硬脆材料旋转超声加工高频振动效应及刀具磨损智能监测系统研究

Due to its the small size, low bearing capacity, the hard-brittle micro-structure component could repture and fracture easily in the rotary ultrasonic machining process. In addition, the abrasives on the tool tended to detach, which brought it difficult to guarantee the machining quality. BK7 glass micro-structures are investigated in this topic, and a novel dimensionless parameter is defined to characterize the effects of the ultrasonic vibration to realize the quantitative characterization of the high-frequency vibration effects in the rotary ultrasonic machining process of hard-brittle material. Based on these investigations, a new surface formation process model adopted to the RUM prosess is proposed. The nucleation mechanisms of incipient cracks and the formation mechanisms of material pulverization are analysized, also the evolutionary mechanisms of the pulverization layer to the chipping/cracking layer is investigated, and the brittle-ductile transition critical condition is determined. By considering the combined effects of the elastic and plastic stress fields on the cracks, a nonlinear model between the surface roughness and the subsurface damage depth is proposed in order to realize the effective prediction of subsurface damage depth. The relationships between signal characteristics in multi-physics domain and tool wear are established using the empirical mode decomposition method and generalized hidden Markov model，and a novel intelligent monitoring system of tool wear based on multi-physics domain information fusion is developed in the end. This research has a great significance to improve the quality and yields of military products.

硬脆材料微结构件的尺寸小、承载能力低，在旋转超声加工时易发生边角破裂和脆性断裂，刀具磨粒容易脱落，加工质量难以保证。本课题以BK7玻璃材料微结构件为研究对象，用自定义的无量纲特征参数表征高频振动效应的强弱，实现对硬脆材料旋转超声加工中高频振动效应的定量表征，提出适用于旋转超声加工的表面形成过程新模型。分析初始裂纹成核的原因以及材料粉末化的形成机理，研究材料的惯性效应引起的粉末化层向切屑层/裂纹层的演化机制，确定旋转超声加工过程中脆性材料的脆-塑转变临界条件。综合考虑弹性应力场和塑性应力场对裂纹扩展深度的影响，建立工件表面粗糙度与亚表层裂纹扩展深度之间的非线性关系模型，实现对亚表层损伤深度的有效预测。利用经验模式分解法、广义隐Markov模型建立不同信号特征与刀具磨损状态之间关系模型，开发出基于多物理域信息融合的刀具磨损智能监测系统。本项研究工作对提高军工产品质量和成品率具有极其重要的意义。

硬脆材料微结构件的尺寸小、承载能力低，在旋转超声加工时易发生边角破裂和脆性断裂，刀具磨粒容易脱落，加工质量难以保证。本课题以硬脆材料微结构件为研究对象，用自定义的无量纲特征参数表征高频振动效应的强弱，实现了对硬脆材料旋转超声加工中高频振动效应的定量表征，提出了适用于旋转超声加工的表面形成过程新模型。分析了初始裂纹成核的原因以及材料粉末化的形成机理，研究了材料的惯性效应引起的粉末化层向切屑层/裂纹层的演化机制。基于脆性材料的动态压痕断裂力学理论，推导了磨粒加载过程中材料应变率的计算公式，实现了对旋转超声加工中材料应变率的定量描述，确定了旋转超声加工过程中脆性材料的脆-塑转变临界条件。对旋转超声加工中刀具端面磨粒的微观破碎形貌进行了分析，发现磨粒的破碎形态主要为材料崩碎所产生的浅表层贝壳状缺陷。借助于霍普金森杆冲击实验测试了材料的动态力学性能，结果表明高应变率效应使材料的弹性模量增加，从而降低了磨粒内部裂纹的成核深度，最终导致磨粒的微观破碎。综合考虑弹性应力场和塑性应力场对裂纹扩展深度的影响，建立了工件表面粗糙度与亚表层裂纹扩展深度之间的非线性关系模型，实现了对亚表层损伤深度的有效预测。对旋转超声加工中刀具整个磨损阶段进行了划分，提取出能够反映刀具磨损状态的振动和切削力信号的时域、频域和相对小波包能量特征等信息，获得了对刀具磨损敏感而对工艺参数不敏感的联合多特征向量组。利用改进GA-BP神经网络对旋转超声加工过程中刀具磨损状态进行识别，显著提高了刀具磨损状态的识别率。通过多传感器信息融合技术将不同信号的特征信息进行了融合，从多个物理层面反映与刀具磨损状态相关的信息，基于Labview软件开发出基于多物理域信息融合的刀具磨损智能监测系统，并通过仿真实验验证该系统的可靠性。本项研究工作对提高军工产品质量和成品率具有极其重要的意义。