摘要
针对7系超硬铝在传统熔焊过程中易出现热裂纹、气孔和焊接接头软化等问题,研究振动焊接工艺过程中,焊接工艺参数与焊接接头强度系数间的非线性关系机理,建立7075超硬铝振动焊接接头强度系数的双重人工神经网络评估模型,包括以焊接参数作为输入接头的焊缝参数、抗拉强度、伸长率和硬度预测模型,以及以4个子模型的预测输出为输入接头的焊接强度系数预测模型。根据建立的预测模型进行焊接接头强度试验,预测结果可以满足工程需要,具有工程实用价值。
To solve the problems of hot cracks, pores and softening of welded joints in the traditional welding process of 7 series superhard aluminum, the nonlinear relationship mechanism between the welding process parameters and the strength coefficient of welded joints during the vibration welding process was studied. A dual artificial neural network evaluation model was established for the strength coefficient of 7075 superhard aluminum vibration welded joints. The first ANNs model includes four submodels with welding parameters as input： joint weld seam parameter prediction model, joint tensile strength prediction model,joint elongation prediction model and joint hardness prediction model. The second ANNs model is joint welding intensity coefficient prediction model with the predicted output of four submodels in the first ANNs as input. A series of welded joint strength tests based on the established prediction model show that the prediction results of the model can meet the engineering needs and have engineering practical value..
作者
刘政军
张琨
刘长军
LIU Zhengjun;ZHANG Kun;LIU Changjun(School of Material Science and Engineering,Shenyang University of Technology,Shenyang 110870,China;School of Chemical Equipment,Shenyang University of Technology,Liaoyang 111000,China)
出处
《兵器材料科学与工程》
CAS
CSCD
北大核心
2018年第5期53-56,共4页
Ordnance Material Science and Engineering
基金
辽宁省科技资助项目(20131079)
关键词
双重人工神经网络
超硬铝合金
脉冲熔化极氩弧焊
机械振动
接头强度系数
dual artificial neural networks
superhard aluminum alloy
double-pulsed metal inert-gas welding
mechanical vibration
joint strength coefficient
