为了提取高精度磨削干涉中的声发射信号特征, 实现砂轮磨削性能退化评估, 针对熔石英开展全寿命周期金刚石砂轮磨削实验, 基于小波包分析确定砂轮磨损敏感频段为低频段, 然后计算声发射信号在低频段的归一化能量占比, 再利用主成分分析对能量占比进行特征降维获得单值特征, 利用该单值特征绘制砂轮磨削性能退化曲线。研究结果表明, 监测特征能够清晰反映砂轮初期磨损、正常磨损和过度磨损三个阶段, 且监测结果不受加工参数影响; 砂轮磨削材料破裂尺度与声发射频率具有一定关系, 伴随砂轮磨损的加剧, 较大尺度破裂的比例上升, 造成65 kHz以下低频段特征的能量占比增大, 监测特征显著增加, 磨粒崩碎产生新的切削刃, 砂轮的去除能力有所改善, 监测特征数值回落, 但是, 不同样本的声发信号频谱差异性显著增加, 说明砂轮加工状态不稳定, 不利于精密与超精密加工中维持稳定质量的要求。砂轮形貌图像的白像素占比变化曲线验证了声发射特征对砂轮磨损状态判断的正确性。

Acoustic emission is the most direct and sensitive characterization of grinding interference. Therefore, to ensure high-precision grinding, acoustic emission signals were used to evaluate the degradation in grinding performance of a diamond wheel. A fused silica grinding experiment was conducted using a newly trimmed diamond grinding wheel to record the acoustic emission signal and surface morphology of the grinding wheel during its life cycle. A wavelet packet analysis was used to determine the typical low-frequency band of the abrasive wear of the grinding wheel. The frequency energy ratio was then extracted by using principal component analysis to characterize the degradation of grinding wheel grinding performance. The results showed that the acoustic emission frequency energy ratio was completely independent of the processing parameters. As the abrasive wear of the grinding wheel intensified, both the proportion of large-scale rupture and the amplitude of the corresponding low-frequency characteristic increased. When the abrasive particles were worn to the limit, the abrasive grain was broken to produce a new cutting edge, improving the removal ability of the grinding wheel, but the variance of the spectrum was significantly increased, inhibiting the wheel from maintaining a stable quality in ultra-precision machining. The variation of the white pixel ratio of the grinding wheel shape image verified the correctness of the acoustic emission signal analysis results.