Grinding wheel wear monitoring based on wavelet analysis and support vector machine

A novel grinding wheel wear monitoring system based on discrete wavelet decomposition and support vector machine is proposed. The grinding signals are collected by an acoustic emission (AE) sensor. A preprocessing method is presented to identify the grinding period signals from raw AE signals. Root mean square and variance of each decomposition level are designated as the feature vector using discrete wavelet decomposition. Various grinding experiments were performed on a surface grinder to validate the proposed classification system. The results indicate that the proposed monitoring system could achieve a classification accuracy of 99.39% with a cut depth of 10?μm, and 100% with a cut depth of 20?μm. Finally, several factors that may affect the classification results were discussed as well.     

Measurement of the normal acoustic impedance using beamforming method

A beamforming technique is introduced to measure the normal acoustic impedance at both normal and oblique incidence in a free field. In the proposed method, microphone array signals are decomposed into incident and reflected waves using an adaptive nulling algorithm, which is a type of beamforming algorithm. The acoustic impedance can then be calculated from the ratio of these two signals. To obtain better results, the pressure vector commonly used in array signal processing is replaced with the transfer function vector between each microphone, and the white Gaussian noise is suppressed by a wavelet shrinkage technique. For an accurate experimental setup, the incident and reflected angles are estimated by the multiple signal classification method with spatial smoothing. The experiments conducted in a semi-anechoic room show that the proposed method is efficient and accurate in measuring the normal acoustic impedance of sound-absorbing materials under a free field condition. This paper was recommended for publication in revised form by Associate Editor Hong Hee Yoo Jongcheon Sun received his B.S. degree in Mechanical Engineering from Korea University, Korea, in 2002. He is currently in the Unified Master’s and Doctor’s course at the School of Mechanical and Aerospace Engineering at Seoul National University, Korea. His research areas are acoustic holography and beamforming.     

Classification of acoustic noise signals using wavelet spectrum based support vector machine

Harsh noises come from air-conditioning units are chronic complaining issues to their users. Individual perceptions of noise levels have been generally quantified by means of subjective evaluation such as a jury test. This article proposes a classification approach to acoustic noise signals using a wavelet spectrum analysis. We derive energy spectrums of noise signals using a discrete wavelet transform at pre-specified window length. The energy spectrums are a linear form and represented by a Hurst parameter as an informative summary of long-range dependent signal data. The Hurst parameter controls the self-similarity scaling as well as the degree of long-range dependence. We estimate the Hurst parameter through the least squares regression of sample energy against a resolution level in the wavelet spectral domain. In the context of multi-class classification problem, the classification of noise signals is performed by a nonlinear support vector machine (SVM) for parameter estimates of linear energy profiles containing the Hurst parameter. In an application example of air-conditioner noise signals, empirical results show that the proposed method offers the higher level of accuracy in acoustic noise sound classification.     

基于LMD和主分量分析的齿轮损伤识别方法

针对齿轮故障振动信号的非平稳特征,提出了基于局部均值分解(Local mean decompos ition,简称LMD)和主分量分析的齿轮损伤识别方法。该方法首先对齿轮振动信号进行局部均 值分解,将其分解成为若干个PF分量(Product function,简称PF),然后选取包含主要损伤信息的PF分量。从PF分量中提取能量和时域统计量等特征参数,组合成初始特征参数向量矩阵, 并进一步对初始特征参数向量矩阵进行主分量分析,得到齿轮振动信号的主特征分量,建立距 离判别函数,从而对齿轮工作状态进行识别。实验数据分析结果表明,本方法能有效地识别齿 轮损伤类型。     

Application of higher order spectral features and support vector machines for bearing faults classification

Condition monitoring and fault diagnosis of rolling element bearings timely and accurately are very important to ensure the reliability of rotating machinery. This paper presents a novel pattern classification approach for bearings diagnostics, which combines the higher order spectra analysis features and support vector machine classifier. The use of non-linear features motivated by the higher order spectra has been reported to be a promising approach to analyze the non-linear and non-Gaussian characteristics of the mechanical vibration signals. The vibration bi-spectrum (third order spectrum) patterns are extracted as the feature vectors presenting different bearing faults. The extracted bi-spectrum features are subjected to principal component analysis for dimensionality reduction. These principal components were fed to support vector machine to distinguish four kinds of bearing faults covering different levels of severity for each fault type, which were measured in the experimental test bench running under different working conditions. In order to find the optimal parameters for the multi-class support vector machine model, a grid-search method in combination with 10-fold cross-validation has been used. Based on the correct classification of bearing patterns in the test set, in each fold the performance measures are computed. The average of these performance measures is computed to report the overall performance of the support vector machine classifier. In addition, in fault detection problems, the performance of a detection algorithm usually depends on the trade-off between robustness and sensitivity. The sensitivity and robustness of the proposed method are explored by running a series of experiments. A receiver operating characteristic (ROC) curve made the results more convincing. The results indicated that the proposed method can reliably identify different fault patterns of rolling element bearings based on vibration signals.     

基于FOA-MKSVM的滚动轴承故障分类方法

由于滚动轴承实际各状态数据一般具有不均衡的特点,所以分类时采用单一核函数存在一定的局限性。针对此问题以及支持向量机多参数选择的盲目性,建立一种基于果蝇优化算法的多核支持向量机模型。该模型可以通过核函数权值来调节全局核函数和局部核函数在该模型中的作用,兼具了良好的学习能力和泛化能力。同时,将多核支持向量机参数与果蝇算法中食物的味道浓度值建立一定关系,通过模仿果蝇觅食行为,对各参数进行优化选择。为了验证所提方法的有效性,先利用UCI标准数据集进行实验,再将其应用到滚动轴承故障分类中,并对单核核函数与多核核函数及参数优化算法进行比较。结果表明,提出的方法具有初始化参数少、参数设置简单、全局搜索能力强和分类准确率高的优点,可有效地应用到滚动轴承故障分类中。     

基于LCD和KNNCH分类算法的齿轮故障诊断方法

提出了一种基于局部特征尺度分解(LCD)和核最近邻凸包（KNNCH）分类算法的齿轮故障诊断方法。该方法采用LCD对齿轮原始振动信号进行分解得到若干内禀尺度分量(ISC),然后提取包含主要信息的ISC分量的能量作为特征向量输入到KNNCH分类器,根据其输出结果来判断齿轮的工作状态。实验分析结果表明,所提出的方法能有效地提取齿轮故障特征信息,而且在小样本的情况下仍能准确地对齿轮的工作状态进行识别。同时,与支持向量机(SVM)算法的对比分析结果表明,KNNCH算法能取得与SVM算法相当或更高的正确识别率。     

基于遗传算法的多尺度支持向量机及其在机械故障诊断中的应用

通过对支持向量机核函数的分析发现,当对样本的各个特征赋予不同大小的尺度参数时,可以避免冗余特征干扰分类,增强关键特征在分类中的作用,提高支持向量机分类器的学习和泛化能力。在此基础上,提出一种具有不同特征尺度参数的支持向量机(简称多尺度支持向量机),并通过遗传算法最小化LOO(leave-one-out)泛化错误上限估计,根据各个特征的识别能力赋予其不同大小的尺度参数。将多尺度支持向量机用于轴承故障诊断,实验结果表明,与传统的单尺度参数支持向量机相比,多尺度支持向量机具有更好的泛化能力。对压缩机气阀的故障识别表明,尺度参数的大小直接反映了对应特征识别能力的大小,因此可以依据尺度参数的大小进行特征选择,保留关键特征,剔除冗余特征。     

An ultrasonic flaw-classification system with wavelet-packet decomposition, a mutative scale chaotic genetic algorithm, and a support vector machine and its application to petroleum-transporting pipelines

In this paper, a novel system for ultrasonic flaw classification is proposed, which is based on wavelet-packet decomposition (WPD), a support vector machine (SVM), and a new chaotic optimization algorithm (mutative scale chaotic genetic algorithm, MSCGA). In this system, WPD is employed to extract the features of ultrasonic flaw signals, an SVM classifier is used to classify the flaws, and an MSCGA is employed as a feature selector to get rid of redundant and irrelevant features. In an experiment, a petroleum-transporting pipeline sample with various types of flaws is analyzed with this system. Experimental results show that the proposed system can improve the performance of the SVM during classification of the flaws in the petroleum-transporting pipeline. For comparison, we test the system without any feature selectors and the system with different feature selectors, respectively. The results show that the novel system is powerful and effective for ultrasonic flaw classification. This text was submitted by the authors in English.     

基于多重分形与SVM的齿轮箱故障诊断研究

针对齿轮箱振动信号的非平稳性和非线性,提出一种多重分形和支持向量机相结合的故障诊断方法。运用多重分形理论方法对齿轮振动信号进行分析,通过分析发现多重分形谱和广义维数作为故障特征能够很好地反映齿轮箱的工作状态;对支持向量机的参数利用粒子群优化算法进行优化,并将齿轮箱振动信号的多重分形特征量作为支持向量机的输入参数以识别齿轮的故障类型。实验结果表明,该方法在样本较小的情况下能够准确对齿轮箱的故障类型进行分类。     

Experimental study of burn classification and prediction using indirect method in surface grinding of AISI 1045 steel

Grinding burn is a discoloration phenomenon according to the thickness of oxide layer on the ground surface. This study tries to establish an automatic grinding burn detection system with robust burn features that are caused by burn and not by the design parameters. To address this issue, a method based on acoustic emission sensor, accelerator, electric current transducers, and voltage transducers was proposed in an attempt to extract burn signatures. A trial-and-error experimental procedure was presented to find out burn threshold. Vitrified aluminum oxide grinding wheel and AISI 1045 steel workpiece were used in the grinding test, as they were the most commonly used wheel–workpiece combinations in conventional grinding process. With the help of fast Fourier transform and discrete wavelet transform, the spectral centroid of AE signal, the maximum value of power signal, and the RMS of the AE wavelet decomposition transform from wavelet decomposition levels d1 to d5 were extracted as burn features. The spectral centroid of AE signal was believed not to be affected by grinding parameters. A classification and prediction system based on support vector machine was established in order to identify grinding burn automatically. Results indicate that the classification system performs quite well on grinding burn classification and prediction.     

基于SMO-SVM的单点金刚笔钝化监测

针对单点金刚笔在砂轮修整过程中易于钝化且难以检测的问题,使用支持向量机建立智能模型。为了得到建立模型所需的样本库,使用小波包分析等方法在线提取修整时声发射信号中的特征信息,并引入钝化平台直径定义钝化临界值。模型本身选用基于串行优化算法的支持向量分类机,使用交叉验证法搭配遗传算法以达到优化模型参数的目的。实验结果表明,该模型在分类精度和计算时间上均优于一般的智能模型,可以有效地监测金刚笔的钝化。     

Target detection and classification by measuring and processing bistatic UWB radar signal

This paper presents a novel method to detect and classify targets obscured by foliage based on real data collected by a bistatic ultra-wideband (UWB) radar system. The type of the target which passes between the transmitter and receiver can have significant effects on the shape of the received waveform. The signal measured by the bistatic UWB radar is related to the type of the target. From these received signals, we extract features that are representative of the target types. Then, we develop target type classification and recognition algorithm based on machine learning techniques. An improved support vector machine (SVM) classifier is developed to perform target types classification and recognition. A novel chaotic differential evolution (CDE) optimization approach using tent map is adopted to determine the parameters of SVM. The effectiveness of the proposed approach is verified by experiments taken in the forest.     

钢筋混凝土框架近断层速度脉冲地震响应分析

针对滚动轴承的故障诊断问题,提出了一种基于栈式稀疏自编码网络(stacked sparse auto encoder,简称SSAE)、改进灰狼智能优化算法(improved grey wolf optimization,简称IGWO)以及支持向量机(support vector machine,简称SVM)的混合智能故障诊断模型。首先,利用栈式自编码网络强大的特征自提取能力,实现故障信号深层频谱特征的自适应学习,通过引入稀疏项约束提高特征学习的泛化性能;其次,利用改进的灰狼算法实现支持向量机的参数优化;最后,基于优化后的SVM完成对故障特征向量的分类识别。所提混合智能故障诊断模型充分结合了深度神经网络强大的特征自学习能力和支持向量机优秀的小样本分类性能,避免了手工特征提取的弊端,可对不同故障类型的振动信号实现更精准的识别。多组对比实验表明,相比传统方法,笔者所提出的模型具有更优秀的故障识别能力,诊断准确率可达98%以上。     

基于支持向量机的控制图在线检测和分析系统的研究

针对控制图在线检测和分析的要求，提出了系统基本框架。利用一对一算法的多类分类支持向量机进行控制图模式识别和异常模式下参数估计。在模型构造中，采用混合核函数，并利用遗传算法优化混合核函数支持向量机参数。仿真结果和实际应用表明：该方法结构简单、收敛速度快，识别准确率高，能够满足控制图在线检测和分析的需要。     

Tool wear condition monitoring using a sensor fusion model based on fuzzy inference system

One of the biggest problems in manufacturing is the failure of machine tools due to loss of surface material in cutting operations like drilling and milling. Carrying on the process with a dull tool may damage the workpiece material fabricated. On the other hand, it is unnecessary to change the cutting tool if it is still able to continue cutting operation. Therefore, an effective diagnosis mechanism is necessary for the automation of machining processes so that production loss and downtime can be avoided. This study concerns with the development of a tool wear condition-monitoring technique based on a two-stage fuzzy logic scheme. For this, signals acquired from various sensors were processed to make a decision about the status of the tool. In the first stage of the proposed scheme, statistical parameters derived from thrust force, machine sound (acquired via a very sensitive microphone) and vibration signals were used as inputs to fuzzy process; and the crisp output values of this process were then taken as the input parameters of the second stage. Conclusively, outputs of this stage were taken into a threshold function, the output of which is used to assess the condition of the tool.     

基于MLMW和CWT灰度矩向量的滚动轴承故障诊断

提出了基于极大提升形态小波(MLMW)降噪的CWT灰度矩向量-LSSVM的轴承故障诊断方法。先利用MLMW对信号进行降噪处理,再将降噪信号的CWT灰度图划分为若干区域,计算各分区的灰度矩组成灰度矩向量,将其作为LSSVM的输入进行故障分类。试验结果表明:相对于原始信号的灰度图,MLMW降噪后的灰度图特征突出、区分显著,相应的灰度矩向量可有效刻画轴承状态;随着分区数增加,诊断准确率升高;相对于原始灰度矩向量-LSSVM方法和小波降噪的灰度矩向量-LSSVM方法,所提出方法准确率高、泛化性好、所需训练样本少,可准确识别轴承故障类型。     

Fault diagnosis of rotating machine by thermography method on support vector machine

Feature-based classification techniques consist of data acquisition, preprocessing, feature representation, feature calculation, feature selection, and classifiers. They are useful for online, real-time condition monitoring and fault diagnosis / features, which are now available with the development of information technologies and various measurement techniques. In this paper, an intelligent feature-based fault diagnosis is suggested, developed, and compared with vibration signals and thermal images. Fault diagnosis is performed using thermal imaging along with support vector machine (SVM) classification to simulate machinery faults, resulting in an accuracy level comparable to vibration signals. The observed results show that fault diagnosis using thermal images for rotating machines can be applied to industrial areas as a novel intelligent fault diagnostic method with plausible accuracy. It can be also proposed as a unique non-contact method to analyze rotating systems in mass production lines within a short time.