Predicting residual stresses in multi-pass weldments with the finite element method☆☆☆
Predicting residual stresses in multi-pass weldments with the finite element method☆☆☆
复制标题
有限元法预测多道焊件残余应力☆☆☆
DOI：
10.1016/0045-7949(89)90048-5
复制
发表时间：
1989
影响因子：
4.7
通讯作者：
R. Goff
中科院分区：
文献类型：
--
作者：
J. Free;R. Goff
焊接结构中的脆性断裂和疲劳裂纹扩展很大程度上受到焊接过程产生的残余应力和缺陷的影响。本研究涉及使用有限元技术预测复杂多道焊件的残余应力。该理论模型是通过对市售热弹塑性有限元包进行广泛修改而开发的，与独立实验工作具有良好的一致性。该计算模型能够预测制造条件下多道焊件厚度范围内的三维焊接残余应力。为了减少计算时间，保留了简化的分析方法。这就对过于复杂的分析的必要性提出了质疑，该分析需要对可变热性能、相变、随时间变化的塑性、超过熔点的温度以及随温度变化的机械性能（屈服强度除外）进行建模。很少有研究尝试确定必须包含的参数才能保持准确的结果。此外，关于多道焊件厚度方向残余应力分布预测的公开数据很少。本文中给出的实验样本的机械和焊接细节都清晰地表明，试图提供可靠的数据进行比较。
Brittle fracture and fatigue crack propagation in welded structures are largely affected by residual stress and defects created by the welding process. This study is concerned with the prediction of residual stresses in complex multi-pass weldments with the use of the finite element technique. The theoretical model, developed by extensively modifying a commercially available thermal-elastic-plastic finite elements package, provides good agreement with independent experimental work. The computational model is capable of predicting three-dimensional welding residual stresses through the thickness of a multi-pass weldment in the as-fabricated condition.A simplified approach to the analysis was maintained in order to reduce computation time. This questions the necessity for an overly complex analysis which models variable thermal properties, phase changes, time-dependent plasticity, temperatures beyond melting point and temperature-dependent mechanical properties (except yield strength). Very few studies attempt to identify the parameters which must be included in order to maintain an accurate result. In addition, there is little published data concerning the prediction of the through-thickness residual stress distribution in multi-pass weldments. The mechanical and welding details of the experimental specimens presented in this paper are clearly indicated in an attempt to provide reliable data for comparison.