Page 5 - 应用声学2019年第2期
P. 5

第 38 卷 第 2 期                                                                       Vol. 38, No. 2
             2019 年 3 月                          Journal of Applied Acoustics                    March, 2019


             ⋄ 研究报告 ⋄



               HDP-HSMM的磨削声发射砂轮钝化状态识别                                                                    ∗




                      钟利民     1,2   李丽娟     1,2   杨 京     1,2†  梁 彬     1,2  程建春      1,2  刘翔雄     3



                                               (1 南京大学声学研究所      南京   210093)
                                         (2 人工微结构科学与技术协同创新中心           南京   210093)
                                          (3 华辰精密装备 (昆山) 股份有限公司       昆山   215337)
                摘要    在高精度金属材料磨削加工中,刀具即砂轮的状态对加工效率和加工质量具有重要的影响。钝化程度
                较高的砂轮不适于加工精密工件,需提前预警并修整更换砂轮。该文提出一种通过磨削声发射信号来检测砂
                轮钝化状态的方法。首先,对于采集到的信号进行小波软阈值降噪。然后,将其分割成多个有重叠的帧,并提
                取每帧信号的 8 个特征组成声发射数据集。最后,通过分层 Dirichlet 过程 -隐半马尔可夫模型来建立声发射数
                据集和不同的砂轮钝化状态之间的非线性关系,旨在识别砂轮钝化状态。结果表明,上述检测方法能有效识别
                砂轮的不同钝化状态并能对整个加工过程中的砂轮钝化程度进行自动划分,其在测试数据集上的准确率达到
                93.7%,可以为实际工业应用提供理论指导。
                关键词     砂轮钝化,分层 Dirichlet 过程 -隐半马尔可夫模型,磨削声发射,小波阈值降噪
                中图法分类号: O429           文献标识码: A          文章编号: 1000-310X(2019)02-0151-08
                DOI: 10.11684/j.issn.1000-310X.2019.02.001




               The blunt state identification of acoustic emission for grinding wheel based on

                                                     HDP-HSMM

                              ZHONG Limin   1,2  LI Lijuan 1,2  YANG Jing 1,2  LIANG Bin 1,2

                                          CHENG Jianchun   1,2  LIU Xiangxiong 3

                                   (1 Institute of Acoustic, Nanjing University, Nanjing 210093, China)
                           (2 Collaborative Innovation Center of Advanced Microstructures, Nanjing 210093, China)
                               (3 Huachen Precision Equipment (Kunshan) Co., Ltd., Kunshan 215337, China)

                 Abstract  In the grinding process, the different blunt states of the grinding wheel significantly affect the
                 processing efficiency and quality. A seriously blunted grinding wheel would even lead to the occurrence of
                 waste products. Therefore, attention has been aroused on how to monitor the blunt state of the grinding
                 wheel in the grinding process. In this paper, an online monitoring method based on acoustic emission signal is
                 proposed. Firstly, the signal collected by the acoustic emission sensor is de-noised by the wavelet soft threshold
                 denoising method, following by the segmented analysis for dividing the denoised acoustic emission signal
                 into multiple overlapping segments. In the second step, by setting a threshold voltage, the acoustic emission
                 hits are intercepted for each frame of acoustic emission signal and 8 statistical features of each acoustic emission


             2018-05-05 收稿; 2018-12-28 定稿
             国家自然科学基金项目 (11374157)
             ∗
             作者简介: 钟利民 (1991- ), 男, 甘肃天水人, 硕士研究生, 研究方向: 声发射检测信号处理。
             † 通讯作者 E-mail: yangj@nju.edu.cn
   1   2   3   4   5   6   7   8   9   10