An adaptive signal classification procedure. Application to aircraft engine condition monitoring

A method (C) for adaptive signal classification is described, and closely related to actual diagnostic problems in monitored aircraft engines for which an input-output model is available. This new method is based on the prediction in parallel of the output residual and of discriminant functions, thus yielding a predictive state classification into overall degradation classes. Updating is achieved by monitoring the actual output and requesting an a posteriori classification from a “teacher”. The main applications are in the field of degradation analysis, and early warning of overall degradation in monitored control systems. (C) has been implemented in the engine maintenance department of an airline.     

An automotive generator fault diagnosis system using discrete wavelet transform and artificial neural network

This paper describes a fault diagnosis system for automotive generators using discrete wavelet transform (DWT) and an artificial neural network. Conventional fault indications of automotive generators generally use an indicator to inform the driver when the charging system is malfunction. But this charge indicator tells only if the generator is normal or in a fault condition. In the present study, an automotive generator fault diagnosis system is developed and proposed for fault classification of different fault conditions. The proposed system consists of feature extraction using discrete wavelet analysis to reduce complexity of the feature vectors together with classification using the artificial neural network technique. In the output signal classification, both the back-propagation neural network (BPNN) and generalized regression neural network (GRNN) are used to classify and compare the synthetic fault types in an experimental engine platform. The experimental results indicate that the proposed fault diagnosis is effective and can be used for automotive generators of various engine operating conditions.     

Feature extraction and health status prediction in PV systems

Diagnosis aims at predicting the health status of components and systems. In photovoltaic systems, it is vital to guarantee energy production and extend the useful life of photovoltaic power plants. Multiple prediction and classification algorithms have been proposed for this purpose in the literature. The accuracy of these algorithms depends directly on the quality of the data and the features with which they are tuned or trained. In this paper, an innovative approach for predicting the health status of photovoltaic systems is proposed, which includes a feature selection stage. This approach first discriminates severely affected photovoltaic panels using basic electrical features. In a second step, it discriminates the other faulty panels using more elaborated time–frequency features and selecting the most relevant features through correlation and variance analysis. Finally, the approach predicts the health status of photovoltaic panels using a nonlinear regression method named partial least squares. This later is then combined with linear discriminant analysis and compared. The approach is validated with real current data from a photovoltaic plant composed of twelve photovoltaic panels with power between 205 and 240 Wp in three health states, namely broken glass, healthy, and big snail trails. The results obtained show that the proposed approach efficiently predicts the three health states. It determines the level of degradation of the panels, which indicates priorities to corrective and predictive maintenance actions. Furthermore, it is cost-effective since it uses only electrical measurements that are already available in standard photovoltaic data acquisition systems. Above all, the approach is generic and it can be easily extrapolated to other diagnosis problems in other domains.     

Classification and Feature Extraction by Simplexization

Techniques for classification and feature extraction are often intertwined. In this paper, we contribute to these two aspects via the shared philosophy of simplexizing the sample set. For general classification, we present a new criteria based on the concept of -nearest-neighbor simplex (), which is constructed by the nearest neighbors, to determine the class label of a new datum. For feature extraction, we develop a novel subspace learning algorithm, called discriminant simplex analysis (DSA), in which the intraclass compactness and interclass separability are both measured by distances. Comprehensive experiments on face recognition and lipreading validate the effectiveness of the DSA as well as the -based classification approach.     

Adaptive linear discriminant regression classification for face recognition

Linear discriminant regression classification (LDRC) was presented recently in order to boost the effectiveness of linear regression classification (LRC). LDRC aims to find a subspace for LRC where LRC can achieve a high discrimination for classification. As a discriminant analysis algorithm, however, LDRC considers an equal importance of each training sample and ignores the different contributions of these samples to learn the discriminative feature subspace for classification. Motivated by the fact that some training samples are more effectual in learning the low-dimensional feature space than other samples, in this paper, we propose an adaptive linear discriminant regression classification (ALDRC) algorithm by taking special consideration of different contributions of the training samples. Specifically, ALDRC makes use of different weights to characterize the different contributions of the training samples and utilizes such weighting information to calculate the between-class and the within-class reconstruction errors, and then ALDRC seeks to find an optimal projection matrix that can maximize the ratio of the between-class reconstruction error over the within-class reconstruction error. Extensive experiments carried out on the AR, FERET and ORL face databases demonstrate the effectiveness of the proposed method.     

An integrated strategy for efficient business plan and maintenance plan for systems with a dynamic failure distribution

The purpose of this study is to propose an integrated strategy to determine jointly efficient business and maintenance plans. The studied system is subject to random failures with a dynamic failure law. It must perform a set of missions (among M possible missions) over a finite planning horizon. Each mission may have different characteristics that depend on operational and environmental conditions. The determination of a business plan consists in choosing and scheduling the missions to be performed. To maximize the net profit (profits generated by the achievement of missions minus maintenance costs), two meta-heuristics based on genetic algorithms are developed. The first genetic algorithm is used to determine the business plan and the second one generates an efficient maintenance plan. Two maintenance policies are studied: a minimalist policy which involves only corrective maintenance actions and another policy, called sequential, which involves several imperfect preventive maintenance activities performed at predetermined times. Two cases are studied for the latter strategy. The first one considers the maintenance effectiveness factor as being the same for all preventive maintenance actions and we search for the best factor. In the second case, we consider maintenance actions with different efficiency factors and we look for the optimal value of each factor. Finally, a numerical example illustrates the proposed approach and the difference between the maintenance policies.     

Optimal subset-division based discrimination and its kernelization for face and palmprint recognition

Discriminant analysis is effective in extracting discriminative features and reducing dimensionality. In this paper, we propose an optimal subset-division based discrimination (OSDD) approach to enhance the classification performance of discriminant analysis technique. OSDD first divides the sample set into several subsets by using an improved stability criterion and K-means algorithm. We separately calculate the optimal discriminant vectors from each subset. Then we construct the projection transformation by combining the discriminant vectors derived from all subsets. Furthermore, we provide a nonlinear extension of OSDD, that is, the optimal subset-division based kernel discrimination (OSKD) approach. It employs the kernel K-means algorithm to divide the sample set in the kernel space and obtains the nonlinear projection transformation. The proposed approaches are applied to face and palmprint recognition, and are examined using the AR and FERET face databases and the PolyU palmprint database. The experimental results demonstrate that the proposed approaches outperform several related linear and nonlinear discriminant analysis methods.     

High temperature dynamic viscosity sensor for engine oil applications

This paper presents a new high temperature dynamic viscosity sensor for in situ condition monitoring of engine lubricants. The sensor is used to measure the variation in the quality factor of a vibrating piezoelectric cantilever beam due to viscous damping. The sensor was used to measure the dynamic viscosity of various single and multi-grade engines oils up to 180 cP from 25 °C to 60 °C. The sensor is capable of detecting degradation and dilution of engine oil for both new and used samples of 5W-30 and 10W-40 and diluted SAE 30 engine oils. All of the viscosity measurements presented are within 0.13-9.8% of the results obtained using the standard Walther equation at various temperatures. An equation relating dynamic viscosity of an oil sample to the quality factor of the beam is presented. The quality factor measurement circuit presented in this research can be implemented in automotive applications for in situ condition monitoring of lubricant viscosity.     

基于观测器的发动机转矩跟踪模型预测控制

针对如何实现发动机转矩快速精准地跟踪期望转矩的问题,提出一种基于观测器的模型预测控制策略.首先,利用均值模型对汽油发动机的进气歧管压力动态、转矩和转速动态进行建模,考虑到发动机真实转矩不可测的情况,采用Lyapunov稳定性理论和可测转速信号设计观测器对进气歧管压力进行在线估计,进而获得发动机的实时估计转矩;然后,利用基于观测器的模型预测控制算法设计转矩跟踪控制器,通过C/GMRES数值优化算法在线求解滚动时域优化问题,实现转矩的实时跟踪控制;最后,利用汽油发动机实验台进行实验验证以表明所提出算法的有效性.     

Unsupervised real-time constrained linear discriminant analysis to hyperspectral image classification

We have proposed a constrained linear discriminant analysis (CLDA) approach for classifying the remotely sensed hyperspectral images. Its basic idea is to design an optimal linear transformation operator which can maximize the ratio of inter-class to intra-class distance while satisfying the constraint that the different class centers after transformation are aligned along different directions. Its major advantage over the traditional Fisher's linear discriminant analysis is that the classification can be achieved simultaneously with the transformation. The CLDA is a supervised approach, i.e., the class spectral signatures need to be known a priori. But, in practice, these informations may be difficult or even impossible to obtain. So in this paper we will extend the CLDA algorithm into an unsupervised version, where the class spectral signatures are to be directly generated from an unknown image scene. Computer simulation is used to evaluate how well the algorithm performs in terms of finding the pure signatures. We will also discuss how to implement the unsupervised CLDA algorithm in real-time for resolving the critical situations when the immediate data analysis results are required.     

Comparing statistical and machine learning classifiers: alternatives for predictive modeling in human factors research

Multivariate classification models play an increasingly important role in human factors research. In the past, these models have been based primarily on discriminant analysis and logistic regression. Models developed from machine learning research offer the human factors professional a viable alternative to these traditional statistical classification methods. To illustrate this point, two machine learning approaches--genetic programming and decision tree induction--were used to construct classification models designed to predict whether or not a student truck driver would pass his or her commercial driver license (CDL) examination. The models were developed and validated using the curriculum scores and CDL exam performances of 37 student truck drivers who had completed a 320-hr driver training course. Results indicated that the machine learning classification models were superior to discriminant analysis and logistic regression in terms of predictive accuracy. Actual or potential applications of this research include the creation of models that more accurately predict human performance outcomes.     

基于Spark的油藏数据挖掘与分析

为了方便油藏数据特征的分析和石油的勘探开发过程,本文利用Spark并行计算框架分析油藏数据,并通过数据挖掘算法分析油藏属性之间的潜在关系,对油藏的不同层段进行了分类和预测.本文的主要工作包括：搭建Spark分布式集群和数据处理、分析平台,Spark是流行的大数据并行计算框架,相对传统的一些分析方法和工具,可以实现快速、准确的数据挖掘任务;根据油藏数据的特点建立多维异常检测函数,并新增渗孔比判别属性Pr;在处理不平衡数据时,针对逻辑回归分类提出交叉召回训练模型,并优化代价函数,针对决策树,提出KR-SMOTE对小类别样本进行过采样扩充,这两种方法都可以有效处理数据不平衡问题,提高分类精度.     

Distributed joint dynamic maintenance and production scheduling in manufacturing systems: Framework based on model predictive control and Benders decomposition

Scheduling the maintenance based on the condition, respectively the degradation level of the system leads to improved system's reliability while minimizing the maintenance cost. Since the degradation level changes dynamically during the system's operation, we face a dynamic maintenance scheduling problem. In this paper, we address the dynamic maintenance scheduling of manufacturing systems based on their degradation level. The manufacturing system consists of several units with a defined capacity and an individual dynamic degradation model, seeking to optimize their reward. The units sell their production capacity, while maintaining the systems based on the degradation state to prevent failures. The manufacturing units are jointly responsible for fulfilling the demand of the system. This induces a coupling constraint among the agents. Hence, we face a large-scale mixed-integer dynamic maintenance scheduling problem. In order to handle the dynamic model of the system and large-scale optimization, we propose a distributed algorithm using model predictive control (MPC) and Benders decomposition method. In the proposed algorithm, first, the master problem obtains the maintenance scheduling for all the agents, and then based on this data, the agents obtain their optimal production using the distributed MPC method which employs the dual decomposition approach to tackle the coupling constraints among the agents. The effectiveness of the proposed method is investigated on two case studies.     

From projection pursuit and CART to adaptive discriminant analysis?

While many efforts have been put into the development of nonlinear approximation theory and its applications to signal and image compression, encoding and denoising, there seems to be very few theoretical developments of adaptive discriminant representations in the area of feature extraction, selection and signal classification. In this paper, we try to advocate the idea that such developments and efforts are worthwhile, based on the theoretical study of a data-driven discriminant analysis method on a simple-yet instructive-example. We consider the problem of classifying a signal drawn from a mixture of two classes, using its projections onto low-dimensional subspaces. Unlike the linear discriminant analysis (LDA) strategy, which selects subspaces that do not depend on the observed signal, we consider an adaptive sequential selection of projections, in the spirit of nonlinear approximation and classification and regression trees (CART): at each step, the subspace is enlarged in a direction that maximizes the mutual information with the unknown class. We derive explicit characterizations of this adaptive discriminant analysis (ADA) strategy in two situations. When the two classes are Gaussian with the same covariance matrix but different means, the adaptive subspaces are actually nonadaptive and can be computed with an algorithm similar to orthonormal matching pursuit. When the classes are centered Gaussians with different covariances, the adaptive subspaces are spanned by eigen-vectors of an operator given by the covariance matrices (just as could be predicted by regular LDA), however we prove that the order of observation of the components along these eigen-vectors actually depends on the observed signal. Numerical experiments on synthetic data illustrate how data-dependent features can be used to outperform LDA on a classification task, and we discuss how our results could be applied in practice.     

A Hierarchical Fuzzy Clustering-based System to Create User Profiles

The rapid development of the World Wide Web as a medium of commerce and information dissemination has generated a growing interest of web portal managers in systems able to identify user profiles from the web access logs. The interpretation of these profiles can help re-organize the web portal, e.g., by restructuring the site’s content more efficiently, or even to build adaptive web portals, i.e., portals whose organization and presentation change depending on the specific visitor’s needs. In this paper, we assume that the pages of the web portal have been prearranged in a number of different categories. We introduce a systematic approach to determine a hierarchy of user profiles from the history of users’ accesses to the categories. First, we filter the access log by removing both occasional users and categories of poor interest. Then, we apply an Unsupervised Fuzzy Divisive Hierarchical Clustering (UFDHC) algorithm to cluster the users of the web portal into a hierarchy of fuzzy groups characterized by a set of common interests and each represented by a prototype, which defines the profile of the group typical member. To identify the profile a specific user belongs to, we propose a novel classification method which completely exploits the information contained in the hierarchy. To prove the effectiveness of our approach, we apply the UFDHC algorithm to access log data collected over a period of 15 days and use the classification method to associate a profile with the users defined by access log data collected during subsequent 60 days. Finally, we highlight the good characteristics of our system by comparing our results with the ones obtained by applying a profiling system based on a modified version of the fuzzy C-means.     

Regularized query classification using search click information

Hundreds of millions of users each day submit queries to the Web search engine. The user queries are typically very short which makes query understanding a challenging problem. In this paper, we propose a novel approach for query representation and classification. By submitting the query to a web search engine, the query can be represented as a set of terms found on the web pages returned by search engine. In this way, each query can be considered as a point in high-dimensional space and standard classification algorithms such as regression can be applied. However, traditional regression is too flexible in situations with large numbers of highly correlated predictor variables. It may suffer from the overfitting problem. By using search click information, the semantic relationship between queries can be incorporated into the learning system as a regularizer. Specifically, from all the functions which minimize the empirical loss on the labeled queries, we select the one which best preserves the semantic relationship between queries. We present experimental evidence suggesting that the regularized regression algorithm is able to use search click information effectively for query classification.     

A survey of models of degradation for control applications

The analysis of equipment degradation has traditionally developed in two main directions. One approach, of great interest for control system design, has been to consider that degradation causes fundamental changes to the behaviour of a system. Another approach, used in optimal maintenance planning and production scheduling, considers degradation as a separate process that affects performance but does not necessarily change the behaviour. This article provides a review of mathematical models of degradation that will facilitate the integration of degradation modelling into control and optimisation schemes. To this end, a new unified classification is proposed. It takes into account the influence of degradation on the behaviour of the system, as well as the factors influencing degradation. Understanding these mutual influences will enable improved optimization, design and operation of control systems. The flexibility of the proposed classification is demonstrated in an industrial application to a multi-product batch scheduling process.     

仿真技术在汽车发动机故障检测中的研究

针对目前汽车发动机的传感器易损坏而导致发动机状态分析的结果产生重大偏差的特点,对人工神经BP网络模型做了改进,使其具有很强的自适应能力而能使网络的收敛方向和速度得到优化,并编制了相应的程序.作为实例,文章对某一实际发动机进行了仿真试验,结果表明该改进的BP网络具有很强的自适应能力,所有的误差控制在3%以内,可以满足工程实际的需要.由于人工神经网络在实际应用中不涉及具体的物理模型,因此该模型对发动机的状态参数在线仿真、减少传感器的维护量,特别是对发动机故障诊断技术水平的提高有很大的意义.     

Feature scaling for kernel fisher discriminant analysis using leave-one-out cross validation

Kernel fisher discriminant analysis (KFD) is a successful approach to classification. It is well known that the key challenge in KFD lies in the selection of free parameters such as kernel parameters and regularization parameters. Here we focus on the feature-scaling kernel where each feature individually associates with a scaling factor. A novel algorithm, named FS-KFD, is developed to tune the scaling factors and regularization parameters for the feature-scaling kernel. The proposed algorithm is based on optimizing the smooth leave-one-out error via a gradient-descent method and has been demonstrated to be computationally feasible. FS-KFD is motivated by the following two fundamental facts: the leave-one-out error of KFD can be expressed in closed form and the step function can be approximated by a sigmoid function. Empirical comparisons on artificial and benchmark data sets suggest that FS-KFD improves KFD in terms of classification accuracy.     

基于时点分割的核Fisher判别分析–顺序回归机多类分类建模方法

针对多维时间序列的多类分类问题,本文提出基于时点分割思想的核Fisher判别分析-顺序回归机(KFDA-ORM)多类分类建模方法.该方法利用核Fisher判别分析(KFDA)与顺序回归机(ORM)的互补性得到分类决策函数;对分类样本的多维时间序列进行时点分割处理,使用决策函数得到各时点的分类级别;通过指数平滑分析得到采样周期内样本的最终分类结果.通过实例验证,该方法对多维时间序列的分类具有较好效果,是一种有效的多类分类方法.