铝合金弯曲损伤实验与BP神经网络识别  

Aluminum alloy bend damage test and BP neural network identification

作　　者：孙昌立[1,2] 朱荣华[2] 

机构地区：[1]黑龙江科技学院计算机与信息工程学院,哈尔滨150027 [2]哈尔滨工业大学材料科学与工程学院,哈尔滨150001

出　　处：《黑龙江科技学院学报》2012年第6期589-592,共4页Journal of Heilongjiang Institute of Science and Technology

摘　　要：为准确识别铝合金的弯曲损伤,通过铝合金7N01三点弯曲实验,根据所提取声发射信号特征,采用三层BP神经网络对铝合金进行损伤识别。结果表明,铝合金弯曲损伤检测正确率达87.5%。BP神经网络技术与声发射方法能够准确识别7N01的弯曲损伤,为多参数、大数据量智能检测技术提供了参考依据。Aimed at an accurate identification of the bending damage of aluminum alloy,this paper discusses the extraction of acoustic emission signals through three-point bending experiments on aluminum alloy 7N01 and the identification of the bending damage of aluminum alloy according to the signal characteristics and using BP neural network.The results show the correct detection rate of 87.5%.BP neural network technology,combined with acoustic emission method,enables accurate identification of 7N01 bending damage,significantly contributing to intelligent detection technologies designed for large amount of data and multiple parameters.

关 键 词：神经网络 铝合金 声发射 损伤识别 

分 类 号：TG115.28[金属学及工艺—物理冶金] TP183[金属学及工艺—金属学]