基于奇异值分解的激光超声晶粒尺寸检测研究

Research on grain size detection based on singular value decomposition using a laser ultrasonic technique

激光超声检测金属材料晶粒尺寸的信号中包含了大量噪声,而无法实现平均晶粒尺寸相近时的快速检测。本文以冷轧低碳钢为研究对象,采用了奇异值分解降噪方法对激光超声检测晶粒尺寸的信号进行处理,实现了平均晶粒尺寸偏差在4μm范围内的精确识别,提高了激光超声检测金属材料平均晶粒尺寸的检测精度和分辨率。

The signal of the laser ultrasonic detection of the grain size of metallic materials contains a large amount of noise,while rapid detection cannot be achieved when the average grain size is similar.In this paper,cold rolled low carbon steel is used as the research object,and the singular value decomposition noise reduction method is used to process the signal of the laser ultrasonic detection of the grain size,which realizes the accurate identification of the average grain size deviation within 4μm,and improves the detection accuracy and resolution of the average grain size of metal materials using the laser ultrasonic.