【作者】 李娜；

【导师】 柳建新；

【作者基本信息】 中南大学 ， 生物医学工程， 2010， 博士

【摘要】 汽车和高速交通的发展让人们的生活越来越方便快捷,也给人们带来了安全隐患。全世界每年死于交通事故的人数为120万,在美国,车祸死亡已经成为儿童和年轻人死亡的第一原因。在中国,随着汽车的普及和车速的加快,汽车的快速和重量成为所有乘员和行人的潜在安全威胁。据公安部交管局统计,仅2009年上半年全国发生交通事故107193起,造成29866人死亡,128336人受伤,直接财产损失4.1亿人民币；根据美国2005年的死亡年龄和死亡原因统计数据显示,由于交通事故而死亡的人数主要分布在3岁到34岁之间,由此产生的非正常死亡人数为4万人,高居该年龄段总死亡人数的首位,其中脑部、颈部和胸部的损伤造成的死亡是致命的主要原因。研究头-颈部损伤机理是对交通事故人员损伤进行预测的有效手段,在安全带作为加强约束技术在新型汽车中得到应用以后,这一研究意义更为重要。数学模型是人体力学创伤研究的有效工具,能够预测创伤发生时的人体响应,弥补了动物和尸体实验不可重复、不可测量,以及价格昂贵的缺陷,更为重要的是,数学模型是验证动物或尸体实验数据、预测乘员碰撞损伤的唯一方法。包括有限元分析在内的数学模型对乘员和交通工具进行模拟,基于类似人体测量测试设备(Anthropometric Test Devices, ATD)的刚性多段系统对汽车碰撞中的人体力学响应仍然难以精确预测,因此各种碰撞中实际情况需要更为详细的人体组织和骨骼的生物材料特性来进行模拟。有限元模型用于分析人体复杂的几何结构和材料特性的最适宜方法,该方法以往的局限性在于几何结构的简化和生物材料特性的不精确,这是由于当时的计算机发展水平不够和动物或尸体实验数据不够而造成。本文从有限元模型的头骨解剖学几何精度的提高入手,建立头-颈部模型,研究头颈部损伤的生物力学原理,深入到组织层面进行分析并模拟汽车运行和碰撞环境下的头-颈部损伤实验,能大大增强模型对头骨骨折、脑部和颈部损伤程度的预测精度。有限元分析采用HYPERMESH、LS-DYNA等软件,显性算法反映碰撞中的巨大形变,拉格朗日动态有限元显性积分方法广泛应用于动态碰撞模拟和非线性结构分析。有限元分析的隐式求解法是模拟计算的经典算法,针对其存在的速度缓慢的不足,本论文提出了更适合生物软组织特性和高速碰撞大形变模拟的显式迭代改进算法,经过多次设置与尝试,生成了稳健的人体头颈部有限元模拟的方法。提出了采用非线性粘弹性材料对脑组织进行求解的方法,针对广泛应用于有限元模型分析的线性粘弹性材料求解方法,由于其表达加载速度范围的局限,我们将之进行改进,得到了高效的多应变率的脑组织求解方法。对颈部鞭甩式损伤进行分析,并细化到小关节囊部位,建立了更细节化的颈部有限元模型,实现了基于显性求解器的颈部有限元模型模拟。实现了对脑组织实验的精确计算和拟合,并通过对一系列拟合数据的实验证明该方法的精确性。将多目标遗传算法和有限元分析相结合,试验结果证实该方法的模拟结果令人满意。本文提出的优化方法对所有脑材料力学研究团队所汇报的数据都可得出结论,不同的实验加载条件,如蠕动压缩、平移剪切,不同的脑材料试件形状,如圆柱形、立方体,都可以进行优化反求。人体有限元模型设计者最为关注的应变率问题,本文首次定量地总结了应变率与应力应变曲线、损伤响应的关系,并提出了两个公式,第一是粘弹性参数β与应变率的线性关系,第二是粘弹性模量G与应变率的幂函数关系。优化反求不仅仅能够对脑材料的非线性粘弹性特性进行完整的表达,还能满足有限元模型设计者对不同加载速率条件下的模型设计要求。本文将优化结果放在头部有限元模型中进行验证,选用LS-DYNA的第77号材料模型,得到了令人满意的模拟结果。更多还原

【Abstract】 The development of automotive and high-speed traffic made people’s lives more convenient, following security problem. The death in traffic accidents around the world are about 120 million every year, in the United States, the death in car crash is the major fatality for children and young men. In China,107193 traffic accident happened,29866 people death,128336 people injury in the first half of 2009, direct financial loss 410,000,000 Chinese Yuan. Statistic also show that traffic accidents in United States is the first major cause of non-normal deaths in the age of 3 to 34 years old, which is about 40000 people dead in car crash in 2005, the head injury, neck injury and thorax injury are the major cause of death.Head and neck injuries are the most frequent severe injury resulting from traffic accidents. Neck injury mechanisms are difficult to study experimentally due to the variety of impact conditions involved, as well as ethical issues, such as the use of human cadavers and animals. Finite element analysis is a comprehensive computer aided mathematical method through which human head and neck impact tolerance can be investigated. A representative finite element human neck model would allow the assessment of the injurious effects of different impact conditions and enable the development of injury protection criteria in the automotive industries. This paper reviewed the articles regarding to using finite element analysis (FEA) model to study neck injury mechanisms and methodology validation. A three-dimensional finite element model established based on the human anatomical radiological image data can be implemented into the finite element analysis software including PAM-CRASH, NASTRAN, LS-DYNA etc. After comparisons and necessary adjustments are made to the model, the model will be able to predict the risk of head and neck injuries in a crash event, which is difficult to achieve experimentally. Once validated, the mathematical model can help understanding injury mechanisms and quantify mechanical parameters related to a specific impact event so that injury tolerances can be formulated.Automobiles allow us to move both farther and faster, however, the speed and weight of vehicles became an inherent safety hazard to their occupants and other road users, more than 1.2 million people die annually on world’s roads. In the U.S., motor vehicle related crashes are the number one cause of death for people in the age group between 4 and 34 and account for more than 40,000 deaths each year. In addition, the brain, neck, and thoracic injury are the main death reason, biomechanical investigation worked for finite element model to develop crashworthy cars, in this paper, the neck finite element model were reviewed for vehicles safety research in the future.Head injury is the most frequent type of injury experienced by all seriously injured road users, especially car occupants. This remains true even with the introduction of enhanced restraint technology into all new cars and with the high levels of seat belt use. Most numerical simulation models of vehicle occupants, including finite element models, are based on rigid multi-segment systems resembling current Anthropometric Test Devices (ATD). As the human body’s responses to impact are still very different from the responses of current ATD’s, models of human body parts or segments are needed to obtain more realistic information about each body part’s responses in crash events and to analyse various impact situations.Mathematical models are valuable tools in the study of trauma. They can be used to predict body response to injury-producing conditions that cannot be simulated experimentally, and they can predict responses that cannot be measured in surrogate and animal experiments. Most important, mathematical modeling is the only means by which valid experimental animal and cadaveric data can be extrapolated to living man. Finite element analysis appears to be the most appropriate technique for analysing the complex geometrical and mechanical properties of human structures. Limitations to finite element modelling in the past were due to the assumptions and approximations incorporated in those models. These simplifications were necessary as finite element methods and CPU were restricted.The accurate representation of the anatomically specific geometry in the finite element model enhances its function as predictive tool for skull fracture and brain injuries. With the aim of the head model we will study and gain insights to the biomechanics of head injury by simulating the real head injuries experienced in the automotive crash environment. The analysis was performed with the with the FE code PAM-CRASH, an explicit, large deformation, Lagrangian dynamic finite element program that is used widely for dynamic crash simulation and for nonlinear structure analysis. The implicit method of Finite Element Analysis is the classic method of solution in simulation, however the calculation speed of implicit method is slow, in this paper, the explicit method which is more effective and suitable the human tissue and large strain impact in traffic accident, through several times attempt and set, was developed to simulate the human head-neck injury response. The new non-linear visco-hyperelastic material model was deduced and utilized in human brain material presentation to simulate the variable strain rate in traffic accident. To analysis the neck sub-failure injury, we build more detailed neck finite element model and study the capsule strain in rear impact, validated the test result and represented the accident scenes. A optimization method was developed to deduce the all the brain test in the last 50 years through reversing engineering, although these tests was in different loading condition, such as compression, simple shear, in different shape, such as cylinder, cube. The biggest concern for finite element model designer is the relationship between strain rate and strain-stress curve, a series of equation was developed based on the assumption that the strain-stress curve was governed by strain rate in the test, one is about viscoelastic parameterβwhich is linear to the strain rate, the other is viscoelastic modulus G which is power function of strain rate. We validate this result in human brain model and predict the brain displacement in craniotomy because of acceleration of gravity, the simulation prediction have well agreement with the realistic MRI image after the craniotomy. We set the material as the No.77 MAT in LS-DYNA solution.更多还原