基于支持向量机的机械故障特征选择方法研究

在机械故障诊断中,对机器状态信号进行处理可得到故障特征集。但是此特征集中通常含有冗余特征而影响诊断效果。特征选择可以去除原始特征中的冗余特征,提高诊断精度和诊断效率。本文提出采用支持向量机(SVM)作为决策分类器,研究了使用SVM的错误上界如半径-间距上界代替学习错误率作为特征性能评价,并且使用遗传算法对特征集进行寻优的特征选择方法。此方法由于只需要训练一次SVM,相比常用的分组轮换方法有较高的计算效率。数值仿真和减速器的轴承故障特征选择试验中,采用此方法对生成特征集进行选择,并与常用的分组轮换法进行了对比。结果显示此方法有较好的选择性能和选择效率。     

Application of the PSO–SVM model for recognition of control chart patterns

Control chart patterns are important statistical process control tools for determining whether a process is run in its intended mode or in the presence of unnatural patterns. Accurate recognition of control chart patterns is essential for efficient system monitoring to maintain high-quality products. This paper introduces a novel hybrid intelligent system that includes three main modules: a feature extraction module, a classifier module, and an optimization module. In the feature extraction module, a proper set combining the shape features and statistical features is proposed as the efficient characteristic of the patterns. In the classifier module, a multi-class support vector machine (SVM)-based classifier is proposed. For the optimization module, a particle swarm optimization algorithm is proposed to improve the generalization performance of the recognizer. In this module, it the SVM classifier design is optimized by searching for the best value of the parameters that tune its discriminant function (kernel parameter selection) and upstream by looking for the best subset of features that feed the classifier. Simulation results show that the proposed algorithm has very high recognition accuracy. This high efficiency is achieved with only little features, which have been selected using particle swarm optimizer.     

A novel scheme for fault detection of reciprocating compressor valves based on basis pursuit,wave matching and support vector machine

A scheme for fault detection of compressor valves based on basis pursuit (BP), wave matching and support vector machine (SVM) is presented. BP is applied to extract the main vibration component in the signal and suppress background noise. Wave matching is a new feature extraction method proposed in this paper. Instead of extracting features through commonly used indicators such as statistic measures or information entropy, wave matching extracts features by matching the vibration signal with parameterized waveform optimized by differential evolution (DE) algorithm. It only produces a small number of features and the features have clear physical meaning. SVM is employed in the fault classification because of its superiority in dealing with small sample problems. The results of real compressor valve signal analysis confirm that the proposed scheme can differentiate compressor valve faults with high accuracy and reliability.     

基于GMKL-SVM的模拟电路故障诊断方法

提出了一种新颖的基于广义多核支持向量机(GMKL-SVM)的模拟电路故障诊断方法。首先,应用Haar小波分析提取被测电路时域响应信号的小波系数作为特征参量,并生成样本数据;然后,基于样本数据,应用量子粒子群算法对GMKL-SVM的参数进行优化,并以此建立基于GMKL-SVM的故障诊断模型,用于区分模拟电路的各个故障。实例电路的单故障和双故障诊断实验结果表明,所提出的GMKL-SVM方法能较好地实现模拟电路故障诊断,与传统的GMKL-SVM方法相比,表现出了更好的性能,获得了更高的故障诊断正确率。     

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.     

基于局域均值分解包络谱和SVM的滚动轴承故障诊断方法研究

论述了局域均值分解(Local mean decomposition,LMD)的定义和算法。结合局域均值分解、包络分析和支持向量机(Support vector machine,SVM)的各自特点,提出了一种基于LMD包络谱和SVM的滚动轴承故障诊断方法,该方法先对滚动轴承振动信号进行分解,得到一系列的生产函数分量,然后,再对前面几个生产函数分量进行包络分析,从包络谱中提取特征幅值比作为特征向量输入到SVM分类器中进行识别。实验结果验证了提出的方法的有效性,可以有效地识别滚动轴承的不同故障。     

Predicting the spontaneous termination of atrial fibrillation based on poincare section in the electrocardiogram phase space

Atrial fibrillation (AF) is a commonly encountered cardiac arrhythmia. Predicting the conditions under which AF terminates spontaneously is an important task that would bring great benefit to both patients and clinicians. In this study, a new method was proposed to predict spontaneous AF termination by employing the points of section (POS) coordinates along a Poincare section in the electrocardiogram (ECG) phase space. The AF Termination Database provided by PhysioNet for the Computers in Cardiology Challenge 2004 was applied in the present study. It includes one training dataset and two testing datasets, A and B. The present investigation was initiated by producing a two-dimensional reconstructed phase space (RPS) of the ECG. Then, a Poincare line was drawn in a direction that included the maximum point distribution in the RPS and also passed through the origin of the RPS coordinate system. Afterward, the coordinates of the RPS trajectory intersections with this Poincare line were extracted to capture the local behavior related to the arrhythmia under investigation. The POS corresponding to atrial activity were selected with regard to the fact that similar ECG morphologies such as P waves, which are corresponding to atrial activity, distribute in a specific region of the RPS. Thirteen features were extracted from the selected intersection points to quantify their distributions. To select the best feature subset, a genetic algorithm (GA), in combination with a support vector machine (SVM), was applied to the training dataset. Based on the selected features and trained SVM, the performance of the proposed method was evaluated using the testing datasets. The results showed that 86.67% of dataset A and 80% of dataset B were correctly classified. This classification accuracy is in the same range as or higher than that of recent studies in this area. These results show that the proposed method, in which no complicated QRST cancelation algorithm was used, has the potential to predict AF termination.     

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KPCA和PSOSVM的异步电机故障诊断

针对异步电机故障振动信号具有较强的非线性特征,而传统的线性分析方法易造成振动信号非线性成分的丢失这一情况,提出一种核主元分析和粒子群支持向量机相结合的异步电机故障诊断方法。利用核函数实现输入空间到高维特征空间的非线性映射以及对映射数据的主元分析,得到原始样本的非线性主元,实现特征提取和数据压缩,将获得的核主元特征通过支持向量机进行模式识别。采用距离比值法和粒子群算法分别对核主元分析和支持向量机的参数进行双重优化选择。实验结果表明,该方法能有效提取故障信号的非线性特征,具有较强的非线性模式识别能力,相比主元分析和支持向量机方法,分类效果更好,实时性更强,可快速有效实现异步电机故障诊断。     

基于支持向量机和模态频率的结构损伤诊断

支持向量机（support vector machine,SVM）是一种基于统计学习理论的机器学习算法,能够较好地解决小样本的学习问题。文中介绍支持向量机回归算法,并应用于结构损伤诊断领域;构造基于模态频率的损伤标识量,作为特征参数训练支持向量机实现对结构损伤的定位和程度标识;最后以梁的损伤识别为例进行验证。结果表明,支持向量机在结构损伤诊断领域中具有很好的应用前景。     

基于蚁群节点寻优的贝叶斯网络结构算法研究

K2算法是学习贝叶斯网络结构的经典算法。针对K2算法依赖最大父节点数和节点序的不足,以及蚁群算法搜索空间庞大的问题,提出了一种新的贝叶斯结构学习算法-MWST-ACO-K2算法。该算法通过计算互信息建立最大支撑树(MWST),得到最大父节点数;然后利用蚁群算法(ACO)搜索最大支撑树,获得节点顺序;最后结合K2算法得到最优的贝叶斯网络结构。仿真实验结果表明,该方法不仅解决了K2算法依赖先验知识的问题,而且减少了蚁群算法的搜索空间,简化了搜索机制,得到较好的贝叶斯结构。最后将该算法应用到冀东水泥回转窑的实际数据中,构建水泥回转窑的贝叶斯网络结构,提高了故障诊断的准确率。     

QPSO-WT和QPSO-SVM在滚动轴承故障诊断中的应用

为了解决小波降噪软阈值选择非最优以及SVM算法中惩罚参数、核函数参数的设置问题,将小波变换、支持向量机分别与量子行为粒子群优化算法QPSO(quantum-behaved particle swarm optimization,)相结合,利用QPSO优化小波阈值以及优化SVM输入参数,进行全局寻优,并将之应用到滚动轴承故障识别中。实验中,QPSO-WT滤波后信号具有更高的信噪比和更低的MSE,QPS0-SVM对10种不同状态的轴承进行故障诊断,对于多分类的情况该方法的识别精确度达到了87.67%,与SVM和RBF神经网络对比,从而进一步证明了该方法的有效性,说明该方法能够满足实际工况下的故障诊断要求。     

A novel approach for recognition of control chart patterns: Type-2 fuzzy clustering optimized support vector machine

Unnatural patterns in the control charts can be associated with a specific set of assignable causes for process variation. Hence, pattern recognition is very useful in identifying the process problems. In this study, a multiclass SVM (SVM) based classifier is proposed because of the promising generalization capability of support vector machines. In the proposed method type-2 fuzzy c-means (T2FCM) clustering algorithm is used to make a SVM system more effective. The fuzzy support vector machine classifier suggested in this paper is composed of three main sub-networks: fuzzy classifier sub-network, SVM sub-network and optimization sub-network. In SVM training, the hyper-parameters plays a very important role in its recognition accuracy. Therefore, cuckoo optimization algorithm (COA) is proposed for selecting appropriate parameters of the classifier. Simulation results showed that the proposed system has very high recognition accuracy.     

基于DRGAN和支持向量机的合成孔径雷达图像目标识别

为解决SAR图像目标识别中样本缺乏和方位角敏感问题,提出了一种基于DRGAN和SVM的SAR图像目标识别算法。首先,采用多尺度分形特征对SAR图像进行增强,经过分割得到目标二值图像,基于Hu二阶矩估计目标的方位角。然后对估计得到的目标方位角进行量化编码,结合原始图像作为输入,对设计的DRGAN模型参数进行训练与优化。由于DRGAN中的深度生成模型将目标姿态与外观表示进行解耦设计,故可利用该模型将SAR图像样本变换到同一方位角区间。基于变换后的训练样本分别提取归一化灰度特征,利用SVM训练分类器。采用MSTAR数据集在多个不同操作条件下对提出的算法进行测试,实验结果表明,在带变体的标准操作条件下,能够达到97.97%的分类精度,优于部分基于CNN模型的分类精度,在4种扩展操作条件下的分类精度分别为97.83%,91.77%,97.11%和97.04%,均优于传统方法的分类精度。在SAR图像目标方位角估计存在一定误差的情况下,训练得到的GAN模型作为SAR图像目标旋转估计器,能够使得在不进行复杂样本预处理的前提下,仍然取得较高的SAR图像目标识别精度。     

人脸特征选择中的SVM泛化误差估计

根据统计学习理论,特征选择可以通过有效的特征搜索策略最小化某个预测泛化误差及其它相关性能来实现。本文研究通过递归特征排除法（Recursive Feature Elimination,RFE）最小化SVM VC留一法(Leave-One-Out, LOO)误差或支持向量span误差估计选择优化特征子集问题,并将最小化VC LOO误差或支持向量span误差估计作为Wrapper特征选择模型的选择判据。人脸识别实质是稀疏超高维空间、典型的小样本模式识别问题。解决这类问题的关键在于如何获得对分类有意义的特征。将特征选择与分类器设计结合,理论上优于传统的特征提取或特征选择方法。为此,本论文将WT和KPCA作为过滤模型（Filter）,最小化SVM泛化误差估计作为封装模型（Wrapper）,结合这两种模型的优势提出人脸特征选择及识别的新框架。并在UMIST人脸数据库上进行了相应的实验,结果显示提出的特征选择方法和特征搜索策略及人脸特征选择构架有效可行。     

BP neural network classification on passenger vehicle type based on GA of feature selection

This paper has concluded six features that belong to passenger vehicle types based on genetic algorithm(GA)of feature selection.We have obtained an optimal feature subset,including length,ratio of width and length,and ratio of height and length.And then we apply this optimal feature subset as well as another feature set,containing length,width and height,to the network input.Back-propagation(BP)neural network and support vector machine(SVM)are applied to classify the passenger vehicle type.There are four passenger vehicle types.This paper selects 400 samples of passenger vehicles,among which 320 samples are used as training set(each class has 80 samples)and the other 80 samples as testing set,taking the feature of the samples as network input and taking four passenger vehicle types as output.For the test,we have applied BP neural network to choose the optimal feature subset as network input,and the results show that the total classification accuracy rate can reach 96%,and the classification accuracy rate of first type can reach 100%.In this condition,we obtain a conclusion that this algorithm is better than the traditional ones[9].     

改进的网格搜索算法在SVM故障诊断中的应用

针对网格搜索法在支持向量机参数寻优时存在复杂度高、运算量大等不足,提出了一种改进的网格搜索SVM分类器的最佳参数选择算法,并将其应用于田纳西-伊士曼过程。实验表明,与改进前的算法相比,改进的网格搜索SVM分类器参数选择算法可以有效地减少SVM分类器的运算量、改进学习性能并提高识别率。     

基于CS-BBO优化SVM的汽轮机转子故障诊断

为了提高汽轮机转子故障诊断的准确率和识别效率,提出了一种基于混沌的生物地理学优化算法(biogeography-based optimization with chaos,简称CS-BBO)和支持向量机(support vector machine,简称SVM)相结合的故障诊断方法。首先,将混沌理论引入到生物地理学优化算法(biogeography-based optimization,简称BBO)中,得到CS-BBO算法;其次,通过CS-BBO算法优化SVM得到诊断模型的最优参数,增强SVM的学习能力和泛化能力;最后,通过ZT-3转子试验台模拟汽轮机转子故障,利用得到的4种状态下的试验数据验证优化模型的有效性。结果表明:CS-BBO算法优化SVM的模型可以准确、高效地对汽轮机转子进行故障诊断;与BBO算法优化SVM模型相比,该方法的故障诊断准确率和识别效率更高。     

基于特征选择的支持向量机在故障诊断中的应用

机械设备因为其本身结构的复杂性,故障很难简单地进行诊断,所以智能诊断成为一个热点的研究方向。以前的工作中多是通过神经网络甚至支持向量机等方法进行诊断,本文提出了基于支持向量机集成的特征选择算法,通过该算法可以有效去除故障数据集中所提取的不相关特征,并在新的更少特征的数据集上进行建模。在实际某柴油机故障数据上的计算表明:在通过特征选择后的数据集上利用支持向量机集成的方法建模可以得到比不进行选择更好的结果,也得到了比单个支持向量机建模更好的结果。     

A new approach for rule extraction of expert system based on SVM

Based on the SVM’s excellent generalization performance, a new approach is proposed to extract knowledge rules from Support Vector Clustering (SVC). In this method, the first step is to choose the features of the sample data by using Genetic Algorithm for improving the comprehensibility of the knowledge rules. Then the SVC algorithm is adopted to obtain the Clustering Distribution Matrix of the sample data whose features have been chosen. Finally, hyper-rectangle rules are constructed using the Clustering Distribution Matrix. To make the rules more concise, and easier to explain, hyper-rectangle rules are simplified further by using rules combinations, dimension reduction and interval extension. In addition, the SMOTE (Synthetic Minority Over-sampling Technique) algorithm is adopted to resample fault samples in order to solve the serious imbalance problem of samples. The UCI datasets are used to validate the new method proposed in this paper, the results compared with other rules extraction methods show that the new approach is more effective. The new method is used to extract knowledge rules for aero-engine oil monitoring expert system, and the results show that the new method can effectively extract knowledge rules for expert system, and break through the bottleneck in expert system knowledge dynamic acquisition.