本文运用模式识别中的统计决策方法对焊接缺陷超声回波信号进行分类识別,以达到区分平面型焊接缺陷(裂纹)和体积型焊接缺陷(夹渣、气孔)的超声回波信号之目的。试验研究表明,本文采用的方法具有一定的可靠性和可行性。经过对32个实际焊接缺陷超声回波信号的分类识别,平面型缺陷的正确分类率为86%,体积型缺陷的正确分类率为78%;总正确分类率为81%。 In this paper, the statistical decision-making method in pattern recognition is used to classify the ultrasonic echo signals of welding defects so as to achieve the purpose of distinguishing ultrasonic echo signals between planar welding defects (cracks) and volumetric welding defects (slag inclusions and stomata). Experimental results show that the method used in this paper has certain reliability and feasibility. After the classification and identification of 32 ultrasonic echo signals of actual welding defects, the correct classification rate of planar defects is 86%, the correct classification rate of volume defects is 78% and the total correct classification rate is 81%.