Robust bearing degradation assessment method based on improved CVA

The SVDD (support vector data description) is one of the most well-known one-class support vector learning methods, in which one tries the strategy of utilizing balls defined on the feature space in order to distinguish a set of normal data from all other possible abnormal objects. The usual strategy of the SVDD depends on the process of finding the region for the normal-class training data with somewhat neglecting the distribution of the abnormal data, thus it may not work well if applied to the binary classification problems in which two classes have similar number of data. In this paper, we consider the problem of performing binary classification based on the SVDD techniques, and in order to overcome the possible drawback of the usual SVDD strategy focusing on the normal-class data only, we propose a new SVDD algorithm which is based on the use of two different SVDD balls for the positive and negative classes along with the SVDD projection and nearest neighbor rule. To investigate how the proposed method works, we compared the performance of the proposed method with SVC (support vector classifier) and conventional SVDD using several real datasets.

In this paper an online identification and control strategy is developed for chemical processes. The procedure used for system identification is canonical variate analysis, a method that operates on input-output data sequences batch-wise to provide an optimized state-space model; the resulting models are then used in a multivariable predictive control procedure to yield excellent overall closed-loop results. The resulting adaptive control procedure is demonstrated to have excellent properties for control of a continuous stirred tank reactor with inverse dynamics, delays and nonlinearities.

Abstract: Multivariate control charts are essential tools in multivariate statistical process control (MSPC). “Shewhart-type” charts are control charts using rational subgroupings which are effective in the detection of large shifts. Recently, the one-class classification problem has attracted a lot of interest. Three methods are typically used to solve this type of classification problem. These methods include the k−center method, the nearest neighbor method, one-class support vector machine (OCSVM), and the support vector data description (SVDD). In industrial applications, like statistical process control (SPC), practitioners successfully used SVDD to detect anomalies or outliers in the process. In this paper, we reformulate the standard support vector data description and derive a least squares version of the method. This least-squares support vector data description (LS-SVDD) is used to design a control chart for monitoring the mean vector of processes. We compare the performance of the LS-SVDD chart with the SVDD and T2 chart using out-of-control Average Run Length (ARL) as the performance metric. The experimental results indicate that the proposed control chart has very good performance.

Pneumatic control valves are the most frequently used actuators in industrial processes. Its property definitely affects the performance of processes and therefore process monitoring of the pneumatic control valve is of great importance. Canonical Variate Analysis (CVA) is a multivariate data-driven method which considers time correlations and has been demonstrated to be superior to some methods in certain benchmark case studies. Inspired by the successful application of the CVA technique in process industry, we firstly monitor the pneumatic control valve using CVA and demonstrate its effectiveness and practicability on the DAMADICS benchmark.

This paper considers an incremental support vector learning for the abnormality detection problems. One of the most well-known support vector learning methods for abnormality detection is the so-called SVDD(support vector data description), which seeks the strategy of utilizing balls defined on the kernel feature space in order to distinguish a set of normal data from all other possible abnormal objects. The major concern of this paper is to modify the SVDD into the direction of utilizing the relation between the optimal solution and incrementally given training data. After a thorough review about the original SVDD method, this paper establishes an incremental method for finding the optimal solution based on certain observations on the Lagrange dual problems. The applicability of the presented incremental method is illustrated via a design example.

Information entropy based sample reduction for support vector data description

The SVDD (support vector data description) is one of the most well-known one-class support vector learning methods, in which one tries the strategy of utilizing balls defined on the feature space in order to distinguish a set of normal data from all other possible abnormal objects. The major concern of this paper is to extend the main idea of the SVDD for the problem of pattern de-noising. Combining the projection onto the spherical decision boundary resulting from the SVDD together with a solver for the pre-image problem, we propose a new method for pattern de-noising. In the proposed method, we first solve the SVDD for the training data, then for each noisy test pattern, perform de-noising by projecting its feature vector onto the decision boundary on the feature space, and finally find the location of the de-noised pattern by obtaining the pre-image of the projection. The applicability of the proposed method is illustrated via an example dealing with noisy handwritten digits.

To improve the efficiency and usability of adaptive anomaly detection system, we propose a new framework based on Support Vector Data Description (SVDD) method. This framework includes two main techniques: online change detection and unsupervised anomaly detection. The first one enables automatically obtain model training data by measuring and distinguishing change caused by intensive attacks from normal behavior change and then filtering most intensive attacks. The second retrains model periodically and detects the forthcoming data. Results of experiments with the KDD’99 network data show that these techniques can handle intensive attacks effectively and adapt to the concept drift while still detecting attacks. As a result, false positive rate is reduced from 13.43% to 4.45%.

Generation-dependent patterns of variation and population differentiation in <i>Abies amabilis</i> and <i>A</i>. <i>lasiocarpa</i> (Pinaceae) from north-coastal British Columbia

Chapter 5 - Analysis of several groups: Canonical variate analysis

Genetic variability and diversity of genotypes are very important for all living organisms. Knowledge of the genetic diversity is a potential tool for pre-breeding parental selection. The present experiment was conducted at two locations (Isfahan, Khuzestan) under field conditions during the 2017–2018 growing season, with fifteen short day onion genotypes which were evaluated by multivariate methods. Nine quantitative traits were studied. MANOVA showed that the locations, varieties and location × variety interaction were significantly different for all nine traits. Significant positive correlation observed for two locations for yield and single weight (0.85 in Khuzestan and 0.61 in Isfahan), yield and bulb height (0.52 in Khuzestan and 0.55 in Isfahan), bulb height and index shape (0.68 in Khuzestan and 0.70 in Isfahan) and bulb diameter and single weight (0.81 in Khuzestan and 0.66 in Isfahan). Further, yield was significantly correlated with dry matter: positively in Isfahan (0.62), and negatively in Khuzestan (–0.54). In Khuzestan, the first two canonical variants explained 79.19% of the total variation between the varieties; however, the greatest variation was found for the Saba and Behbahan improved population. The first two canonical variables explained 86.76% of the total variation between the varieties in Isfahan. Saba and Behbahan improved population varieties were the smallest, while Paliz and Early Super Select were the largest. The Saba and Behbahan improved population, as the most diverse genotypes, were recommended for further inclusion in future crop improvement programs.

A compound fault diagnosis model for photovoltaic array based on 1D VoVNet-SVDD by considering unknown faults

Support Vector Data Description (SVDD) has a limitation for dealing with a large data set in which computational load drastically increases as training data size becomes large. To handle this problem, we propose a new fast SVDD method using K-means clustering method. Our method uses divide-and-conquerstrategy; trains each decomposed sub-problems to get support vectors and retrains with the support vectors to find a global data description of a whole target class. The proposed method has a similar result to the original SVDD and reduces computational cost. Through experiments, we show efficiency of our method.

Linear and non-linear pattern recognition models for classification of fruit from visible–near infrared spectra

Isolated small theropod teeth are abundant in vertebrate microfossil assemblages, and are frequently used in studies of species diversity in ancient ecosystems. However, determining the taxonomic affinities of these teeth is problematic due to an absence of associated diagnostic skeletal material. Species such as Dromaeosaurus albertensis, Richardoestesia gilmorei, and Saurornitholestes langstoni are known from skeletal remains that have been recovered exclusively from the Dinosaur Park Formation (Campanian). It is therefore likely that teeth from different formations widely disparate in age or geographic position are not referable to these species. Tooth taxa without any associated skeletal material, such as Paronychodon lacustris and Richardoestesia isosceles, have also been identified from multiple localities of disparate ages throughout the Late Cretaceous. To address this problem, a dataset of measurements of 1183 small theropod teeth (the most specimen-rich theropod tooth dataset ever constructed) from North America ranging in age from Santonian through Maastrichtian were analyzed using multivariate statistical methods: canonical variate analysis, pairwise discriminant function analysis, and multivariate analysis of variance. The results indicate that teeth referred to the same taxon from different formations are often quantitatively distinct. In contrast, isolated teeth found in time equivalent formations are not quantitatively distinguishable from each other. These results support the hypothesis that small theropod taxa, like other dinosaurs in the Late Cretaceous, tend to be exclusive to discrete host formations. The methods outlined have great potential for future studies of isolated teeth worldwide, and may be the most useful non-destructive technique known of extracting the most data possible from isolated and fragmentary specimens. The ability to accurately assess species diversity and turnover through time based on isolated teeth will help illuminate patterns of evolution and extinction in these groups and potentially others in greater detail than has previously been thought possible without more complete skeletal material.