摘要
随着高速铁路的不断提速,高铁轻量化设计中广泛采用高强铝合金材料,但高速列车齿轮箱体服役安全评价亟待完善.本文针对高速列车齿轮箱体使用的铝合金材料服役特性,搭建了声发射检测拉伸试验系统,运用BP神经网络算法对声发射信号进行训练与识别,实现对箱体材料拉伸损伤表征识别与材料服役状态的安全预警.本研究为材料损伤状态的无损实时识别提供了一种识别与预警方法.
With the rapid development of high-speed rails, high-strength aluminum alloys are widely used in the lightweight design, but the service safety assessment of gear boxes in high-speed trains needs to be improved in China. An acoustic emission tensile test system was built for high-speed train gearbox shells made of aluminum alloys. After training and recognition by a BP neural network, acoustic emission signal was used for characterizing tensile damage in the materials and warning the materials service status. The research provides a method of nondestructive real-time characterization and warning for damage in aluminum alloys.
出处
《北京科技大学学报》
EI
CAS
CSCD
北大核心
2013年第5期626-633,共8页
Journal of University of Science and Technology Beijing
基金
"十一五"国家科技支撑计划资助项目(2009BAG12A07-D07)
国家自然科学基金资助项目(61273205
51005014)
教育部中央高校基本科研业务专项(FRF-SD-12-028A)
关键词
铝合金
声发射
损伤探测
神经网络
模式识别
aluminum alloys
acoustic emissions
damage detection
neural networks
pattern recognition
