摘要
以锆4合金电子束焊接熔深为研究对象,用BP神经网络的方法建立电子束焊接工艺参数与熔深之间的数学模型;本研究采用MATLAB语言实现BP算法,其程序设计主要是调用MATLAB神经网络工具箱的相关函数来完成。结果表明,所建BP神经网络模型具有较好的自适应和预测能力。
This paper presents the application of software package MATLAB to establish model of the welding bead depth of electron beam welding Zirconium alloy based on BP neural networks.The model was trained with actual samples obtained from high power electron beam welding Zirconium alloy.The simulation results indicated that the model has strong abilities on nonlinear adaptivity and predicting.The model not only possesses the ability to estimate the welding bead depth by the parameters,but also can optimize the welding technics according to predicted results.
出处
《电焊机》
北大核心
2013年第12期64-67,共4页
Electric Welding Machine
关键词
锆4合金
电子束焊接
BP神经网络
zirconium alloy 4
electron beam welding
BP neural networks
zirconium alloy
