration system using nonthermal plasma technology was established. The regeneration carbon removal mass and peak temperatures of DPF internal measuring points were monitored to evaluate the effect of regeneration. The mechanism explaining the influence of residual ash on DPF capture and regeneration was thoroughly investigated. Results indicate that the DPF trapping efficiencies of the nuclear-mode particles and ultrafine particles have significant improvements with the increase quantity of residual ash, from 90 % and 96.01 % to 94.17 % and 97.27 %, respectively. The exhaust backpressure of the DPF rises from 9.41 kPa to 11.24 kPa. Heat transfer in the DPF is improved with ash, and the peak temperatures of the measuring points accordingly increase. By comparing the regeneration trials, the elapsed time for complete regeneration and time difference for reaching the peak temperature between adjacent reaction interfaces are extended with increased quantity of ash. The carbon removal mass rises by 34.00 %.     

Influence of the geometric parameters of the vehicle frontal profile on the pedestrian’s head accelerations in case of accidents

The goal of this paper is to determine how the geometry of the vehicle’s frontal profile is influencing the pedestrian’s head accelerations (linear and angular) in car-to-pedestrian accidents. In order to achieve this goal, a virtual multibody dummy of the pedestrian was developed and multiple simulations of accidents were performed using vehicles with different frontal profile geometry, from different classes. The type of accidents considered is characteristic for urban areas and occur at relatively low speed (around 30 km/h) when an adult pedestrian is struck from the rear and the head acceleration variation are the measurement of the accident severity. In the accident simulation 3D meshes were applied on the geometry of the vehicles, in order to define the contact surface with the virtual dummy, similar with real vehicles. The validation of the virtual pedestrian dummy was made by performing two crash-tests with a real dummy, using the same conditions as in the simulations. The measured accelerations in the tests were the linear and angular accelerations of the head during the impact, and these were compared with the ones from the simulations. After validating the virtual model of the car-to-pedestrian accident, we were able to perform multiple simulations with different vehicle shapes. These simulations are revealing how the geometric parameters of the vehicle’s frontal profile are influencing the head acceleration. This paper highlights the main geometric parameters of the frontal profile design that influence the head injury severity and the way that the vehicles can be improved by modifying these parameters. The paper presents an approach to determine the “friendliness” of the vehicle’s frontal profile in the car-to-pedestrian collision.     

Tire wear estimation based on nonlinear lateral dynamic of multi-axle steering vehicle

This paper presents a novel nonlinear dynamic model of a multi-axle steering vehicle to estimate the lateral wear amount of tires. Firstly, a 3DOF nonlinear vehicle dynamic model is developed, including dynamic models of the hydropneumatic suspension, tire, steering system and toe angle. The tire lateral wear model is then built and integrated into the developed vehicle model. Based on the comparison of experimental and simulation results, the nonlinear model is proved to be better than a linear model for the tire wear calculation. In addition, the effects of different initial toe angles on tire wear are analyzed. As simulation results shown, the impact of the dynamic toe angle on the tire wear is significant. The tire wear amount will be much larger than that caused by normal wear if the initial toe angle increases to 1° - 1.5°. The results also suggest that the proposed nonlinear model is of great importance in the design and optimazation of vehicle parameters in order to reduce the tire wear.     

Schlieren,Shadowgraph, Mie-scattering visualization of diesel and gasoline sprays in high pressure/high temperature chamber under GDCI engine low load condition

Three visualization methods, Schlieren, Shadowgraph, and Mie-scattering, were applied to compare diesel and gasoline spray structures in a constant volume chamber. Fuels were injected into a high pressure/high temperature chamber under the same in-cylinder pressure and temperature conditions of low load in a GDCI (gasoline direct injection compression ignition) engine. Two injection pressures (40 MPa and 80 MPa), two ambient pressures (4.2 MPa and 1.7 MPa), and two ambient temperatures (908 K and 677 K) were use. The images from the different methods were overlapped to show liquid and vapor phases more clearly. Vapor developments of the two fuels were similar; however, different liquid developments were seen. At the same injection pressure and ambient temperature, gasoline liquid propagated more quickly and disappeared more rapidly than diesel liquid phase. At the low ambient temperature and pressure condition, gasoline and diesel sprays with higher injection pressures showed longer liquid lengths due to higher spray momentum. At the higher ambient temperature condition, the gasoline liquid length was shorter for the higher injection pressure. Higher volatility of gasoline is the main reason for this shorter liquid length under higher injection pressure and higher ambient temperature conditions. For a design of GDCI engine, it is necessary to understand the higher volatility of gasoline.     

天然气发动机冷却系统校核

商要:发动机冷却系统的作用是使发动机的温度在所有工况下均保持在适当的范围内。冷却系统不仅要防止发动机过热,也要防止发动机过冷。本文主要是对某重卡车型的发动机冷却系统进行了校核计算,为整个车型的匹配计算提供数据支持。     

基于SLP和生产物流的F公司车间设施布局改善

一个良好的设施布置设计可以使物料搬运的成本降低,使人员走动的距离缩短,从而使运作效率提高。运用SLP技术,对F公司CG片生产车间布局进行分析改善。首先,对该车间的生产流程和物流量进行分析,发现存在物料重复搬运、线路迂回、物流作业流程复杂等问题。然后,运用物流关系和作业单位相互关系图,经过尽可能小的调整,在能适应产品生产、工艺、流程等方面的情况下,降低物料搬运成本,求得合理的布置,从而使生产物流运作顺畅,达到提高整个生产系统经济效益最优的目的。最后,以物料搬运成本和厂房租金为评价指标,说明新布局的优越性。     

基于等效应力的试验载荷谱开发与研究

以某型轿车稳定杆为研究对象,采集试车场道路载荷谱并对其进行统计分析,在传统基于广义力的载荷谱编制方法基础上,考虑结构的真实损伤计算,提出一种基于等效应力的编谱方法。通过CAE和台架试验验证可知,寿命与失效位置与实车路试结果基本一致。新方法相比传统方法能更准确预估损伤等效结果,减少试验频次,更快速地模拟出汽车结构件实际受载情况,试验加速效果明显。     

竖向预应力作用效果的数值模拟与预应力损失的试验研究

对箱梁竖向预应力筋张拉引起的混凝土竖向应力和螺纹钢筋的应力进行了有限元计算和现场测试分析,测试结果表明预应力损失离散性很大,而且预应力损失可能高达50%.这是由于箱梁的高度有限,竖向预应力筋的长度都比较短,因而在达到张拉控制应力时高强精轧螺纹粗钢筋的伸长量有限.目前的竖向预应力锚固技术尚存在不足,在锚固时稍有不慎造成钢筋回缩量偏大,很容易造成预应力损失.