基于导波频散特征的超声导波模态识别方法

发布时间：2018-04-02 15:19

本文选题：超声导波　切入点：频散　出处：《机械工程学报》2017年18期

【摘要】：针对超声导波检测应用中模态识别难的问题,提出一种基于导波频散特征的模态识别方法。通过估计导波信号的频散量来识别导波模态,而导波的频散量与该模态导波波数在激励频率处的二次泰勒展开系数有关。根据导波信号的频散特点,构造带时间斜变的Chirplet匹配原子库,基于该Chirplet原子库,对导波信号进行匹配追踪分解,并算得该导波信号波数在激励频率处二次泰勒展开系数的最优估计值,再根据预先算得的各模态导波的波数-频率频散关系,区分导波信号的不同模态。数值模拟和试验验证都表明,该导波模态识别方法是准确且有效的。而该方法的不足之处是尚不能识别混叠严重的导波信号。研究结果有助于提高人们对复杂导波检测信号的解析能力,并推动导波检测技术的推广应用。
[Abstract]:Aiming at the difficulty of modal identification in ultrasonic guided wave detection, a modal identification method based on dispersion characteristics of guided wave is proposed.The pilot mode is identified by estimating the dispersion of the pilot wave signal, and the dispersion of the pilot wave is related to the quadratic Taylor expansion coefficient of the guided wave number at the excitation frequency.According to the dispersion characteristics of the guided wave signal, a Chirplet matching atom library with time-skew variation is constructed. Based on the Chirplet atom library, the guided wave signal is decomposed by matching tracing.The optimal estimate of the quadratic Taylor expansion coefficient of the guided wave signal at the excitation frequency is obtained, and the different modes of the guided wave signal are distinguished according to the wavenumber frequency dispersion relation of the pilot wave calculated in advance.Numerical simulation and experimental results show that the method is accurate and effective.The disadvantage of this method is that it can not recognize the serious mixed guided wave signal.The results are helpful to improve the analytical ability of complex guided wave detection signals and promote the popularization and application of guided wave detection technology.
【作者单位】：武汉理工大学能源与动力工程学院;海军工程大学动力工程学院;
【基金】：国家自然科学基金(51205404,51709216) 国防预研基金(9140A27020115JB11001) 湖北省自然科学基金(2016CKB700)资助项目
【分类号】：TB559;TN911.7

【相似文献】