基于神经网络的袋鼠软骨组织材料参数反演及非线性特性分析

Material parameter inversion and nonlinear analysis of kangaroo cartilage based on neural network

软骨损伤和退变会导致骨关节炎等疾病的发生,因此深入研究软骨的力学性能并确定其材料参数是生物力学领域重点关注的课题之一。本文首先建立了袋鼠肩关节软骨组织的有限元模型,将软骨组织视为超弹性材料并采用二阶缩减多项式本构模型,同时在有限元分析中考虑了应变率对力学行为的影响,对软骨组织的压痕过程进行了有限元仿真,得到其变形特征及应力分布规律。其次,搭建了对软骨组织进行材料参数反演的神经网络智能算法,对其本构参数进行了预测,得到了能够准确描述软骨变形行为的材料参数。研究结果表明:神经网络与有限元相结合是预测软骨超弹性材料参数的一种高效、准确的方法,可以有效评估软骨组织力学性能,为关节软骨类病理研究和修复提供参考和帮助。

Cartilage damage and degeneration can lead to osteoarthritis and other diseases.Therefore,in-depth study of the mechanical properties of cartilage and determine its material parameters is one of the most important topics in the field of biomechanics.In this paper,the finite element model of kangaroo shoulder joint cartilage is established firstly.The cartilage tissue is regarded as hyperelastic material and the second-order reduced polynomial constitutive model is adopted.At the same time,the influence of strain rate on mechanical behavior is considered.The finite element simula⁃tion of the indentation process of cartilage tissue is done,and the deformation characteristics and stress distribution law are obtained.Secondly,a neural network intelligent algorithm for the material parameter inversion of cartilage tissue is built,and the constitutive parameters are predicted,and the material parameters that can accurately describe the defor⁃mation behavior of cartilage are obtained.The results show that the combination of neural network and finite element method is an efficient and accurate method to predict the parameters of cartilage hyperelastic materials,which can effec⁃tively evaluate the mechanical properties of cartilage tissue,and provide reference and help for pathological research and repair of articular cartilage.