点击次数：

论文类型：期刊论文

发表时间：2024-10-19

发表刊物：焊接学报

卷号：45

期号：09

页面范围：30-41

ISSN号：0253-360X

关键字：friction stir welding; infrared thermal imager; monitoring; support vector regression; temperatures of the weld zone

摘要：At present，2219-T8 high strength aluminum alloy is usually used in the fuel tank of heavy launch vehicle，and the thickness of the tank is up to 18 mm．Friction stir welding (FSW) is used for longitudinal and girth welding of tank shell．The extreme temperatures of the weld zone directly affect the welding quality，but it is difficult to obtain the temperatures of weld zone
during the welding due to the rotation of the tool and the shielding of the shoulder．This paper proposes an in-situ characterization method of extreme temperatures in FSW weld zone．Firstly，based on Deform一3D，the FSW temperature field simulation model of 18 mm thick 2219-T8 aluminum alloy is established，which realizes the temperature field characterization of 18 mm thick 2219-T8 aluminum alloy fiat plates and girth welding．Then，based on the temperature field simulation model，the data sets of the surface feature point temperature and extreme temperatures of the weld zone are extracted．Combined with the support vector regression algorithm，the extreme temperatures prediction model of the weld zone is established．Finally，the surface temperature of the weldment is measured in real time by infrared thermal imager．Combined with the extreme temperatures prediction model of the weld zone，the extreme temperatures of the weld zone in the FSW process are monitored．The experimental results show that the relative error of the peak temperature detection in the FSW weld area is less than 7.22％，and the relative error of the minimum temperature in the weld area is less than 9.54％，and the results confirm the effectiveness of the extreme temperature detection method jn the FSW weld area．