【作者】 董立强；

【导师】 李光耀；

【作者基本信息】 湖南大学 ， 机械工程， 2014， 博士

【摘要】 汽车碰撞是造成儿童颈部损伤的重要因素,大约60%-80%的儿童脊椎损伤发生在颈部,而且儿童颈部损伤与成人相比有更高的死亡率。这归因于儿童特殊的生理结构,比如松弛的韧带、平的关节面、未成熟的椎骨及相对较重的头部等,这些导致儿童颈部损伤与成人有很大的不同。但是由于道德及法律等原因,目前对儿童的损伤研究严重不足。为了提高儿童在汽车事故中的安全性,迫切需要对儿童颈部损伤进行研究。本文使用有限元方法分析10岁儿童颈部在碰撞中的损伤,提出获得儿童数据的方法和建立有限元颈部模型的方法,并通过模拟提出儿童颈部损伤特点及机理。本文方法可帮助研究者建立更加精确的儿童有限元模型,研究儿童颈部在汽车碰撞中的损伤特点和机理,从而帮助设计和评估防护措施,提高儿童在汽车碰撞中的安全性。本文的主要内容和创新点包括：(1)提出获得儿童颈椎材料数据的方法。由于缺少儿童实验样本,很难通过实验获得儿童材料数据,而现有儿童颈椎模型缺少获得材料数据的方法,限制了对儿童颈部损伤的研究。本文提出将缩放方法和Pareto及主效应分析相结合的方法,该方法通过数据缩放初步得到儿童的材料数据,之后使用Pareto及主效应分析确定材料参数值。该方法既考虑了儿童与成人几何的差异,又考虑了儿童材料特性较弱的特点。通过与儿童实验数据对比证明了该方法的可靠性,为研究者开发儿童颈椎模型扫除了障碍。(2)提出建立10岁儿童颈椎有限元模型的方法,提高儿童颈椎模型预测损伤的精度。目前儿童颈椎模型多缩放自成人模型,材料模型使用线性材料,建模方法上也存在局限性。本文通过测量24个儿童颈椎尺寸,得到10岁儿童椎骨平均尺寸,通过数据对比证明选定几何数据的正确性。根据几何建立包括儿童特有结构的颈椎有限元模型,实现儿童颈椎模型和解剖结构的一致性。为了提高模拟材料力学特性的准确性,各部件使用非线性材料本构模型,使用本文得到的儿童数据获得各本构模型的参数值,并实现对软组织撕裂的模拟。使用三段颈椎(C0-C2, C4-C5, C6-C7)及整体(CO-T1)的儿童实验数据验证了模型,证明了该建模方法的可靠性。本文预测的软组织失效位置及顺序,可为儿童颈椎实验提供数据支持。(3)提出建立一种新的颈部肌肉有限元模型的方法,提高肌肉模型精度；提出基于径向基函数的代理模型方法对主动收缩肌肉激励曲线进行优化,解决肌肉激励信号寻优问题；提出使用敏感性分析方法研究主动收缩肌肉对头部响应的影响。现有肌肉模型在几何上进行了简化,并且目前的研究仍无法确定各主动收缩肌肉激励曲线的形式,目前也缺少各肌肉在碰撞中对响应影响的了解。本文建立了三种肌肉模型来分析几何对响应的影响,通过与儿童低速正面碰撞实验数据对比,显示几何显著地影响头部响应,本文提出的新的肌肉模型能更好的预测头颈部的响应。使用基于径向基函数的响应面法,对新肌肉模型的肌肉激励曲线进行优化,该方法提高了计算效率,得到了优化的肌肉激励曲线。本文提出评价各肌肉对头颈部响应影响的评价方法,通过计算发现了在正面碰撞中对头颈部响应影响最大的肌肉。(4)针对儿童未发育完全的结构和材料特点,使用椎骨有限元模型进行几何和材料的敏感性分析,获得影响响应的关键因素和儿童颈椎响应及损伤的特点。目前缺少从几何和材料角度研究儿童颈椎响应及损伤的特点。本文使用开发的C4-C6椎段模型进行分析,考虑了所有材料的变化及小关节面和尺寸的几何变化,使用方差分析方法计算各参数对响应的影响程度。根据分析结果提出了儿童颈椎在弯矩和力载荷下响应及损伤的特点,此研究为建立更精确的儿童颈部模型提供了重要的数据支持,并可对儿童颈部损伤诊治提供帮助。(5)针对儿童颈部在碰撞中的损伤问题,使用10岁儿童颈部有限元模型分析儿童颈部在碰撞中的响应,获得儿童颈部的损伤机理,弥补目前对儿童颈部损伤的研究不足。由于儿童尸体样本和模型的缺少,目前仍对儿童颈部在碰撞中的损伤机理缺乏了解,特别是在后面碰撞中。本文使用开发的10岁儿童颈部有限元模型模拟了在正面碰撞和后面碰撞中的儿童头颈部的响应,通过与实验数据的对比证明了模型计算结果的可靠性,通过分析各韧带应变及骨应力分布,获得了儿童颈部损伤的机理。使用腿骨模型模拟轴向冲击下骨折的骨折力和骨折模式,通过与实验数据的对比证明了骨材料模型和参数的可靠性,将此数据应用于儿童椎骨骨折研究中,获得了压缩骨折力与材料参数的关系。更多还原

【Abstract】 Motor vehicle crashes (MVCs) are the main factors that cause the child neck injuries. About60%to80%of all pediatric spinal injuries are in the cervical region. Compared to adults, the mortality rate among victims of pediatric spinal trauma is higher. Children have different anatomical and physiological features compared to adults, such as the laxity of ligaments, decreased angle of facet joints, immature vertebral bodies, insufficiently developed neck musculatures, and relatively large head mass. These make the pediatric neck injuries different from that of adult in MVCs. Due to the ethics and laws, very few studies focused on pediatric injury in MVCs. In order to improve the safety of children in MVCs, more studies on the child neck injuries need to be conducted urgently.This study analyzed the10-year-old (10YO) child neck injuries in MVCs using the finite element (FE) method. The methods of obtaining the child material data and modeling the child neck model were provided. The injury characteristics and mechanisms for child neck were studied by simulations. These methods could help researchers to create more accurate child FE model to study the injury characteristics and mechanisms for child neck in MVCs. Using the FE model, the protective measures could be designed and evaluated to improve the safety of children in MVCs. The detailed contents and innovative points for this study are as follow:(1) A method obtaining the material data for children was provided. Due to lack of pediatric specimens, it was difficult to obtain the child material data through experiments. The existing child cervical spine models didn’t yet provide the method to obtain the child material data. Those limited the studies on child neck injuries. This study provided the method to derive the child material data by combining the scaling method, the Pareto method, and main chart plot through two steps. Step1:get the child data by scaling method based on the adult data; Step2:determine the main material factors using the analysis method of Pareto and main chart plot. Such a two-step method considered not only the geometry difference between child and adult, but also the characteristics of the child neck material. The validity of this method was verified by comparing the simulation results with the experiment data. This method could help researchers to create more accurate child cervical spine models.(2) A method of modeling a10YO child ligamentous cervical spine FE model was presented to improve the accuracy for predicting child cervical spine injuries. The existing child cervical spine models were mostly created by scaling down the adult models and using the elastic material properties. In this study, the average dimensions of cervical vertebral body for10YO children were obtained from24pediatric subjects. A child CT dataset to get the neck geometry was selected by comparing the vertebral body sizes of the selected dataset with the average values. Based on the geometries, the cervical spine FE model was created, including the unique structures for child. To improve the accuracy of predicting the material properties, nonlinear material constitutive models were used. The parameters for the material models were determined using the child data obtained from this study. The failure of soft tissue was also considered in the model by defining the failure parameters. The model was validated at three segments (C0-C2, C4-C5, and C6-C7) as well as the whole cervical spine (C0-T1) against the experimental data. The consistency between the simulation results and experimental data proved the reliability of this modeling method. The model-predicted failure positions and failure progressions of soft tissues provided the data that corroborated child cervical spine tests.(3) An original modeling method for neck muscle was presented to improve the accuracy of neck model. The method of optimizing the muscle activation curves using response surface method (RSM) that’s based on radial basis functions (RBFs) was provided to solve the optimization problem for muscle activation signals. The application of the sensitivity analysis method to study the effects of activation muscles on head responses was presented. The existing muscle models were simplified in geometry, and did not consider the effects of muscle geometries. The existing studies did not determine the types of muscle activation curves, and the effects of activation muscles on neck responses could not obtained by tests. In this study, three neck muscle models were simulated to study the effects of muscle geometry. The comparison results between simulations and experiment data showed that the geometry remarkably affected the head responses. The new muscle model could predict the head and neck responses better. The RSM that’s based on RBF was used to obtain optimized muscle activation curves for the new muscle model. This method improved the computational efficiency to find the optimized solutions. A method to evaluate the effects of active muscles on the head and neck responses was used. The main effect activation muscles were determined by calculating the evaluation values.(4) For children who have not fully developed the structure and material, a material and geometry sensitivity study was conducted using the cervical spine segment model to obtain the main factors affecting the responses and the characteristics of injuries for child cervical spine. Children have different geometry and material properties compared to adults. However, very few researches conducted the injury characteristic study from the perspective of geometry and material. As such, a C4-C6vertebral segment model obtained from the10YO cervical spine model was used to conduct the analysis. The factors included all the materials, the shape of facet joint, and the size of vertebral body. The analysis of variance was used to calculate the contribution of each factor on the responses. Based on the analysis results, the characteristics of responses and injuries for child cervical spine were obtained. The study results could help researchers to create more accurate child cervical spine models, and also provided the needed data for improving child injury analysis or treatment.(5) The analyses of child neck responses in impacts using the10YO child neck FE model were conducted to obtain the injury mechanisms of child neck. These analyses could make up for the inadequacies of the current studies. Due to lack of child cadaver samples and models, there is still a lack of understanding of the child neck injury mechanisms in MVCs, especially in the rear impacts. Using the developed10YO child neck FE model, the head and neck responses in frontal and rear impact were predicted. The consistence between the simulation results and experimental data proved the reliability of the simulation results. The child neck injury mechanisms were obtained by analyzing the ligament strains and bone stress distributions. The leg model was used to simulate the bony fracture force and fracture mode under axial impacts. The reliability of the material constitutive model and material parameters for the bones was verified by comparing the simulation results with the experimental data. Referring to these material data, the child data for vertebra were obtained to study the child vertebra fracture in compression. The relationship between the bony fracture force and material parameters was obtained.更多还原