以金刚石压头划刻BK7光学玻璃为研究对象, 分析了脆性材料脆性去除过程中的声发射机制, 研究了声发射信号的特征提取技术。多种切深实验显示: BK7光学玻璃发生脆性去除的特征主要集中在［100, 200］kHz、［300, 400］kHz两个频段, 对应不同的声发射机制, 其中［100, 200］kHz频带的滤波信号呈现明显的、时间间歇的突发式声发射现象, 与脆性材料裂纹的生成与扩展密切相关。基于上述实验结果, 提出了以突发式声发射事件为单位的特征监测方法。针对该带通滤波信号的均方根值(RMS), 研究了基于凸优化理论的声发射事件识别算法, 得到了脆性材料裂纹扩展的时刻及能量信息。得到的结果表明: 以声发射事件为单位的特征监测具有明确的物理意义, 能够更加客观地表征脆性材料的去除过程。

By focusing on the diamond scratching of BK7 optical glass, the Acoustic Emission(AE) mechanism in the brittle removal of optical brittle materials was analyzed and a feature extraction technique of AE signals used in processing and monitoring of optical brittle materials was studied. Various cutting depth test results show that features of brittle removal for optical glass BK7 mainly focus on two frequency bands that are ［100, 200］ kHz and ［300, 400］ kHz, and they correspond to different AE mechanisms. In which, filtered signal of frequency band ［100, 200］ kHz presents obvious burst-type AE phenomenon with a time interval, which is closely related to the production and extension of cracks for optical brittle materials. On the basis of the results mentioned above, a monitoring method that uses burst-type AE events as unit was proposed. Aimed at RMS(Root Mean Square) signals of the band-pass filtering, a recognition algorithm of AE events based on convex optimization theory was studied to get the time and energy information of crack growth for optical brittle materials. It concludes that the feature monitoring method that uses AE events as unit has specific physical meanings and represents removal process of optical brittle materials more objectively.