金属材料超声探伤缺陷智能识别方法的研究

According to flaws that are difficult to accurately identification during ultrasonic testing, signal analysis and intelligence classification methods are mainly studied in this paper, due to the transient virtue of ultrasonic pulse echo, wavelet transform is used in this paper to extract the feature information of flaw signals and to suppress the noise in the signals. To realize flaws classified intelligently and automatically, artificial neural network is used to distinguish different flaws automatically. Based on the achievement, wavelet packet and fit dynamic envelope method have given to extract the feature of the flaw in this paper. In order to apply practice, we build a virtual instrument to aid identifying flaw type. The research result could assist to analyze ultrasonic flaw detection, thus perfect existing ultrasonic flaw detector tool and offer strongly means for flaw determined the nature and identified, improve the efficiency of ultrasonic inspection enormously.

目前超声探伤的主要趋势是智能化和自动化。本项研究的主要目标是研究超声探伤中缺陷的定性分类的智能化方法。采用小波分析及模式识别方法，从常规超声检测（A型脉冲反射式┑娜毕莼夭ㄐ藕胖刑崛》从橙毕菪灾实奶卣髦担ν纪牙肴宋鞴垡蛩兀凸劭蒲У囟匀毕萁惺侗鸸槔唷Ｎ迪种悄芑瞧鞯氖涤没⒉坊於ɑ。锏阶远⒏咝У厥侗鸶骼嗳毕莺吞岣呒觳饪煽啃约疤缴诵实哪康摹?.