基于小波包和倒频谱分析的颤振特征提取方法

立铣加工过程中的颤振会严重影响工件表面质量和材料去除率,加剧刀具磨损和恶化工作环境。虽然大部分颤振监测系统可以监测到颤振发生,但颤振发生时已经对工件和刀具产生了严重的损伤,因此,需要提前监测到颤振特征。由于加工过程的非线性导致振动信号频率成分复杂,单一的时频分析方法难于得到可靠的颤振特征。通过小波包分解确定颤振发生频段并重构该频段信号,通过颤振发生频段的倒频谱辨识稳定、过渡和颤振状态。研究结果表明,该方法可以有效识别立铣加工过程的稳定、过渡和颤振状态。     

玻璃纤维机加工刀具失效的在线监测准则

玻璃纤维增强聚合物复合材料(GFRP)由于其特殊的材料属性被广泛应用于轻型结构中,然而其加工过程会发生刀具过度磨损,直接影响了加工工件材料的完整性。基于此,提出了一种GFRP材料加工过程中刀具磨损失效的在线监测准则,对刀具失效时的振动信号和主轴功率信号进行了采集,开展了刀具和工件的切削试验,对测量的振动信号进行分析,对该监测准则进行了试验验证。进一步建立了刀具颤振监测准则,从颤振信号分析了刀具的失效,并通过试验验证了该准则的可行性。     

基于小波包分解的刀具磨损特征分析

为了高效准确地在线监测加工高温合金过程中的刀具磨损,有效地提取刀具磨损相关特征显得尤为重要。文章提出了基于小波包分解的刀具磨损特征提取方法,将刀具切削过程中的切削力信号在时频域下分解重构,分析了各频段重构信号能量值与刀具磨损的相关性,提取了信号分解重构后小波包系数能量值中与刀具磨损相关的两个频段信号作为刀具磨损监测的特征参数,最后通过试验结果表明,采用小波包分解方法在切削力信号中提取的切削力特征和切削振动特征可作为刀具磨损特征,从而为后续研究刀具磨损在线监测提供有效输入。     

基于特征频率组的铣削颤振识别

金属切削过程中的颤振,易降低工件加工质量和机床加工精度。为实现对加工过程中颤振的在线监测,设计颤振的在线分析和识别方法十分必要。文章基于对主轴加速度信号的频域分析,提出了用于振动识别的特征频率组的分析方法,特征频率组包含了强迫振动或颤振的典型频谱信息。该方法采用加速度传感器,获取主轴的振动信号,以特征频率组为识别特征,对铣削加工的稳定性进行识别,并通过铣削试验进行验证。试验结果表明:与时域分析的方法相比,该方法利用的频域特征,无须阈值和人为判断,适用于在线的识别与分析。     

基于BP神经网络的外圆磨削颤振在线识别和监测方法

为提高机床磨削加工过程中对颤振现象识别的能力,提出一种基于BP(back?propagation)神经网络模型的颤振识别方法。通过对加工过程中传感器采集到的高频声发射信号以及振动信号相关特征值的提取,获得关于颤振的多特征参数样本库,并用其对BP神经网络模型进行学习和训练,建立BP神经网络在线识别颤振的算法模型,实现对机床加工过程中是否发生颤振的在线监测和识别。试验结果表明:这种基于BP神经网络模型的颤振识别测试结果与磨削加工试验中的磨削颤振现象结果相符合。该方法能够有效地识别磨削加工过程中的颤振,并起到在线监测识别的作用。      

钛合金车削加工过程中刀具磨损状态监测的小波包子带能量变换特征提取新方法

在小波包分析的基础上,提出对小波包子带能量特征抽取的新算法.考虑到小波包能量子带的动态特性和统计特性可以作为刀具磨损状态识别特征提取的来源,提出将小波包子带能量相对比率、小渡包子带能量相对比率的变化值、小波包子带能量相对比率的变化值的统计偏差(能量距)作为三个新特征值.建立刀具磨损状态监测实验平台,采集刀具三维力反馈、振动信号作为监测信号.按常规特征抽取方法和本研究中提出的方法抽取特征值,形成网络训练、识别特征值空间.用梯度下降法训练建立BP人工神经网络,对27具四种磨损状态进行识别,验证小波包子带能量变换提取到的特征的有效性.     

铣刀的力学和动力学广义建模

本文介绍工业上使用最多的立铣和镶齿铣刀的广义数学模型.立铣刀的几何形状用环绕刀体参量外形的螺旋槽建模.镶齿铣刀的刀刃几何,用每一刀片的局部坐标系定义,并用刀具总坐标系在刀体上对其定位和定向.对两种情况用数学表达了刀刃的坐标.使用铣削时的纯运动学,包括刀具和工件两者的结构振动,估计每一切削点处的切屑厚度.用沿与工件接触的每一切削刃或刀齿积分处理,可预断出任意立铣刀和镶齿铣刀的切削力、振动、表面粗糙度及颤振稳定性图.对螺旋锥球头、大圆孤立铣刀和镶片铣刀,预断和测量出的切削力、表面粗糙度和稳定性图,为提出的广义立铣分析的有效性作了说明.集成到先进切削加工模拟程序中的算法,可用于铣削加工中的工艺计划,以避免颤振、扭矩和功率极限约束及尺寸形成误差.     

数控机床刀具磨损状态特征参数提取

基于振动信号小波包分解理论对不平稳信号特征提取的优势,提出了一种利用振动信号的能量变化来监测刀具磨损状态的方法.该方法利用db4小波基对振动信号进行4层小波包分解,并将分解后的各频带能量值作为刀具磨损状态判断的特征参数.在新刀和刀具磨损的状态下提取特征向量,并根据频段能量的变化判断刀具磨损程度.试验结果证明该方法在刀具磨损状态判断中的可行性.     

成本低廉的智能刀柄传感器系统

在切削加工过程中，为了有效地跟踪刀具磨损、监测刀具径跳和预测次生颤振的发生，从而能在工件损坏之前提前调整切削参数，在刀具的刀尖处直接观测刀具的响应特性是非常有用的。为了以较低的成本实现这一目标，     

基于电流信号和振动信号的刀具失效监测研究

针对单一信号监测刀具失效状态存在的缺陷问题,提出电流信号和振动信号的融合方式辨识刀具失效的监测方法。研究了监测信号的预处理和特征提取方法,利用有限元分析软件建立了刀具失效状态监测的多参数融合模型,得出了刀具失效实时判定准则。研究结果表明:电流信号和振动信号的特征融合方式能有效提高刀具失效在线监测实时辨识的准确性和稳定性。     

On-line chatter detection and identification based on wavelet and support vector machine

Chatter is very harmful to precision machining process. To avoid cutting chatter effectively, a method based on wavelet and support vector machine is presented for chatter identification before it has fully developed. Wavelet transform, which can image the information in both the time and frequency domain, is applied as an amplification for the chatter premonition. Each wavelet packet's energy regularly changes during the development of the chatter, which has a time advantage for the identification. Therefore, a two-dimensional feature vector is constructed for chatter detection based on the standard deviation of wavelet transform and the wavelet packet energy ratio in the chatter-emerging frequency band. A support vector machine (SVM) is designed for pattern classification based on the feature vector. The intelligent recognition system, composed of the feature extraction and the SVM, has an accuracy rate of 95% for the identification of stable, transition and chatter state after being trained by the experiment data. The method can be applied in different machining processes.     

An expert troubleshooting system for the milling process

Common problems experienced in milling processes include forced and chatter vibrations, tolerance violations, chipping and premature wear of the tools. This paper presents an expert system which attempts to troubleshoot the source of milling problems by utilising dynamics data coupled with the opinion of the operator and acoustic Fourier spectrum data taken from the cutting process. The expert system utilises a fuzzy logic based process to interpret the signals and information, and recommends possible alterations to the process to achieve high-performance milling operations.Specific inference engines were developed to assess the chatter stability, variation in cutting force coefficient, tool run-out and forced vibration characteristics of the system. Lastly, a stability lobe plot interpretation engine to automate the lobe selection process and recommend new, chatter free cutting conditions, was also developed. The chatter stability inference engine was tested with real cutting data, through acoustic measurements taken from various cutting conditions on an aluminium milling process. The chatter inference engine successfully determined the stability of the system for each sampled cutting condition. The robustness of the troubleshooting system depends on the accuracy of acoustic and frequency response measurements.     

Chatter identification in end milling process based on EEMD and nonlinear dimensionless indicators

Vibration analysis is widely used to reveal the fundamental cutting mechanics in machining condition monitoring. In this work, vibration signals generated in different chatter conditions as well as stable cutting are studied to understand chatter characteristics. Considering the nonlinear and non-stationary properties of chatter vibration in milling process, a self-adaptive analysis method named ensemble empirical mode decomposition (EEMD) is adopted to analyze vibration signals and two nonlinear indices are extracted as chatter indicators. Firstly, the vibration signal is preprocessed with a comb filter to eliminate the interference of rotation frequency, tooth passing frequency and their harmonics. Secondly, EEMD is applied to decompose the filtered signal into a set of intrinsic mode functions (IMFs). Sensitive IMFs containing rich chatter information are selected. With the development of chatter, an accumulation phenomenon appears in the spectrum of sensitive IMFs and chatter frequencies are modulated by the rotation frequency and tooth passing frequency. Finally, two nonlinear dimensionless indices within the range of [0, 1], i.e., C₀ complexity and power spectral entropy, are extracted from the sensitive IMFs in both time domain and frequency domain. The proposed method is verified with well-designed cutting tests. It is found that, the stochastic noise dominates in the sensitive IMFs of stable cutting and both the C₀ complexity value and power spectral entropy are the largest; with the increase of chatter severity level, the periodic chatter components dominate gradually and the proportion of stochastic noise decreases, and thus these two indicators decrease.     

Experimental research on cutting force variation during primary chatter vibration occuring in plain milling operation

Plain milling operation is characterized by a transient and intermittent cutting process, in which undeformed chip thickness varies continuously. The undeformed chip thickness variation is opposite in the up milling and down milling processes. First, the property of primary chatter vibration in plain milling operation is investigated. In the up milling process, the transient vibration generated in the initial stage of the cutting operation develops into primary chatter vibration, along with the chip thickness increase. On the other hand, a large amount of vibration energy is supplied during the initial collision of the cutting edge with the workpiece at a large undeformed chip thickness in the down milling process; and immediately after this collision, the primary chatter vibration of almost stable amplitude continues. Secondly, the vibration energy supply during the primary chatter vibration of plain milling operations is investigated on the basis of the experimental results. The exciting mechanism can be explained by considering the interference between the tool flank and the workpiece surface accompanying the arbor vibration. An unusual phenomenon is also discussed, in which the normal cutting force component has two maxima during one period of vibration in up milling. From the above results, the cutting edge shapes (effective relief angle and cutting edge radius), and the torsional rigidity of the milling arbor must be carefully determined, to prevent the primary chatter vibration in plain milling operation.     

Experimental research on cutting force variation during regenerative chatter vibration in a plain milling operation

A plain milling operation is characterized by a transient and intermittent cutting process, in which undeformed chip thickness varies continuously. The reverse is the case in variations of undeformed chip thickness in the processes of up- and down-milling. In the present study, the property of regenerative chatter vibration in a plain milling operation is investigated from the viewpoint of cutting force variation. With primary chatter vibration, the vibration energy supply is closely related to the collision of the cutting tool flank against the workpiece surface during vibration, which is induced by the bending vibration or the torsional vibration of the arbor. In addition to this factor, the regenerative effect is considered to be one of the main causes of the chatter excitation in regenerative chatter vibration. The simulation result of the cutting force variation during regenerative chatter vibration agrees well with the experimental result, when considering these factors. It is shown that the regenerative chatter vibration in the-down-milling process occurs more easily than in the up-milling process.     

Analysis of acoustic emission in chatter vibration with tool wear effect in turning

The progressive wear of cutting tools and occurrence of chatter vibration often pose limiting factors on the achievable productivity in machining processes. An effective in-process monitoring system for tool wear and chatter therefore offers the unique advantage of relaxing the process parameter constraints and optimizing the machining production rate. This research presents a dynamic model of the cutting RMS acoustic emission (AE) signal when chatter occurs in turning, and it determines how this motion is related to the RMS AE signal in the presence of tool flank wear. The tool wear effect on acoustic emission generated in turning is expressed as an explicit function of the cutting parameters and tool/workpiece geometry. The AE generated from the sliding contact on the flank wear flat during chatter is investigated based on the energy dissipation principle. This model offers an explanation of the phenomenon of chatter vibration in the neighborhood of the chatter frequency of the tool. It also sheds light on the variation of the RMS AE signal power in close correlation to the characteristic of the state of wear. Cutting tests were conducted to determine the amplitude relationship between RMS AE and cutting parameters. It is shown that RMS AE is quite sensitive to the dynamic incremental changes in the friction and the wear flat mechanism active in machining processes.     

Why is it hard to identify the onset of chatter? A stochastic resonance perspective

A stochastic dynamical model is presented to identify the difficulties in chatter detection during cutting processes. The theoretical implications are based on measurements related to the stochastic character of the cutting force. The stochastic model is validated in a Hardware-In-the-Loop (HIL) environment where the multiplicative component of the stochastic cutting force is varied parametrically. In case of an industrial machine tool, the stochastic resonance effect is also demonstrated quantitatively by means of high-resolution vibration measurements for various spindle speeds in full immersion milling. The proposed method predicts the noise induced peaks in the spectrum of the vibration signals, which occur already within the chatter-free parameter domains and might be misjudged as chatter.     

Prediction of chatter in NC machining based on a dynamic cutting force model for ball end milling

Ball end milling is one of the most widely used cutting processes in the automotive, aerospace, die/mold, and machine parts industries, and the chatter generated under unsuitable cutting conditions is an extremely serious problem as it causes excessive tool wear, noise, tool breakage, and deterioration of the surface quality. Due to the critical nature of detecting and preventing chatter, we propose a dynamic cutting force model for ball end milling that can precisely predict the cutting force for both stable and unstable cutting states because our uncut chip thickness model considers the back-side cutting effect in unstable cutting states. Furthermore, the dynamic cutting force model considers both tool runout and the penetration effect to improve the accuracy of its predictions. We developed software for calculating the cutting configuration and predicting the dynamic cutting force in general NC machining as well as single-path cutting. The chatter in ball end milling can be detected from the calculated cutting forces and their frequency spectra. A comparison of the predicted and measured cutting forces demonstrated that the proposed method provides accurate results.     

模具钢硬铣削过程振动和声发射信号监测与分析

硬切削作为绿色切削的重要组成部分,已成为金属切削的一个研究热点。为了对硬切削过程进行监测,建立了一套信号采集系统,通过该系统采集模具钢铣削过程中的振动信号和声发射信号,并从时域、频域对其进行了分析研究。研究结果表明：模具钢硬铣削过程的振动信号和声发射信号的时域波形呈现不同的特点;振动信号和声发射信号的均方根值随切削速度的增大均呈明显的增大趋势,而受每齿进给量和铣削深度影响很小;随着切削速度的提高,振动信号各频段的幅值均增大,但频谱分布基本不变;随切削速度的提高,声发射信号的频谱成分增多,并导致了均方根值的增大。