摘要
降噪是超声信号处理的重要环节,正交匹配追踪是一种常用的降噪方法,传统正交匹配追踪算法计算量大、分解精度不高,无法提取强噪声背景下的超声信号.本文提出了一种结合果蝇优化算法和正交匹配追踪的超声信号降噪算法,将正交匹配追踪中的“贪婪”搜索转换为Gabor函数的参数优化问题,利用果蝇优化算法估计Gabor函数的最优值,采用自适应步长以提高果蝇优化算法的全局遍历性,同时引入高维广义CAT映射以跳出局部最优,最后由寻找到的最佳原子重构超声信号.为验证算法的有效性,对仿真的多频超声回波信号和实验采集的锻件试块超声回波信号进行了降噪处理,结果表明,本文提出的方法能有效提取强噪声背景下的超声信号.
Noise reduction is the key technology of ultrasonic testing whose performance affects the evaluation of ultrasonic testing results directly. Orthogonal matching pursuits(OMP) is one of the most popular methods for the purpose of noise reduction;however, OMP suffers from a number of disadvantages of high complexity and time-consuming of the atom searching, and it is unable to extract the ultrasonic signal under background of powerful noise. An improved fruit fly optimization algorithm based orthogonal matching pursuit(IFOA-OMP) for ultrasonic noise reduction method is proposed, the parameters of Gabor function are estimated by IFOA, and the global ergodicity is improved by the adaptive step function, at the same time, the high-dimensional extensive CAT map is introduced to jump out of the local optimum. Finally, the ultrasonic signal is reconstructed from the best matching atoms. In order to validate the effectiveness of the proposed method,the simulated multi frequency ultrasonic signal and the real ultrasonic signal of forging block are used. The processing results show the proposed method can accurately extract the ultrasonic signal under background of powerful noise.
作者
董明
李敬
索永录
唐恩贤
马宏伟
陈渊
张广明
万翔
DONG Ming;LI Jing;SUO Yong-lu;TANG En-xian;MA Hong-wei;CHEN Yuan;ZHANG Guang-ming;WAN Xiang(School of Mechanical Engineering,Xi’an University of Science and Technology,Xi’an,Shaanxi 710054,China;Technology Center,Shaanxi Huangling Coal Mine Co.,Ltd.,Yan’an,Shaanxi 727307,China;School of Energy Engineering,Xi’an University of Science and Technology,Xi’an,Shaanxi 710054,China;Shaanxi Key Laboratory of Mine Electromechanical Equipment Intelligent Monitoring,Xi’an,Shaanxi 710054,China;General Engineering Research Institute,Liverpool John Moores University,Liverpool L33AF,United Kingdom)
出处
《电子学报》
EI
CAS
CSCD
北大核心
2022年第2期493-501,共9页
Acta Electronica Sinica
基金
国家自然科学基金(No.51705418,No.52175518)
中国博士后科学基金(No.2019M653696)
陕西省自然科学基础研究计划陕煤联合基金(No.2021JLM-07)
陕西省自然科学基础研究计划(No.2019JQ-801,No.2019JM-024,No.2019JM-212)。
关键词
正交匹配追踪
果蝇优化算法
自适应步长
高维广义CAT映射
数字信号处理
orthogonal matching pursuit(OMP)
fruit fly optimization algorithm(FOA)
adaptive step
high-dimensional extensive CAT map
digital signal process
