摘要
切削过程中严重的刀具磨损会直接影响工件的加工质量。针对该问题,构建了基于CNN-ABiGRU的刀具磨损状态监测模型,引入注意力机制合理分配对不同时段信号特征的注意力。实验结果表明:文中方法能精确识别刀具的磨损状态,较CNN和ABiGRU有更好的识别精度与稳定性。
In the milling process,the wear of the tool will affect the quality of the workpiece directly.For this problem,a tool wear monitoring model based on CNN-ABiGRU was built to monitor the tool wear condition,and attention mechanism was introduced to allocate attention to signal characteristics at different time step.Experimental results show that this method can identify the tool wear condition accurately,and has better identification accuracy and stability than CNN and ABiGRU.
作者
吴琪文
周学良
吴瑶
Wu Qiwen;Zhou Xueliang;Wu Yao(School of Mechanical Engineering,Hubei University of Automotive Technology,Shiyan 442002,China)
出处
《湖北汽车工业学院学报》
2021年第4期59-64,69,共7页
Journal of Hubei University Of Automotive Technology
基金
国家自然科学基金(52075107)
湖北汽车工业学院博士科研启动基金(BK201601)
中国博士后科学基金(2018M6409120)
湖北省高等学校优秀中青年科技创新团队计划项目(T2020018)。
