基于小波包和支持向量机的滚动轴承故障模式识别

为了解决对故障轴承的特征提取和故障特征准确分类问题,提出了应用小波包变换和支持向量机相结合进行滚动轴承故障诊断的方法.小波包变换具有良好的时-频局部化特征,非常适于对瞬态或时变信号进行特征提取.而支持向量机可完成模式识别和非线性回归.利用上述原理根据轴承振动信号的频域变化特征,采用小波包变换对其提取频域能量特征向量,然后利用建立的支持向量机多故障分类器完成滚动轴承故障模式的识别.试验结果表明,支持向量机可以有效、准确地识别轴承的故障模式,为轴承故障诊断向智能化发展提供了新的途径.     

Chatter identification in end milling process using wavelet packets and Hilbert–Huang transform

Chatter detection is an important task to improve productivity and part quality in the machining process. Since measured signals from sensors are usually contaminated by background noise and other disturbances, it is necessary to find efficient signal processing algorithms to identify the chatter as soon as possible. This paper presents an effective chatter identification method for the end milling process based on the study of two advanced signal processing techniques, i.e., wavelet package transform (WPT) and Hilbert–Huang transform (HHT). The WPT works as a preprocessor to denoise the measured signals and hence the performance of the HHT is enhanced. The proposed method consists of four steps. First, the measured signals are decomposed by the WPT, so that the chatter signals are allocated in a certain frequency band. Secondly, wavelet packets with rich chatter information are selected and are used to reconstruct new signals. Thirdly, the reconstructed signals are analyzed with HHT to obtain a Hilbert–Huang spectrum, which is a full time–frequency–energy distribution of the signals. Finally, the mean value and standard deviation of the Hilbert–Huang spectrum are calculated to detect the chatter and identify its levels as well. The proposed method is applied to the end milling process and the experimental results prove that the method can identify the chatter effectively.     

基于经验小波变换和流形学习的滚动轴承故障诊断研究

针对振动信号的非线性、非平稳性和早期故障特征信号难以提取的特点,提出一种基于经验小波变换(EWT)和流形学习约简的故障特征提取方法。首先利用EWT将振动信号分解成不同特征时间尺度的单分量固有模态函数(IMF),然后从包含故障信息的IMFs中提取滚动轴承的时域统计特征、频域统计特征、AR模型自回归系数和功率谱熵,构造高维特征集;再利用线性局部切空间排列(LLTSA)流形学习算法将构造的高维特征集约简为故障区分度更好的低维特征集;最后利用支持向量机(SVM)对提取特征进行故障识别。实验结果表明该特征提取方法对滚动轴承故障诊断准确率更高。     

基于支持向量机的火车滚轴故障诊断

火车故障的60%都是由于滚轴问题引起的,现在的诊断方法都是基于知识的,但故障样本的不足从一定程度上制约了基于知识的方法在实际中的应用,针对这一问题,利用支持向量机在小样本情况下具有较强分类能力的特点,本文提出了一种基于支持向量机的滚轴故障诊断方法.该方法采用小波变换对齿轮的震动信号进行处理来构造特征向量,然后输入到支持向量机分类器中进行模式识别.     

基于EMD和支持向量机的滚动轴承故障诊断研究

为解决在复杂噪声和工频及其倍频干扰条件下滚动轴承故障诊断问题,提高诊断准确率,进行了经验模态分解(EMD)和支持向量机(SVM)的研究,给出了相应的决策流程。基于改进的EMD分解的特征提取算法,选取故障特征明显的IMF分量进行特征提取,最大限度地滤除了低频噪声干扰,捕捉到信号的故障特征,然后将特征集输入到SVM分类器中进行识别,结果表明:该方法对于轴承故障识别具有较高的准确率,为确保轴承安全运行和快速故障诊断提供了理论支持。     

基于小波包分解与K L变换的齿轮泵振动信号故障特征提取方法

针对齿轮泵故障成因复杂、模糊性强的特点,结合小波包分解与K-L变换,提出一种适用于支持向量机故障诊断的特征提取方法。通过小波包对样本故障振动信号进行分解得到特征向量,而后利用K-L变换处理得到新的特征向量集,达到降维去噪的目的。将处理后的特征向量集用于支持向量机的模型训练,分析结果表明:该方法能够有效提高故障模式识别准确率和识别效率。     

Machine ensemble approach for simultaneous detection of transient and gradual abnormalities in end milling using multisensor fusion

In a fully automated manufacturing environment, instant detection of the cutting tool condition is essential for the improved productivity and cost effectiveness. This paper studies a tool condition monitoring system (TCM) via machine learning (ML) and machine ensemble (ME) approach to investigate the effectiveness of multisensor fusion technique when machining 4340 steel with multilayer coated and multiflute carbide end mill cutter. In this study, 135 different features are extracted from multiple sensor signals of force, vibration, acoustic emission and spindle power in the time and frequency domain by using data acquisition and signal processing module. Then, a correlation-based feature selection technique (CFS) evaluates the significance of these features along with machining parameters collected from machining experiments. Next, an optimal feature subset is computed for various assorted combinations of sensors. Finally, machine ensemble methods based on majority voting and stacked generalization are studied for the selected features to classify not only flank wear but also breakage and chipping. It has been found in this paper that the stacked generalization ensemble can ensure the highest accuracy in tool condition monitoring. In addition, it has been shown that the support vector machine (SVM) outperforms other ML algorithms in most cases tested.     

A new on-line spindle speed regulation strategy for chatter control

A new on-line control method to suppress regenerative chatter vibration during the machining process by regulating spindle speed is proposed. The dynamic cutting force signal collected from a dynamometer is passed through a low pass filter, and then digitized. The fast Fourier transform is carried out to obtain the corresponding power spectrum. The chatter frequency is identified when the intensity at a certain frequency other than the spindle speed and tooth passing frequency exceeds a critical value. Based on the identified chatter frequency, a new spindle speed is computed by applying the principle of keeping the phase between the present and previous undulations to 90°. The new speed command is executed while the cutting proceeds. It is found from simulation that the chatter vibration can be suppressed by this approach in the shortest time. This method is also verified by experiments through actual cutting of various materials by a computer numerically controlled milling machine. The main feature of this approach is that the feed of the machine tool does not need to be halted during the change of spindle speed. Hence, tool wear can be reduced. Furthermore, no system identification of the machine tool structure is needed, and therefore it has great potential in actual applications.     

基于支持向量机的机械加工误差预测与补偿模型的研究

对加工系统进行补偿是提高机械加工精度的有效手段.通过对加工系统的研究,建立误差预测模型,是进行误差补偿的必要途径.本文以镗孔加工为实验对象,提出了基于支持向量机(Suport Vector Machine,SVM)的加工系统误差预测模型,实验结果显示,支持向量机可以应用于误差预测建模,且在系统误差的预测精度上高于基于径向基(RBF)神经网络的误差预测模型.     

基于ANN和SVM的轴承剩余使用寿命预测

为了提高现代制造业设备的可靠性和高效性,轴承剩余使用寿命(RUL)的预测已经成为越来越重要的研究方向.提出一种基于人工神经网络(ANN)和支持向量机(SVM)的轴承剩余使用寿命预测方法.该方法首先将获取的18维反映轴承衰退的时域特征和频域特征输入到ANN模型中做特征抽取,再将输出的18维特征向量作为SVM模型的输入,进...     

基于多特征提取和LSSVM的轴承故障诊断

针对故障滚动轴承的振动信号具有非线性、非平稳的特点,提出一种基于时域指标、小波包能量和最小二乘支持向量机(LSSVM)的轴承故障诊断方法。分别对滚动轴承的原始信号进行时域分析计算和小波包分解,并提取状态差异较明显的时域指标和小波包分解后能量差异较大的小波包能量作为故障特征向量;将含有多个特征向量的数据样本分为训练样本和测试样本并进行归一化处理;训练样本作为LSSVM的输入来对该模型进行训练,通过训练好的LSSVM模型对测试样本进行分类和诊断。实验结果表明:采用该方法,轴承状态总体识别率为97.5%。     

混沌-支持向量机在加工误差预测中的应用

提出将混沌-支持向量机模型方法应用于加工误差数据预测。利用互信息法和曹氏方法进行相空间重构,并运用小数据量法计算最大Lyapunov指数,对加工误差时间序列进行混沌识别。通过最小二乘支持向量机对历史样本的学习建立预测模型,并将其预测结果与RBF神经网络预测结果进行仿真对比。结果表明,在较少的加工误差数据条件下,该模型能够有效地描述和预测加工误差的变化,具有较高的预测精度。     

基于小波包分解和奇异谱分析方法的立铣加工颤振监测

在立铣加工过程中,颤振是加工过程失稳的一个最重要的原因。颤振将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。在颤振发生过程中,振动信号具有在时域中不断增大,在频域中能量频移的特性。考虑这两个振动信号特征,提出了一种颤振特征提取方法。提取颤振发生频带中振动信号的能量比和奇异谱熵系数作为两个颤振特征,并通过人工神经网络模型实现切削颤振的识别。文中提出的颤振监测系统包括特征提取和分类,能够精确辨识立铣加工中的稳定、过渡和颤振状态。     

基于改进贝叶斯分类的电机轴承故障诊断系统研究

针对电机轴承故障诊断模型构建时间长、准确率不高的问题,提出一种基于改进贝叶斯分类的故障诊断方法。首先通过小波包变化、粗糙集及主成分分析方法分别构造原始故障特征集、降维后的故障特征集,再将原始故障特征集和降维后的故障特征集输入到改进贝叶斯分类模型中实现故障诊断,以此为基础设计一套交流发电机轴承故障诊断系统。最后以国内车辆车载电机轴承振动数据为依据,将改进贝叶斯分类方法和神经网络及最小二乘支持向量机方法作对比分析,结果表明：改进贝叶斯分类方法建模时间更短,故障诊断准确率更高。     

A novel approach to machining condition monitoring of deep hole boring

In the optimization of deep hole boring processes, machining condition monitoring (MCM) plays an important role for efficient tool change policies, product quality control and lower tool costs. This paper proposes a novel approach to the MCM of deep hole boring on the basis of the pseudo non-dyadic second generation wavelet transform (PNSGWT). This approach is developed via constructing a valuable indicator, i.e., the wavelet energy ratio around the natural frequency of boring bar. Self-excited vibration occurs at the frequency of the most dominant mode of the machine tool structure. Via modeling dynamic cutting process and performing its simulation analysis during deep hole boring, it is found that the vibration amplitudes at the nature frequency of the machine tool rise with the tool wear. The PNSGWT that has relative adjustable dyadic time-frequency partition grids, good time-frequency localizability and exact shift-invariance is used to extract the wavelet energy in the specified frequency band. Accordingly, the MCM of deep hole boring can be implemented by means of normalizing the wavelet energy. Finally, a field experiment on deep hole boring machine tool is conducted, and the result shows that the proposed method is effective in the process of monitoring tool wear and surface finish quality for deep hole boring.     

基于全矢最小二乘支持向量机的设备状态趋势预测

在支持向量机(SVM)基础上拓展出的最小二乘支持向量机(LS-SVM)非线性泛化能力更好,具有较高的拟合和预测精度,目前被广泛应用于设备状态趋势预测中。为进一步提高其预测精度,结合基于同源信息融合的全矢谱技术提出一种新的趋势预测方法——全矢LS-SVM。该方法采用全矢谱技术融合双通道信息,相比传统单通道信号提取方法,保障LS-SVM预测数据特征提取的完整性,提高预测精度。将该方法应用于某电厂1号汽轮机振动数据的预测,实验结果表明,全矢LS-SVM方法具有较高的预测精度。     

基于VMD与IFWA-SVM的滚动轴承故障诊断研究北大核心

为有效提取非平稳性、复杂性的滚动轴承振动信号特征,提出一种基于变分模态分解、改进烟花算法(IFWA)优化支持向量机(SVM)的滚动轴承故障诊断方法。利用VMD对原始信号进行分解,计算得到各IMF的样本熵,将原始信号的时域特征与其结合组成特征矩阵。为提高故障诊断效率,采用IFWA优化SVM,建立IFWA-SVM模型。使用训练集特征矩阵训练诊断模型,实现滚动轴承的故障诊断。利用实测信号验证该方法,并与粒子群算法优化进行比较。结果表明:利用该方法进行诊断,正确率提高了3.33%、训练时间缩短了21.55 s,验证了该方法的可行性。     

金刚石滚轮修整时的径向圆跳动状态在线判别

金刚石滚轮修整砂轮时的性能受其径向圆跳动的影响,而其径向圆跳动状态判别的智能化程度较低。为此,对金刚石滚轮修整状态下的径向圆跳动磨削声发射信号,提出一种基于小波分解和SVM的在线检测方法。将磨削声发射信号通过小波变换并分解,提取小波分解系数的有效值、方差及能谱系数3种特征参数。结果表明:将3种特征参数彼此组合输入到SVM中进行状态识别时的准确率都在96.0%以上;3种特征参数同时输入时的准确率最高,达到了98.3%。该检测方法具有实际应用价值。      

基于支持向量机的蠕墨铸铁表面粗糙度预测

目的准确预测蠕墨铸铁加工过程中的表面质量,指导加工参数调整,保证加工过程中加工质量的稳定,运用差分进化算法优化的SVM模型(DE-SVM)构建蠕墨铸铁表面粗糙度(Ra)预测模型和加工参数选择方法。方法采用DE-SVM提高支持向量机回归模型的预测精度,建立针对实际加工材料的表面粗糙度预测模型,基于构建的预测模型,挖掘表面粗糙度与加工参数之间的关系,从而获得较优的加工参数。结果结合蠕墨铸铁的铣削加工实验数据,对比DE-SVM与常用优化算法(粒子群优化算法(PSO)和遗传算法(GA))优化的SVM模型,DE-SVM模型获得的MAPE(0.122)和R2(0.9559)值均优于粒子群和遗传算法优化的支持向量模型获得MAPE和R2值。在给定的加工参数范围内,切削速度和进给速度对表面粗糙度的影响较大,且表面粗糙度与切削速度成正比关系,与进给速度成反比,而切削深度对表面粗糙度影响不显著。结论由实验的对比结果可知,采用DE-SVM模型建立的蠕墨铸铁表面粗糙度模型具有更高的预测精度,基于DE-SVM获得的加工参数对表面粗糙度的影响,可有效指导加工参数的选择与调整,对保持蠕墨铸铁优良的加工质量具有较好的指导意义。     

A new perspective and analysis for regenerative machine tool chatter

The paper contains a practical perspective on regenerative machine tool chatter. Chatter is a well known phenomenon, occurrence of which is undesired in manufacturing. Aggressive machining conditions, in the sense of removing more metal rapidly, usually cause chatter. In most cases, these conditions can be determined a priori to the operation. A chatter stability study and its reasoning based on root locus plot analysis of time delayed systems is presented as a new and practical perspective in the field. At the junction of root locus and chatter concepts an area of particular interest to the authors arises: a new method for active vibration suppression, the Delayed Resonator. It is an active vibration absorber tuning of which is achieved utilizing a simple time delayed feedback. The cross linking between the Delayed Resonator study and the subject matter, machine tool chatter, is exciting to share. This is the primary motivation in pursuing this study. One of the highlights of the work appears at the phenomenon called Dual Frequency Delayed Resonator. This feature has been conjectured in the literature using the well known “stability lobes”, but never discussed with detail.