基于声发射信号短时能量的TIG焊熔透状态识别  被引量:4

Weld Bead Penetration State Recognition in TIG Welding Process Based on Short-time Energy of Acoustic Emission Signal

在线阅读下载全文

作  者:王齐胜 高延峰[1] 龚岩峰 黄林然 赵佳敏 WANG Qisheng;GAO Yanfeng;GONG Yanfeng;HUANG Linran;ZHAO Jiamin(School of Aeronautic Manufacturing Engineering,Nanchang Hangkong University,Nanchang 330063,China)

机构地区:[1]南昌航空大学航空制造工程学院,江西南昌330063

出  处:《热加工工艺》2020年第21期107-110,116,共5页Hot Working Technology

基  金:国家自然科学基金资助项目(51465043);江西省自然科学基金资助项目(20171BAB206033);江西省重点研发计划项目(20171BBE50011)。

摘  要:针对TIG焊接质量在线检测问题,提出了基于声发射信号短时能量的焊缝熔透状态识别方法。采集了TIG焊接过程中母材负载的声发射信号并对其进行了时频域分析。首先采用限幅滤波法降低信号中的脉冲干扰。为了提高信噪比并凸显信号的能量突变信息,对声发射信号进行了短时能量分析。结果表明,未熔透、临界熔透、过熔透三种状态的信号短时能量依次变大。最后,根据平均短时能量,利用支持向量机进行熔透状态的识别。试验结果表明,声发射信号的平均短时能量可以很好地将三种熔透状态进行区分,对熔透状态的识别率达到了98%以上,从而为焊接质量在线检测提供了一种新方法。Aiming at quality detection of TIG welding on line, a weld bead penetration state recognition method based on short-time energy of acoustic emission signal was proposed. The loading acoustic emission signal of the base metal in the process of TIG welding was collected and analyzed by time-frequency domain. Firstly, the amplitude limit filtering method was used to reduce the pulse interference in the signal. In order to improve the signal-to-noise ratio and highlight the energy mutation information of the signal, a short-time energy analysis of acoustic emission signal was performed. The results show that the short-time energy of the three states which are non-penetration, critical penetration, and excessive penetration becomes larger in turn. Finally, the support vector machine was used to realize the penetration state recognition based on the average short-time energy. The test results show that the average short-time energy of the acoustic emission signal can distinguish the three penetration states well, and the average recognition rate of the penetration state reaches over 98%. This research provides a new method for the welding quality online detection.

关 键 词:声发射信号 短时能量 熔透状态 状态识别 

分 类 号:TG444[金属学及工艺—焊接]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象