摘要
为分析网格不敏感结构应力的焊接疲劳S-N数据分布具有高收敛性的机理,基于钛合金焊接接头疲劳试验数据,分别基于名义应力、主S-N曲线法中的结构应力及等效结构应力建立焊接疲劳数据坐标,并以最小二乘法拟合S-N曲线,应用粗糙集分析方法,以接头类型、相对板厚、载荷比、应力比等焊接疲劳主要影响因素作为条件属性,以疲劳数据对于S-N曲线的偏离度作为决策属性,对基于三种应力的焊接疲劳数据分布规律对比分析.结果表明,基于网格不敏感等效结构应力所建立的S-N数据分布更为集中和均匀,在预测焊接接头疲劳寿命方面具有更高的准确性.
To study the high convergence of welding fatigue data distribution based on the mesh-insensitive structural stress,welding fatigue data coordinates were established based on nominal stress, structural stress and equivalent structural stress in principal S- N curve method from fatigue data of titanium alloy welded joints. S- N curves were fitted by least square method. The influences of welding factors on fatigue life were analyzed with rough set method by taking joints type,thickness,load ratio,ratio set as main influencing factors and taking the deviation of S- N data to S- N curve as decision attribute. Results indicated that the S- N data distribution based on mesh-insensitive structural stress are more concentrated and uniform. So it can be concluded that this method has a higher accuracy in predicting fatigue life of welding joint.
出处
《焊接学报》
EI
CAS
CSCD
北大核心
2015年第2期11-14,113,共5页
Transactions of The China Welding Institution
基金
国家自然科学基金资助项目(5117504)
辽宁省科学技术计划资助项目(2011220039)
教育部科学技术研究重点资助项目(210045)
计算智能重庆市重点实验室资助项目(CQ-LCI-2013-05)
关键词
焊接
网格不敏感结构应力
疲劳
数据分布
welding
mesh-insensitive structural stress
fatigue
data distributions
