基于颤振预报的螺杆转子智能自抑振加工理论与方法研究

Chatter often occurs during precision screw rotor grinding processing, and it results in low quality of finished surface, low processing efficiency and high scrap rate. Prediction control is the most effective chatter suppression method, which can suppress chatter rapidly to prevent damage to the surface of precision screw rotor. An approach includes chatter recognition, prediction and suppression for precision screw rotor grinding processing is presented in this project, and a chatter self-suppression processing system include signal sensing, processing, analyzing and feedback control is established. 1) In order to reveal the self-excitation mechanisms of regenerative chatter, the dynamic model of screw rotor grinding process is established, and the stability lobe diagram is derived. 2) Empirical mode decomposition (EMD) is introduced to decompose the vibration signal and extract the chatter feature, and the signal-noise separation is accomplished by independent component analysis (ICA), so the chatter symptom is separated rapidly and effectively. 3) To suppress the grinding chatter quickly, a magneto-rheological (MR) fluid based intelligent fixture is designed to change the structural damping and stiffness of screw rotor grinding system online.

由于金属切削加工的再生效应，螺杆转子精磨过程中往往会发生颤振，导致了螺杆转子加工精度低、表面质量差、报废率高、加工效率不高等问题，针对上述问题，本项目提出一种基于颤振预报的螺杆转子智能自抑振加工方法，将对螺杆转子磨削颤振的在线识别、预报及其控制技术进行深入研究，建立一个集信号传感、调理、处理分析和反馈控制于一体的智能自抑振加工系统。1）建立螺杆转子磨削加工系统颤振动力学模型，推导金属切削稳定性极限曲线，从而探索螺杆转子加工颤振产生机理；2）构建一种新式的颤振征兆识别方法，采用经验模态分解直接从加工刀具振动信号中分离有效的时频特征量，并通过独立成分分析分离得到与原始振动信号相关度较高的颤振征兆特征，提高了颤振识别的准确率；3）设计基于磁流变液的支撑刚度/阻尼可变型螺杆转子夹具，在颤振被识别预报出来时实时调控磨削加工系统的结构阻尼/刚度，破坏颤振形成的条件，达到快速有效抑制颤振的效果。

针对螺杆转子加工过程容易产生加工颤振而导致表面加工质量降低的问题，本项目提出了一种基于颤振预报的螺杆转子智能自抑振加工方法，通过开展颤振产生机理及抑制方法、颤振征兆提取、模式识别等研究，建立了集信号传感、调理、处理分析和反馈控制于一体的智能自抑振加工系统。1）建立螺杆转子加工系统动力学模型，推导出金属切削稳定性极限曲线，从而探索螺杆转子加工颤振产生机理及抑制方法；2）提出一种基于EMD和HHT的颤振征兆提取方法，并利用ICA进行信噪分离，提高颤振征兆分离效果；3）提出一种基于HMM-SVM混合模型的加工状态识别方法，对螺杆转子加工状态进行快速、准确识别；4）设计了一种基于磁流变液的智能自抑振刀架，通过实时调节加工系统的阻尼刚度而达到颤振抑制效果。开展了相关实验研究，实验结果表明：本项目所提出的方法能对颤振进行快速、准确识别，识别时间1.5秒以内，识别准确率超过95%，所提出基于磁流变液的颤振抑制效果良好，振幅降低50%以上，工件表面粗糙度Ra从10.3μm降低到1.5μm。