摘要
激光喷丸强化是一项新型的表面处理技术,在这个处理过程中会产生残余应力,从而有效抑制材料疲劳裂纹的萌生以及减缓裂纹扩散速率,有效提高材料的疲劳寿命。为有效地控制金属表面残余应力,结合激光喷丸技术的特点,利用神经网络强大的非线性映射能力,将金属材料主要的力学性能参数和激光参数作为网络输入,金属材料表面残余应力作为网络输出,建立金属材料表面残余应力的优化控制模型。最后选用7050Al、A304不锈钢和AM50镁铝合金这三种金属材料对此模型进行验证,验证结果表明此模型可以有效地控制金属材料表面的残余应力。
Laser shot peening is a new technique for strengthening metals. This process induces a compressive residual stress field, which increase fatigue crack initiation life and reduces fatigue crack growth rate. In order to effectively control metal surface residual stress, consider the characteristics of laser shot technology, then use neural networks a powerful nonlinear mapping ability to establish a surface residual stress of metallic materials of the optimal control model. In this model, taking the metal of the mechanical properties of material parameters and laser parameters as the network input, the residual metal surface as the network output. Finally, selecting the 7050Al, A304 stainless steel and AM50 magnesium alloy to verify this model, and the validation results show that this model can effectively control metal surface residual stress.
出处
《应用激光》
CSCD
北大核心
2010年第1期15-19,共5页
Applied Laser
基金
国家自然科学基金项目(项目编号:50705038)
江苏省自然基金青年科技创新人才学术带头人项目(项目编号:BK2007512)
关键词
激光喷丸
表面残余应力
神经网络
BP算法
laser shock
the surface residual stress
neural network
BP algorithm