摘要
针对常规方法难以检测镁合金涂层缺陷的问题,提出采用非线性超声检测技术对含有不同程度微缺陷的镁合金聚乳酸涂层试件进行检测。同时,由于镁合金涂层试件较小,可容纳的正弦脉冲周期数较少,导致超声非线性系数β的计算易受到噪声和频率偏移的干扰而产生较大误差,提出采用截断-最小二乘法(truncation least square method,简称TC-LS)等4种信号处理方法对涂层试件检测信号进行处理,并进行仿真试验研究。结果表明:TC-LS计算β的误差在5%以内,在抗噪能力和处理少周期数信号能力上比其他3种方法更强、精度更高。对实测信号研究分析表明,TC-LS比离散傅里叶变换(discrete Fourier transform,简称DFT)得到的β值趋势变化更加明显,能够更有效计算β值,为非线性超声信号处理提供了一种新方法。
In view of the difficulty encountered by conventional methods in detecting defects in magnesium alloy coatings,the nonlinear ultrasonic detection technology is proposed to detect the samples of polylactic acid coating of magnesium alloy with different degree of microdefects.Meanwhile,due to the small size of the magnesium alloy coating specimen and the small number of sinusoidal pulse cycles that can be accommodated,the calculation of the ultrasonic nonlinear coefficientβis highly susceptible to noise interference and frequency shift,resulting in significant errors.To process the detection signals of the coating specimen,4 signal processing methods,including truncation least square method are proposed,and simulation experiments are conducted.The results show that the error in calculatingβusing truncation least square method is within 5%,and the method demonstrates higher noise robustness and greater accuracy in processing signals with fewer cycles compared with the other 3 methods.Through the analysis of the measured signals,the results show that the trend change ofβis more obvious than that of the discrete Fourier transform,and theβcan be calculated more effectively.
作者
颜丙生
黄致远
张柏
YAN Bingsheng;HUANG Zhiyuan;ZHANG Bai(School of Mechanical and Electrical Engineering,Henan University of Technology Zhengzhou,450001,China;Xuchang Tobacco Machinery Co.,Ltd.Xuchang,461143,China)
出处
《振动.测试与诊断》
北大核心
2025年第6期1098-1104,1270,共8页
Journal of Vibration,Measurement & Diagnosis
基金
国家自然科学基金联合基金资助项目(U1604134)
河南省科技攻关资助项目(212102210327)
河南工业大学创新基金支持计划专项资助项目(2020ZKCJ28)。
关键词
镁合金
非线性超声检测
聚乳酸涂层
最小二乘法
信号处理
magnesium alloy
nonlinear ultrasonic detection
polylactic acid coating
least square method
signal processing
