基于粒子群优化的超声共振谱法测量材料弹性常数  

Particle swarm optimization resonant ultrasound spectroscopy to identify elastic coefficients

作　　者：沈飞 樊璠[1] 王蕊 张强[1] 牛海军[1] SHEN Fei;FAN Fan;WANG Rui;ZHANG Qiang;NIU Haijun(School of Biological Science and Medical Engineering,Beihang University,Beijing 100191,China)

机构地区：[1]北京航空航天大学生物与医学工程学院,北京100191

出　　处：《无损检测》2020年第3期1-4,共4页Nondestructive Testing

摘　　要：超声共振谱法是一种基于材料样品的自由共振频率,再反推其弹性常数的高精度材料力学属性求取方法,该方法在高Q值(品质因素)固体材料的弹性常数计算方面应用广泛。然而对于具有较高黏弹性阻尼的材料来说,受共振谱线平缓、共振频率提取困难等影响,该方法的应用存在一定困难。建立了关于弹性常数的贝叶斯模型,结合粒子群优化(PSO)的全局搜索能力,寻找弹性常数的最大后验概率密度点,最终实现材料的弹性估计。测量了皮质骨模拟材料的弹性常数,与传统方法相比较,基于PSO的超声共振谱方法对弹性常数初始值不敏感,可用于高阻尼材料弹性常数的测量。Resonant ultrasound spectroscopy method is a high-precision method for identifying the full elastic coefficients of materials by using the resonant frequencies.The method is widely used in the calculation of the elasticity of solid materials with high Q value.However,for materials with higher viscoelastic damping,the application of this method meets certain difficulties due to the difficulty in identifying the resonant modes with the overlapped resonant spectrum.This study established a Bayesian formulation of elastic coefficients,with the global search ability of Particle Swarm Optimization(PSO),to find the maximum posterior probability density point of elasticity,and finally realized the identification of elasticity of materials.The elastic coefficients of a simulated cortical bone specimen were measured.Compared with the traditional method,the PSO-based resonant ultrasound spectroscopy is not sensitive to the initial guessing value of the elastic coefficients and can be used for the measurement of the elastic properties of high damping materials.

关 键 词：超声共振谱 弹性常数 粒子群优化 贝叶斯模型 频率配对 

分 类 号：TG115.28[金属学及工艺—物理冶金]