摘要
利用温室拉伸、扫描电镜,以及基于深度学习和阈值分割法开发的缺陷自动识别和统计程序,研究了缺陷对高压压铸铝合金机械性能的影响规律。结果表明,高压压铸铝合金铸件不同位置的力学性能具有波动性;对比图像识别程序和人工统计断口缺陷面积结果,验证了图像识别程序的精度;断口缺陷面积和力学性能关系表明,孔隙率和最大缺陷尺寸与延伸率具有相关性,当孔隙率或最大缺陷尺寸上升时,高压铸造铝合金的延伸率呈下降趋势。
In this paper,the influence of defects on mechanical properties of high-pressure die-cast aluminum alloy was studied by means of greenhouse stretching,scanning electron microscopy and automatic defect identification and statistics program based on deep learning and threshold segmentation.The results show that the mechanical properties of high-pressure die-cast aluminum alloy castings fluctuate at different positions.The accuracy of the image recognition program was verified by comparing the results of the image recognition program and the manual statistics of the defect area of the fracture.The relationship between the fracture defect area and mechanical properties shows that the porosity and maximum defect size are correlated with the elongation.When the porosity or maximum defect size increases,the elongation of high-pressure cast aluminum alloy shows a downward trend.
作者
仇中原
蒲亮兮
杨雨童
王显会
王秋锋
黄诗尧
Qiu Zhongyuan;Pu Liangxi;Yang Yutong;Wang Xianhui;Wang Qiufeng;Huang Shiyao(Nanjing University of Science and Technology,Nanjing 210009;Yangtze Delta Region Institute of Advanced Materials,Suzhou 215000;Xi’an Jiaotong-Liverpool University,Suzhou 215000;Nanjing Tech University,Nanjing 210009)
出处
《汽车工艺与材料》
2023年第8期1-6,共6页
Automobile Technology & Material
关键词
高压铸造铝合金
图像识别
缺陷特征
延伸率
High-pressure die casting aluminum alloy
Image recognition
Defect feature
Elongation
