Crack Fault Classification for Planetary Gearbox Based on Feature Selection Technique and K-means Clustering Method

During the condition monitoring of a planetary gearbox, features are extracted from raw data for a fault diagnosis.However, different features have different sensitivity for identifying different fault types, and thus, the selection of a sensitive feature subset from an entire feature set and retaining as much of the class discriminatory information as possible has a directly effect on the accuracy of the classification results. In this paper, an improved hybrid feature selection technique(IHFST) that combines a distance evaluation technique(DET), Pearson's correlation analysis, and an ad hoc technique is proposed. In IHFST, a temporary feature subset without irrelevant features is first selected according to the distance evaluation criterion of DET, and the Pearson's correlation analysis and ad hoc technique are then employed to find and remove redundant features in the temporary feature subset, respectively, and hence,a sensitive feature subset without irrelevant or redundant features is selected from the entire feature set. Further, the k-means clustering method is applied to classify the different kinds of health conditions. The effectiveness of the proposed method was validated through several experiments carried out on a planetary gearbox with incipient cracks seeded in the tooth root of the sun gear, planet gear, and ring gear. The results show that the proposed method can successfully distinguish the different health conditions of a planetary gearbox, and achieves a better classification performance than other methods. This study proposes a sensitive feature subset selection method that achieves an obvious improvement in terms of the accuracy of the fault classification.     

A two-step feature selection method for monitoring tool wear and its application to the coroning process

Coroning is a complex and multi-directional gear finishing process involving metal removal of gear teeth surface, and condition monitoring has not been applied to this process. In order to capture the progress of wear, an acoustic emission (AE) sensor is used, but the large data size of AE requires extensive dimension reduction and feature selection. The conventional method of averaging to reduce the data size may have the risk of losing information as higher frequencies are filtered off. A two-step feature selection method is implemented using class mean scatter criterion and modified relevance/redundancy analysis. This method results in feature dimension reduction and enhances classification performance. It involves first ranking candidate features by calculating their separability. Features which are correlated are then combined to reduce dimensions without averaging. Application of this two-step feature selection technique enables coroning tool wear to be monitored with a classification rate of 98.3 % compared to 94.1 % using conventional feature selection.     

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。     

Design of multisensor fusion-based tool condition monitoring system in end milling

Recent advancement in signal processing and information technology has resulted in the use of multiple sensors for the effective monitoring of tool conditions, which is the most crucial feedback information to the process controller. Interestingly, the abundance of data collected from multiple sensors allows us to employ various techniques such as feature extraction, selection, and classification methods for generating such crucial information. While the use of multiple sensors has improved the accuracy in the classification of tool conditions, design of tool condition monitoring system (TCM) for reduced complexity and increased robustness has been rarely studied. Therefore, this paper studies the design of effective multisensor-based TCM when machining 4340 steel by using a multilayer-coated and multiflute carbide end mill cutter. Multiple sensors tested in this paper include force, vibration, acoustic emission, and spindle power sensor for the time and frequency domain data. In addition, two feature selection methods and three classifiers with a machine ensemble technique are considered as design components. Importantly, different fusion methods are evaluated in this paper: (1) decision level fusion and (2) feature level fusion. The experimental results show that the design of TCM based on the feature level fusion can significantly improve the accuracy of the tool condition classification. It is also shown that the highest accuracy can be achieved by using force, vibration, and acoustic emission sensor together with correlation-based feature selection method and majority voting machine ensemble.     

Diagnosis of artificially created surface damage levels of planet gear teeth using ordinal ranking

Effective diagnosis of damage levels is important for condition based preventive maintenance of gearboxes. One special characteristic of damage levels is the inherent ordinal information among different levels. Retaining the ordinal information is therefore important for diagnosing damage levels. Classification, a machine learning technique, has been widely adopted for automated diagnosis of gear faults. However, classification cannot keep the ordinal information because the damage levels are treated as nominal variables. This paper employs ordinal ranking, another machine learning technique, to preserve the ordinal information in automated diagnosis of damage levels. As to ordinal ranking, feature selection is important. However, most existing feature selection methods are for classification, which are not suitable for ordinal ranking. This paper designs a feature selection method for ordinal ranking based on correlation coefficients. A diagnosis approach based on ordinal ranking and the proposed feature selection method is then introduced. This method is tested on diagnosis of artificially created surface damage levels of planet gear teeth in a planetary gearbox. Experimental results show the effectiveness of the proposed diagnosis approach. The advantages of using ordinal ranking for diagnosing gear damage levels are also demonstrated.     

An ensemble classification of exudates in color fundus images using an evolutionary algorithm based optimal features selection

Atomic recognition of the Exudates (EXs), the major symbol of diabetic retinopathy is essential for automated retinal images analysis. In this article, we proposed a novel machine learning technique for early detection and classification of EXs in color fundus images. The major challenge observed in the classification technique is the selection of optimal features to reduce computational time and space complexity and to provide a high degree of classification accuracy. To address these challenges, this article proposed an evolutionary algorithm based solution for optimal feature selection, which accelerates the classification process and reduces computational complexity. Similarly, three well‐known classifiers that is, Naïve Bayes classifier, Support Vector Machine, and Artificial Neural Network are used for the classification of EXs. Moreover, an ensemble‐based classifier is used for the selection of best classifier on the basis of majority voting technique. Experiments are performed on three well‐known benchmark datasets and a real dataset developed at local Hospital. It has been observed that the proposed technique achieved an accuracy of 98% in the detection and classification of EXs in color fundus images.     

NEW FEATURE SELECTION METHOD IN MACHINE FAULT DIAGNOSIS

Aiming to deficiency of the filter and wrapper feature selection methods, a new method based on composite method of filter and wrapper method is proposed. First the method filters original features to form a feature subset which can meet classification correctness rate, then applies wrapper feature selection method select optimal feature subset. A successful technique for solving optimization problems is given by genetic algorithm (GA). GA is applied to the problem of optimal feature selection. The composite method saves computing time several times of the wrapper method with holding the classification accuracy in data simulation and experiment on bearing fault feature selection. So this method possesses excellent optimization property, can save more selection time, and has the characteristics of high accuracy and high efficiency.     

Fault diagnosis of rotating machinery based on multiple ANFIS combination with GAs

This paper presents a novel method for fault diagnosis based on empirical mode decomposition (EMD), an improved distance evaluation technique and the combination of multiple adaptive neuro-fuzzy inference systems (ANFISs). The method consists of three stages. First, prior to feature extraction, some preprocessing techniques, like filtration, demodulation and EMD are performed on vibration signals to acquire more fault characteristic information. Then, six feature sets, including time- and frequency-domain statistical features of both the raw and preprocessed signals, are extracted. Second, an improved distance evaluation technique is proposed, and with it, six salient feature sets are selected from the six original feature sets, respectively. Finally, the six salient feature sets are input into the multiple ANFIS combination with genetic algorithms (GAs) to identify different abnormal cases. The proposed method is applied to the fault diagnosis of rolling element bearings, and testing results show that the multiple ANFIS combination can reliably recognise different fault categories and severities, which has a better classification performance compared to the individual classifiers based on ANFIS. Moreover, the effectiveness of the proposed feature selection method based on the improved distance evaluation technique is also demonstrated by the testing results.     

Fault classification on vibration data with wavelet based feature selection scheme

Fault classification based upon vibration measurements is an essential building block of a conditional based health usage monitoring system. Multiple sensors are incorporated to assure the redundancy and to achieve the desired reliability and accuracy. The shortcoming of using multiple sensors is the need to deal with a high dimensional feature set, a computationally expensive task in classification. It is vital to reduce the feature dimension via an effective feature extraction and feature selection algorithm. A simple wavelet based feature selection scheme is proposed herein, uniquely built by local discriminant bases and genetic optimization. This scheme overcomes the disadvantages faced by the existing feature selection methods by producing a generic feature set, reducing the dimensionality of features, and requiring no prior information of the problem domain. The proposed feature selection scheme is based upon the strategy of "divide and conquer" that significantly reduce the computation time without compromising the classification performance. The simulation results show the proposed feature selection scheme provides at least 65% reduction of the total number of features at no cost of the classification accuracy.     

基于改进特征选择方法的齿轮故障SVM诊断

为提高齿轮故障诊断的精度,对常用的共享特征选择方法(Share feature selection,SFS)进行改进,提出了改进的特征选择方法(Improved feature selection,IFS)。改进的特征选择方法结合齿轮两两故障类型之间的特点,在齿轮两两故障之间建立独立的故障特征集,用以取代所有故障类型的共享特征集;而后,通过建立多个二分类的支持向量机,对独立的故障特征集进行识别,得到诊断结果。齿轮故障诊断实例表明,改进的特征选择方法排除了无用特征的干扰,提高了诊断精度,具有一定的优势。     

A COMPARATIVE STUDY OF FEATURE SELECTION FOR HIDDEN MARKOV MODEL-BASED MICRO-MILLING TOOL WEAR MONITORING

This paper presents a discriminant feature selection approach for hidden Markov model (HMM) modeling of micro-milling tool conditions. The approach is compared with other popular feature selection methods such as principal component analysis (PCA) and automatic relevance determination (ARD) according to their HMM classification rate. In tool condition monitoring (TCM), there are a lot of features that contain redundant information or less sensitive to tool state discrimination. These features are expected to be deleted for less computation and more robust modeling of tool conditions. Fisher's linear discriminant analysis (FDA) is modified for this purpose. The FDA is generally used for classification, and the features are mapped to another space and lose their physical meanings. In the modified discriminant feature selection, the features are selected in the original feature space by maximizing tool state separation and ranked by their separation ability between different tool states. Experimental results from both micro-milling of copper and steel under different working conditions indicate that the FDA is superior to both PCA and ARD for feature selection in HMM's classification. The reasons behind these differences are also discussed.     

基于改进的快速独立分量分析与支持向量机的ECG分类诊断

提出了一种改进快速独立分量分析与支持向量机相结合的新型心电图分类方法.利用埃特金加速法对快速独立分量分析算法的核心迭代过程进行改造,得到改进的快速独立分量分析算法,减少了迭代次数,提高了算法的收敛速度.新方法运用改进的快速独立分量分析算法提取心电图数据的特征向量,并通过支持向量机实现心电图信号的分类.对取自MIT/BH数据库的7种不同心脏状况的心电图数据进行实验,结果表明该方法整体识别率达到98.8%,改进的快速独立分量分析算法所需迭代时间比现有的快速独立分量分析算法减少48%.     

Optimization of multiple responses using principal component analysis and technique for order preference by similarity to ideal solution

Optimizing multi-response problems has become an increasingly relevant issue when more than one correlated product quality characteristic must be assessed simultaneously in a complicated manufacturing process. This study proposes a novel optimization procedure for multiple responses based on Taguchi’s parameter design. The signal-to-noise (SN) ratio is initially used to assess the performance of each response. Principal component analysis (PCA) is then conducted on the SN values to obtain a set of uncorrelated components. The optimization direction for each component is determined based on the corresponding variation mode chart. Finally, the relative closeness to the ideal solution resulting from the technique for order preference by similarity to ideal solution (TOPSIS) is determined as an overall performance index (OPI) for multiple responses. Engineers can easily employ the proposed procedure to obtain the optimal factor/level combination for multiple responses. A case study involving optimization of the chemical-mechanical polishing of copper (Cu-CMP) thin films from an integrated circuit manufacturer in Taiwan is also presented to demonstrate the effectiveness of the proposed procedure.     

基于SOM网络的特征选择技术研究

讨论了一种SOM网络训练结果的可视化技巧,结合该技巧提出了基于SOM网络的特征选择方法。该方法 通过计算出SOM网络竞争层神经元权值中各维特征对输入模式聚类识别的影响,可以选择出对于模式识别敏感 的特征集。用IRIS和齿轮故障数据对该方法进行了检验,研究结果表明,采用该方法能较好地从原始特征中选择 出有效特征子集,实现不同类别输入数据之间的模式聚类识别。     

基于KCCA虚假邻点判别的非线性变量选择

特征变量选择技术是非线性系统建模过程中降低信息冗余和提高精度的有效方法。提出一种结合核典型相关法(kernel canonical correlation analysis,KCCA)与虚假最近邻法的变量选择法。首先引入核方法,将非线性原始数据映射到线性空间,再采用典型相关法有效合理地消除因子之间的多重共线性,受混沌相空间虚假最近邻点法的启示,通过计算原始数据在KCCA子空间中投影的距离,判断其对主导变量的解释能力,由此进行变量的选择。该方法用氢氰酸生产工艺工程中的非线性模型验证,并与全参数模型进行比较,结果显示该方法有良好的变量选择能力。因此,该研究为非线性系统建模的变量选择方法提供了一种新方法。     

舰用蒸汽透平级和级组的多目标规划决策（MODM）分析

舰用蒸汽透平的设计实践表明,由于设计问题的复杂性,仅用单目标优化方法求得的解不能完全适合设计需要。本文讨论了在舰用蒸汽透平级和级组设计中引入多目标规划决策（MODM）的必要性,介绍了用于单级和/或多级透平优化设计的两种计算机计算模型,讨论了决策变量、目标函数的选取,搜索过程中约束条件的某些假定,以及优化搜索过程。叶栅损失由三种不同的方法确定。计算机程序在DISCDVER Y-1600微机上通过。本文给出了单级和多级级组多目标决策分析一些结果,分析了级数对多级透平效率的影响。计算结果表明,所提出的方法是有效的。     

Evaluation of methods for reduction of surface texture features

The diagnosis of worn and damaged surfaces is an important issue in machine failure analysis and condition monitoring. Of many approaches used, image classification based on feature parameters has often proven to be particularly useful. However, large image databases can be computationally costly to analyse, and the datasets are susceptible to noise. Hence, it is essential to determine which feature parameters hold the most useful information, in order to improve the classification rate and computation time. This paper presents a performance evaluation of dimension reduction techniques currently used in pattern recognition. A comparison of three methods is conducted, in order to determine which is able to produce the best results over a large range of image datasets. The methods analysed are: Non-Linear Fisher, Principal Component Analysis and kernel Principal Component Analysis. These are then tested against four different classifiers to obtain the best combination. These classifiers are: Linear Discriminant Classifier, Quadratic Discriminant Classifier, k-Nearest Neighbour and Support Vector Machine Classifiers. For further analysis, two combined dimension reduction and classification methods are tested: Minimum Classification Error and reduced feature space Support Vectors. For the comparison, four datasets of images with different scales and rotations are used, i.e. Brodatz textures, artificially generated isotropic fractal images and Talysurf images of sandblasted and abraded steel surfaces. The results showed that a combination of the Non-Linear Fisher dimension reduction technique and a Linear Support Vector Machine Classifier gave the best performance overall and are the most promising for the application in automated machine condition monitoring and expert free failure analysis. Further improvement can be achieved by performing a step-wise dimension reduction by first reducing the features using the Principal Component Analysis method, then further reduction with the Non-Linear Fisher technique.     

Automated system for characterization and classification of malaria‐infected stages using light microscopic images of thin blood smears

In this paper, we propose a comprehensive image characterization cum classification framework for malaria‐infected stage detection using microscopic images of thin blood smears. The methodology mainly includes microscopic imaging of Leishman stained blood slides, noise reduction and illumination correction, erythrocyte segmentation, feature selection followed by machine classification. Amongst three‐image segmentation algorithms (namely, rule‐based, Chan–Vese‐based and marker‐controlled watershed methods), marker‐controlled watershed technique provides better boundary detection of erythrocytes specially in overlapping situations. Microscopic features at intensity, texture and morphology levels are extracted to discriminate infected and noninfected erythrocytes. In order to achieve subgroup of potential features, feature selection techniques, namely, F‐statistic and information gain criteria are considered here for ranking. Finally, five different classifiers, namely, Naive Bayes, multilayer perceptron neural network, logistic regression, classification and regression tree (CART), RBF neural network have been trained and tested by 888 erythrocytes (infected and noninfected) for each features’ subset. Performance evaluation of the proposed methodology shows that multilayer perceptron network provides higher accuracy for malaria‐infected erythrocytes recognition and infected stage classification. Results show that top 90 features ranked by F‐statistic (specificity: 98.64%, sensitivity: 100%, PPV: 99.73% and overall accuracy: 96.84%) and top 60 features ranked by information gain provides better results (specificity: 97.29%, sensitivity: 100%, PPV: 99.46% and overall accuracy: 96.73%) for malaria‐infected stage classification.     

Design and control of flexure based XYϑZ stage

This article presents the design and control of an ultraprecision XYϑ_Z stage with nanometer accuracy. The stage has a plane mechanism and symmetric hexagonal structure which consists of a monolithic flexure hinge mechanism with three piezoelectric actuators and six flexures preserving the plane motion. The symmetric design reduces the effect of temperature gradient on the structure. Because the relationship between design variables and system parameters are quite complicated and there are some trade-offs among them, it is very difficult to set design variables manually and optimal design procedure is used. The objective of design is maximizing the 1^st resonant frequency to improve the dynamic characteristics. The reason is that the stage must move with heavy load of about 20 kg. The higher resonant frequency also makes the stage stiffer and stronger against the dynamic force and moment. This paper describes the procedures of selecting parameters for the optimal design and a mathematical formulation for the optimization problem. The stage was designed to attain ±10 um in the X- and Y-direction and ±90arcsec in the yaw direction at the same time and have the 1^st resonant frequencies of 455.5 Hz in X- and Y-direction and 275.3 Hz for yaw direction without load. The stage was fabricated according to the optimal design results and experimental results indicate that the design procedure is effective. A conventional PI control results are presented for ultraprecision motion.