股骨头坏死是导致髋关节病残的常见疾患之一，近年来发病率逐渐上升且有年轻化趋势。目前正研究在股骨头坏死发生前期用干细胞修复坏死骨组织的方法。其中一个关键问题是：在新骨长成之前如何对坏死部位进行有效的力学支撑以避免股骨头塌陷。因此本文提出采用牛松质骨作为力学支撑物兼干细胞植入的载体，采用有限元对股骨头/支撑物耦合系统的力学性能进行仿真模拟，以实现最佳的支撑效果。

    然而，骨是一种多孔分级结构，具有非均质、非连续的特点，且个体差异性大。因此其力学模拟极为复杂。为实现上述研究目标，本文在前人基础上对股骨头模拟的分区域弹塑性方法（Subregional Method with Elastoplasticity, SMEP）进行了改进与发展，建立了一种更适合松质骨的力学模拟方法，重点分析解答了以下3方面问题：

    （1）基于三周期极小曲面建立了类骨模型，通过力学仿真模拟诠释了类骨结构材料的屈服宜采用屈服应变而非屈服应力进行定义这一特点；

    （2）从显微结构层面解释了相同骨密度但不同解剖部位的松质骨具有不同弹性模量的原因；

    （3）通过牛松质骨的压缩实验与模拟验证，在SMEP方法中引入了失效软化机制，更加符合松质骨实际特性。

    然后通过股骨头大尺度球头压入实验验证了改进的SMEP方法的有效性，发现利用改进后的SMEP方法使模拟结果更加符合松质骨大变形后垮塌破坏的实际情况。最后将此方法应用于股骨头坏死治疗中，结合手术方案的限制，针对不同坏死类型，设计了不同的力学支撑方案并模拟计算了股骨头/支撑物耦合系统的力学性能。分析得出在本文讨论的股骨头坏死的支撑方案中，对于C型坏死，靠近负重区且相对较细牛骨柱支撑效果最佳，而对于L型坏死，靠近负重区且相对较粗牛骨柱支撑效果最佳，为制定手术方案提供了力学分析结果的参考。 

  Osteonecrosis of femoral head is one of the common diseases that cause hip joint disability. In recent years, the incidence of the disease increases gradually and the patients tend to be younger. Currently, A method using stem cells to treat necrotic bone tissue in the early stage of osteonecrosis of the femoral head is being studied. To avoid collapse of the femoral head, one of the key questions is how to effectively support the necrotic site before the new bone grows. Therefore, this paper proposes that bovine cancellous bone is used as a mechanical support and a carrier for stem cell implantation. To achieve the best support effect, the mechanical properties of the femoral head/support coupling system are simulated by finite element method.

  However, bone is a porous hierarchical structure with heterogeneous features and large individual differences. So, the mechanical simulation of bone is extremely complicated. In order to achieve the above research goals, the Subregional Method with Elastoplasticity (SMEP) is improved and developed, which is more suitable for cancellous bone mechanical simulation. Three main problems are solved as following:

  (1) The bone-like models are established based on the triply periodic minimal surfaces. The mechanical simulation shows that the yielding of the bone-like structural material should be defined by the yield strain instead of the yield stress.

  (2) The reason that cancellous bone with the same bone density but different anatomical sites has different elastic modulus is explained from the microscopic structure level.

  (3) Through the compression experiment and simulation of bovine cancellous bone, the failure softening mechanism is introduced in the SMEP method, which is more consistent with the actual characteristics of cancellous bone.

  Then, the effectiveness of the improved SMEP is verified by the large-scale spherical indentation experiment of the femoral head. It is also found that the simulation results are more in line with the actual situation of collapse and failure of cancellous bone after large deformation by using the improved SMEP.

Finally, this simulation method is applied to the treatment of osteonecrosis of femoral head. Considering the limitations of surgical procedures, different mechanical support schemes are designed for different types of osteonecrosis of femoral head. The mechanical properties of the femoral head/support coupling system are simulated and calculated. It is concluded that among the supporting schemes of osteonecrosis of the femoral head discussed in this paper, for type C osteonecrosis of femoral head, the supporting effect is the best if bovine bone column close to the weight-bearing area and is relatively thin, while for type L osteonecrosis of femoral head, the supporting effect is the best if bovine bone column close to the weight-bearing area and is relatively thick. These results provide a basis for developing surgical plans.