高级检索
任大鑫, 丛凌翔, 韩荣豪, 宋刚, 刘黎明. 根部加强的镁合金搅拌摩擦焊分析[J]. 焊接学报, 2024, 45(1): 23-30. DOI: 10.12073/j.hjxb.20230104001
引用本文: 任大鑫, 丛凌翔, 韩荣豪, 宋刚, 刘黎明. 根部加强的镁合金搅拌摩擦焊分析[J]. 焊接学报, 2024, 45(1): 23-30. DOI: 10.12073/j.hjxb.20230104001
REN Daxin, CONG Lingxiang, HAN Ronghao, SONG Gang, LIU Liming. Study on friction stir welding of magnesium alloy with backing plate[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2024, 45(1): 23-30. DOI: 10.12073/j.hjxb.20230104001
Citation: REN Daxin, CONG Lingxiang, HAN Ronghao, SONG Gang, LIU Liming. Study on friction stir welding of magnesium alloy with backing plate[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2024, 45(1): 23-30. DOI: 10.12073/j.hjxb.20230104001

根部加强的镁合金搅拌摩擦焊分析

Study on friction stir welding of magnesium alloy with backing plate

  • 摘要: 常规搅拌摩擦焊中,不同厚度的材料所适配的最佳搅拌针长度也不相同,搅拌针过长或过短都会对焊接效果产生不利影响. 为了解决这一局限性,提出了一种在焊缝背面添加适当厚度同种材料垫板的新型焊接工艺,在该工艺中,搅拌针长度大于被焊板材厚度,将垫板与母材焊接在一起,一方面,降低了对搅拌针长度的要求;另一方面,可消除焊缝减薄产生的不利影响. 结果表明,采用该方法分析1.5 mm厚AZ31B镁合金的对接焊,接头抗拉强度最大可达母材的91.19%,此外,分析了焊缝横截面微观组织和显微硬度分布,通过所建立的卷积神经网络模型,对接头抗拉强度随参数变化的分布情况进行了预测,获取了最佳工艺参数.

     

    Abstract: In conventional friction stir welding, the length of the stirring pin needs to be strictly matched with the thickness of the welded plate. A too-long or too-short stirring pin will adversely affect the welding effect. To solve this limitation, this paper proposes a new welding process that adds a backing plate of the same material with appropriate thickness to the back of the weld seam. In this process, the length of the stirring pin is greater than the thickness of the welded plate, and the backing plate is fused with the base metal. On the one hand, the requirement for stirring pin length is reduced. On the other hand, the adverse effects of weld seam thinning can be eliminated. The butt welding of 1.5 mm thick AZ31B magnesium alloy was studied by this method. The maximum tensile strength of the weld joint can reach 91.19% of the base metal. In addition, the microstructure and microhardness distribution of the weld beam cross-section were analyzed. The optimal process parameters were obtained by the established convolutional neural network model, and the distribution of joint tensile strength with the change of parameters was predicted.

     

/

返回文章
返回