>> 自然科学版 >> 当期目录 >> 正文
黏弹性软板拉梅常数的Lamb波检测
时间: 2026-11-20 次数:

禹建功,陈宇浩,张博,等.黏弹性软板拉梅常数的Lamb波检测[J].河南理工大学学报(自然科学版),2027,46(1):98-105.

Yu J G, Chen Y H, Zhang B,et al.Lamb wave detection of Lamé constants of viscoelastic soft plates[J].Journal of Henan Polytechnic University(Natural Science) ,2027,46(1):98-105.

黏弹性软板拉梅常数的Lamb波检测

禹建功, 陈宇浩, 张博, 刘欢

河南理工大学 机械与动力工程学院,河南 焦作 454003

摘要: 目的  生物组织发生病变会导致其结构材料特性(如刚度、硬度等)发生变化,利用这些变化可从力学角度表征生物组织的健康状态。为此,开发一种基于神经网络和Lamb波技术的智能检测方法,检测黏弹性软板的材料常数,为皮肤、角膜等板状组织的疾病早期诊断和辅助治疗提供一种有效的方法。 方法 首先,基于Kelvin-Voigt黏弹性模型,建立黏弹性软板中Lamb波的动力学方程,使用勒让德正交多项式法求解该波动方程,得到黏弹性软板中Lamb波的群速度频散曲线;然后,使用海藻酸钠制作块状和板状实验样本,对样本激励和接收体波以及不同频段下的Lamb波,通过信号分析得到相应的波速;最后,建立材料常数和Lamb波群速度之间的非线性映射关系,构建一种基于Lamb波和人工神经网络的反演模型,使用理论数据训练反演模型,将检测得到的群速度带入训练好的反演模型,反演得到黏弹性软板中的材料常数。 结果 结果表明,除了前两阶模态A0和S0,其余模态都存在截止频率。理论计算与实验结果吻合良好,误差小于2%。拉梅常数的反演结果和体波检测结果一致,误差最高为2.38%。 结论  所提出的基于神经网络和Lamb波技术的检测方法能够有效检测黏弹性软板的材料特性,具有较高的准确度和可靠性。

关键词:软板;Lamb波;黏弹性;频散;反演

doi:10.16186/j.cnki.1673-9787.2024020032

基金项目:国家自然科学基金资助项目(12102131);河南省教育厅创新团队项目(23IRTSTHN016);河南理工大学创新团队项目(T2022-4);河南省科技攻关项目(242102221041)

收稿日期:2024/02/29

修回日期:2025/02/21

出版日期:2026/11/20

Lamb wave detection of Lamé constants of viscoelastic soft plates

Yu Jiangong, Chen Yuhao, Zhang Bo, Liu Huan

School of Mechanical and Power Engineering, Henan Polytechnic University, Jiaozuo 454003, Henan, China

Abstract: Objectives  Pathological changes in biological tissues can lead to alterations in their structural material properties, such as stiffness and hardness. These changes can be utilized to characterize the health status of biological tissues from a mechanical perspective. To this end, an intelligent detection method based on neural networks and Lamb wave technology is developed to detect the material constants of viscoelastic soft plates, providing an effective approach for early diagnosis and adjunctive therapy of diseases in plate-like tissues such as skin and cornea.Methods  First, based on the Kelvin-Voigt viscoelastic model, the dynamic equation of Lamb waves in viscoelastic soft plates is established. The Legendre orthogonal polynomial method is employed to solve the wave equation, yielding the group velocity dispersion curves of Lamb waves in the viscoelastic soft plates. Second, block-shaped and plate-shaped experimental samples are fabricated using sodium alginate. Bulk waves and Lamb waves at different frequency bands are excited and received from the samples, and the corresponding wave velocities are obtained through signal analysis. Finally, a nonlinear mapping relationship between the material constants and Lamb wave group velocities is established. An inversion model based on Lamb waves and an artificial neural network is constructed. The inversion model is trained using theoretical data, and the measured group velocities are fed into the trained model to invert the material constants of the viscoelastic soft plates. Results  The results show that, except for the first two modes A0 and S0, all other modes exhibit cutoff frequencies. The theoretical calculations are in good agreement with the experimental results, with errors below 2%. The inversion results of the Lamé constants are consistent with the bulk wave measurement results, with a maximum error of 2.38%. Conclusions The proposed detection method based on neural networks and Lamb wave technology can effectively detect the material properties of viscoelastic soft plates with high accuracy and reliability.

Key words:soft plate;Lamb wave;viscoelasticity;dispersion;inversion

最近更新