摘要
基于神经网络理论,结合真空扩散焊接过程中的持续时间、温度和压力等焊接参数的影响,建立了可用于铝基复合材料焊接接头力学性能分析的数学模型。该数学模型的预测结果与实测数据间的最大相对误差、平均相对误差和均方误差指标均满足预测要求,且模型有较高的精度和较强的容错能力。利用所建立的真空扩散焊接模型进行仿真,可以分析不同的焊接参数对于焊接接头力学性能的影响,为铝基复合材料的力学性能分析提供了有价值的参考。
Based on the neural network theory,considering the effect of duration,temperature,and pressure welding parameters in the vacuum diffusion welding process,the mathematical model can be built for analyzing the aluminum welding mechanical properties of composites joint.The Predicted result of the mathematical model is consistent with experimental data at the maximum relative error,the average relative error and mean square error.And the model has high accuracy and strong fault tolerance.By using the simulation results of the model,you can analyze the impact of different welding parameters for the mechanical properties of welded joints,which provides a valuable reference for the analysis of the mechanical properties of aluminum matrix composites.
出处
《电焊机》
北大核心
2014年第11期103-105,115,共4页
Electric Welding Machine
基金
江西省教育厅资助项目(GJJ-13503)
关键词
神经网络
金属复合材料
力学性能分析
neural networks
metallic composite materials
mechanical analysis
