【作者】 石向南；

【导师】 曹立波；

【作者基本信息】 湖南大学 ， 车辆工程， 2014， 博士

【摘要】 道路交通事故是造成人类伤亡的重要因素之一。每年有120万人死于道路交通事故,约2000万人遭受终身的伤残性伤害。随着汽车安全技术的不断发展以及乘员安全性问题逐步受到重视,驾乘人员的损伤防护技术取得了很大的进展,特别是50百分位成人的乘员保护技术已趋成熟。然而,相对于50百分位的成人,老龄人群和肥胖人群等特殊群体在汽车碰撞事故中更易遭受重伤和死亡。但是现有的碰撞测试假人和人体有限元模型等损伤评价工具,并没有考虑到人群中年龄和肥胖程度组成的多样性。由于医疗条件以及物质生活水平的提高,世界各国的老龄以及肥胖人群的所占比重呈现出逐渐增加的趋势。因此,老龄和肥胖乘员有限元模型的开发以及损伤机理研究已经成为汽车安全研究领域迫切需要解决的重要课题。本文在文献研究的基础上,利用人体测量学方法、统计学方法、网格变形技术、计算机仿真以及优化技术,分别对老龄和肥胖乘员有限元模型的建模方法以及损伤机理等进行了研究。研究工作的主要内容及创新点如下：1、参数化的肋骨腔几何模型的建立。本研究通过收集大量的临床医用CT(Electronic Computer X-ray Tomography Technique)扫描数据,并进行了数据分割、指定位置标志点坐标数据获取以及Procrustes分析,建立了同一坐标系下的人体肋骨腔标志点空间坐标数据集。在此基础上,对标志点空间坐标数据集进行了主成分分析和多变量回归分析,建立了基于人体特征的参数化肋骨腔几何模型,并用于研究年龄、性别、身高以及BMI (Body Mass Index)对肋骨腔几何模型大小和形状的影响。结果表明,这些人体特征参数对肋骨腔几何模型均有显著性影响；同时,利用统计学方法建立了基于人体特征参数的肋骨横截面面积的线性混合模型。研究发现,除BMI外,其它的人体特征参数对肋骨横截面面积均具有显著性影响。该研究为后续建立参数化的人体有限元模型提供了三维数模基础。2、开发出一套基于径向基函数理论的参数化网格变形技术。本研究利用基于样条曲线的几何变形方法对标志点空间坐标进行逼近,并用于对标志点坐标数据的分析,最终把参考空间里的坐标基准位置映射到新空间的目标位置。坐标映射的过程包括整体的平移、刚度旋转、缩放以及局部弯曲。通过该网格变形技术可以将基准模型与不同的目标模型联系起来,并运用该技术对人体骨骼模型进行了网格变形研究。结果表明该方法快捷有效,可为后续开展参数化人体有限元模型的研究提供技术基础。3、提出了建立任意指定人体特征参数的乘员有限元模型的流程,并完成了指定人体特征的乘员有限元模型的验证工作。本研究首先利用丰田公司开发的第四代人体有限元模型,建立了人体外部轮廓与内部骨骼结构的混合几何模型；在此基础上,利用本文开发的网格变形技术,提出了一套建立参数化人体有限元模型的流程。其后,利用该流程建立了一个预先指定了人体特征的乘员有限元模型,并对该模型的有效性进行了验证。结果表明利用本文提出参数化人体有限元建模流程开发的模型具有较高的生物逼真度,因而可利用该流程自动生成任意指定人体特征参数的乘员有限元模型。这为研究汽车乘坐环境对特殊人群碰撞安全性的影响,提供了有效的新思路。4、研究并分析了老龄人群的损伤机理以及年龄对人体肋骨骨折损伤风险的影响。为分析年龄对人体损伤的影响,本研究建立了指定人体特征参数条件下的胸部有限元模型。该模型经过严格的仿真碰撞验证,并与两具含有不同人体特征参数的尸体样本胸部撞击试验进行了对标。结果显示,建立的基于年龄特征的参数化胸部有限元模型具有较高的仿真度。为了探究与人体年龄相关的材料参数、几何模型以及组织结构属性对胸部生物力学特性的影响,对验证后的胸部有限元统计学模型进行参数化研究。结果表明,由于年龄增长而引起的胸部几何模型、肋骨材料参数以及肋骨骨密质厚度的变化量,导致胸部的线性刚度特性、肋骨骨密质的Von Mises应力和第一主应变呈现出不同的变化趋势；特别是,在冲击锤的撞击作用下,随着人体年龄的增长,胸部几何模型、肋骨材料参数以及肋骨骨密质厚度产生的复合影响会使得肋骨骨密质的Von Mises应力和第一主应变逐渐升高。考虑到随着人体年龄的增长,人体骨骼组织的失效应力和失效应变将逐渐减小,导致老龄人群在碰撞过程中将面临更大的胸部肋骨骨折损伤风险。5、研究并分析了肥胖乘员的损伤机理以及肥胖对乘员损伤风险的影响。基于目前文献中仅存的少量肥胖人群碰撞试验数据,对上述新建的参数化的人体有限元模型分别进行了腹部安全带动态加载验证和尸体台车碰撞仿真碰撞验证。结果表明利用本文提出参数化人体有限元建模流程开发的肥胖人体模型具有较高的生物逼真度。在此基础上,再次利用上文新建的参数化人体有限元模型和网格变形技术,对人体各局部区域逐步变形,得到BM1分别是25kg/m2、30kg/m2、35kg/m2和40kg/m2的人体有限元模型,并用于研究在车辆正面碰撞过程中人体肥胖特征对乘员损伤响应的影响。结果表明,与非肥胖乘员相比,肥胖乘员额外增加的人体质量并由此导致了安全带更差的佩戴路径,使得其在正面碰撞事故中遭受更高的胸部以及下肢的损伤风险。本研究的结论可为以后的乘员约束系统设计开发提供相关的有用信息及理论依据。更多还原

【Abstract】 Motor vehicle crashes (MVCs) are one of the leading causes of injuries and deaths among population. There are approximately1,200,000deaths and20,000,000injured in traffic accidents worldwide every year. Along with the development of vehicle safety technology and the increasing attention on the issue of occupant safety in cars, occupant safety research on the injury prevention of older and obese occupant has been significantly improved over the past century, especially for the mid-size, yong male occupants and child safety reaearch. However, older and obese populations are often at increased risk of death and serious injury compared with mid-size young male, which is especially well documented in motor-vehicle crashes (MVCs). Unfortunately, Injury assessment tools considering aging and obesity effects and capable of simulating the geometrical and biomechanical variations among the population are not currently available. As a result, older and obese populations are generally not considered in current safety design process. Due to longer life expectancy and increasing life level, the projected increase of aging and obese population in the world further necessitate future efforts to develop more advanced injury assessment tools to evaluate safety designs for mitigating injuries that these aging and obese populations may be susceptible to. Based on the literature review, this paper focused on the study of parametric humen finite element model and injury epidemiology and injury prevention of older and obesity population through using human anthropometry, statistic analysis, mesh morphing, computer simulation and optimization.1. A parametric statistical skelekon model was developed. In this study, data from clinical thorax CT scans from89subjects were obtained. Rib cage geometries were extracted from CT images through image segmentation. Landmarks were manually identified on each rib using Hypermesh11.0(Altair, U.S.). After landmarks were collected for all subjects, the whole dataset was processed by a series of numerical analyses, including Procrustes analysis, principal component analysis (PCA) and multivariate regression analysis. The purpose of Procrustes analysis is to align the rib cage landmarks into the same center and orientation. After that, a statistical human body geometry model was developed. Using this model, detailed3D skeleton geometries of human body can be predicted by stature, gender, age, BMI, and some other human parameters. The results show that all of these parameters showed strong effects on rib cage geometry. Also, a linear mixed model was built to test whether age, gender, stature, BMI, rib number, and cross-section location along each rib are significant predictors for rib cross-sectional area. Except for BMI, all parameters also showed significant effects on rib cross-sectional area using a linear mixed model. Statistical results also indicated that the age and sex effects on rib cross-sectional areas varied significantly by the number and location of the ribs. This statistical rib cage geometry model can serve as a geometric basis for developing a parametric human FE model for quantifying effects from different human attributes on thoracic injury risks.2. Development of an automatic mesh morphing method based on radial basis function (RBF). In mesh morphing, RBF will be used as a3D interpolation and smoothing method, in which an interpolation function is calculated based on the data from the corresponding landmarks on the geometry model and the baseline FE model. Based on the landmark locations, a transformation field was calculated by RBF, which was used to transform the baseline mesh into the geometry described by the landmarks predicted by the statistical geometry models. Comparison showed that the morphed rib cage model can accurately represent the target geometry as well as maintain a comparable mesh quality of the baseline model. This algorithm can be applied programmatically, making them an efficient tool for generating a parametric human FE model.3. Development of a procedure for building a parametric human FE model and validation of this model. Using THUMS4as the baseline model, a combined model including the outer body shape model and inner skeleton model was built. Based on the combined model and mesh morphing method, a procedure for building a parametric human FE model was developed and used to build several appointed humen FE model. After that, the morphed human FE models were validated against cadaver tests under belt loading on the abdomen and whole-body frontal crash tests. This study demonstrated the feasibility of using this procedure to build a parametric human FE model. This validated parametric human FE model provides a useful tool to investigate the vehicle inner environment effects and to develop restraint system design guidelines for aging and obese occupants.4. The injury epidemiology of older population and the effect of age on rib fracture were studied and analyzed in this paper. Even though the literature has shown that the thorax injury tolerance reduced with aging, the exact mechanisms of the reduction (material or structural) are not well described. Furthermore, a comprehensive description of the age-related factors that influence thoracic stiffness is unavailable, yet this issue is of critical importance for designing "age-friendly" restraints. Therefore, in this study the effects from age-related thoracic geometric changes and bone material property changes on thorax stiffness and tolerance will be quantified in a parametric study using the newly-developed parametric thorax FE model. A Design of Experiment (DoE) will be conducted with varying thorax geometries and material properties according to different ages under hub loading condition. Results showed that age-related thoracic geometric changes, rib material property and cortical bone thickness changes make some different vary trend on thorax stiffness, rib cortical bone Von Misese stress and first component strain, expecially for the last two index, they will be at lower value as age increased. It means that eldly occupants will have much higher rib fracture injury risk in motor vehicle crashes.5. The injury epidemiology and injury prevention of obese population were studied and analyzed in this paper. Based on mesh morphing method and paremetric human FE model, the THUMS4model was morphed into human models with different BMIs. A parametric study was conducted using these morphed models to investigate the obesity effects on the occupant responses in MVCs. Results showed that obese occupants sustained higher risks of thorax and lower extremity injuries than non-obese occupants due to a combination of increased mass and poor belt fit for obese occupants. The conclusions from this research provide valuable information for future designs of obese occupant restraint systems.更多还原