Nonlinear Principal Component Analysis Research Articles

Novelty detection and dimension reduction via guided ultrasonic waves: Damage monitoring of scarf repairs in composite laminates

This work focuses on structural health monitoring aspects of composite adhesively bonded repairs, evaluating their performance with guided ultrasonic waves. These repairs have shown remarkable potential in addressing repairability demands in new composite aircraft. More specifically, the behavior of a scarf repair under axial tensile loading was monitored with guided ultrasonic waves. The signal post-processing techniques focused on the extraction of the appropriate features, on the application of the pattern recognition and dimension reduction algorithms and on their subsequent correlation with the damage. A principal component analysis was employed that operated as a benchmark for the proposal of a more advanced data reduction method, the nonlinear principal component analysis. Appropriate damage indices were extracted and the results were correlated with images taken through a digital image correlation technique. The correlation of the extracted features with the early stage damage was performed and conclusions about the recovered strength through the scarf repair were deduced. The study focused on the selection of appropriate signal features and on their subsequent investigation through an outlier analysis. The limits of the applied outlier analysis were interpreted through principal component analysis and optimized through the concept of principal curves as derived through the nonlinear principal component analysis.

Spatiotemporal Pattern Recognition and Nonlinear PCA for Global Horizontal Irradiance Forecasting

This letter presents a novel technique for the forecast of the ground horizontal irradiance (GHI) from satellite-based images. To enhance the forecast accuracy, spatial information in addition to temporal information has been considered. This produced an increase in the computational load of the forecast process. Dimensionality reduction techniques based on nonlinear principal component analysis (PCA) are used to project the original data set into low-dimension feature space. A multilayer feedforward neural network classifier is used to model the signal through a training operation involving past history of the considered spatiotemporal signal. Experiments have been carried out on two different data sets. Comparisons with classical forecasting techniques demonstrate that the introduction of the spatial information permits to obtain better short-term forecast measurements for all types of sky conditions. Moreover, further analysis demonstrates that, compared with linear PCA, the nonlinear PCA is more appropriate for dimensionality reduction of spatiotemporal GHI data set.

Long-term behavior of groundwater chemistry in a periodically rewetted fen area covered with macrophytes

We studied the long-term changes in groundwater composition in the context of peat restoration at a degraded water-table managed peatland site typical for many agriculturally used fen areas in the northern hemisphere. At the study site, peatland rewetting with groundwater control and pumped canal water was carried out in two periods: from 1997 to 2002 and from 2011 to 2013. The site was not managed between 2002 and 2011, which led to an unstable groundwater table that had declined in part. The aim of this study was to investigate the consequences of rewetting and desiccation on groundwater chemistry. We pursued a multivariate approach using nonlinear principal component analysis (Isomap) to identify the prevailing processes that control the groundwater quality in this system.Sixteen years after peatland restoration, the groundwater quality had significantly improved. Principal component analysis revealed that hydrological processes had a major impact on groundwater quality, i.e. fluctuations between upwelling of local, salt-influenced groundwater and downwelling of surface and rainwater (first principal component) as well as upwelling of regional groundwater from deeper layers (second principal component) which originated from the catchment. In particular, the upwelling of regional deep groundwater had a strong positive impact on the groundwater quality of upper layers at the Biesenbrow site. Another major impact on groundwater quality was nutrient withdrawal by macrophytes and incorporation into organic matter. In the upper groundwater layer, peat mineralization processes resulted in substantially increased SO4 concentrations.We concluded that potential matter release after rewetting is buffered by hydrological barriers, and seems to be marginal with little impact on adjacent environments in the long term. The ecosystem is sustainably stabilized, and therefore has no negative impact on groundwater quality during periods of water shortage. Due to the strong influence of regional groundwater, management measures in the catchment are very important for maintaining and improving groundwater quality in peatlands.

Exploring the relationship between entheseal changes and physical activity: a multivariate study.

Analyses of entheseal changes (EC) in identified skeletal samples employ a common research strategy based on the comparison between occupations grouped on the basis of shared biomechanical and/or social characteristics. Results from this approach are often ambiguous, with some studies that point to differences in EC between occupational samples and others failing to provide evidence of behavioral effects on EC. Here we investigate patterns of EC among documented occupations by means of a multivariate analysis of robusticity scores in nine postcranial entheses from a large (N = 372) contemporary skeletal sample including specimens from one Italian and two Portuguese identified collections. Data on entheseal robusticity, analyzed by pooled sides as well by separated sides and levels of asymmetry, are converted in binary scores and then analyzed through nonlinear principal component analysis and hierarchical cluster analysis. Results of these analyses are then used for the classification of occupations. Differences between occupational classes are tested by MANOVA and pairwise Hotelling's test. Results evidence three classes which separate occupations related to farming, physically demanding but generalized occupation, and physically undemanding occupations, with the more consistent differences between the first and the last classes. Our results are consistent with differences in biomechanical behavior between the occupations included in each class, and point to the physical and social specificity of farming activities. On the other hand, our study exemplifies the usefulness of alternative analytical protocols for the investigation of EC, and the value of research designs devoid of a priori assumptions for the test of biocultural hypotheses.

Comprehensive Evaluation of Thermal Power Units Based on Nonlinear Principal Component Analysis

In order to improve the operation and management level of thermal power units, and to achieve the goal of saving energy and reducing consumption, the nonlinear principal component analysis (PCA) was proposed to assess the operating condition of fossil-fired power plants comprehensively, which can resolve the problem on the evaluation of multiple indicators. This method improves the traditional PCA by eliminating the relativity. Moreover, this paper applies the nonlinear PCA to evaluate the operating state of power plants, which can reduce the dimensions of the indicators and abstract the principal components. It can provide the high reference value for the power generation group on how to implement saving energy, optimal operation, indicators evaluation and competition in thermal power units. The practical example shows that the proposed method is effective.

Nonlinear Fuzzy Robust PCA Algorithm for Pain Decision Support System

This paper describes particular pain events to be located as in infant face images with feature extraction algorithm. Nonlinear Fuzzy Robust PCA (NFRPCA) feature extraction is implemented to test its effectiveness in recognizing pain in images. In this work, two classifiers, Fuzzy k-nearest neighbors (Fuzzy k-NN) and k-nearest neighbors (k-NN) are employed. Result shows that the NFRPCA and classifier (Fuzzy k-NN and k-NN) can be used for the recognition of infant pain images with the best accuracy of 89.77%.

NEURAL NETWORK APPROACHES FOR INTRUSION DETECTION AND RECOGNITION

Most current Intrusion Detection Systems (IDS) examine all data features to detect intrusion. Also existing intrusion detection approaches have some limitations, namely impossibility to process large number of audit data for real-time operation, low detection and recognition accuracy. To overcome these limitations, we apply modular neural network models to detect and recognize attacks in computer networks. It is based on combination of principal component analysis (PCA) neural networks and multilayer perceptrons (MLP). PCA networks are employed for important data extraction and to reduce high dimensional data vectors. We present two PCA neural networks for feature extraction: linear PCA (LPCA) and nonlinear PCA (NPCA). MLP is employed to detect and recognize attacks using feature-extracted data instead of original data. The proposed approaches are tested using KDD-99 dataset. The experimental results demonstrate that the designed models are promising in terms of accuracy and computational time for real world intrusion detection.

Modelling the quality of work in the Italian social co-operatives combining NPCA-RSM and SEM-GME approaches

The objective of this paper is to describe and analyse with appropriate statistical models the links between work quality latent factors. Due to the complexity of the task, the analysis is carried out through a two-step approach:In the first step, we construct some multidimensional measures of the subjective quality of work, using nonlinear principal component analysis (NPCA) and Rasch analysis with the Rating Scale Model (NPCA-RSM);In the second step, we adopt a Structural Equation Model based on generalized maximum entropy (SEM-GME) to integrate the measures achieved with the previous step and to evaluate the relationships between the subjective work quality latent factors.Therefore, the novel aspects of this paper are the following: (i) The integration between the NPCA-RSM and SEM-GME, which allows reduction of the variables analysed and evaluation of the measurement errors; (ii) The formalization of a Job Satisfaction Model for the study of the relationships between the subjective work quality latent factors in the Italian social services sector.

Reconstruction based approach to sensor fault diagnosis using auto-associative neural networks

Fault diagnostics is an important research area including different techniques. Principal component analysis (PCA) is a linear technique which has been widely used. For nonlinear processes, however, the nonlinear principal component analysis (NLPCA) should be applied. In this work, NLPCA based on auto-associative neural network (AANN) was applied to model a chemical process using historical data. First, the residuals generated by the AANN were used for fault detection and then a reconstruction based approach called enhanced AANN (E-AANN) was presented to isolate and reconstruct the faulty sensor simultaneously. The proposed method was implemented on a continuous stirred tank heater (CSTH) and used to detect and isolate two types of faults (drift and offset) for a sensor. The results show that the proposed method can detect, isolate and reconstruct the occurred fault properly.

Classification of fricative consonants for speech enhancement in hearing devices.

ObjectiveTo investigate a set of acoustic features and classification methods for the classification of three groups of fricative consonants differing in place of articulation.MethodA support vector machine (SVM) algorithm was used to classify the fricatives extracted from the TIMIT database in quiet and also in speech babble noise at various signal-to-noise ratios (SNRs). Spectral features including four spectral moments, peak, slope, Mel-frequency cepstral coefficients (MFCC), Gammatone filters outputs, and magnitudes of fast Fourier Transform (FFT) spectrum were used for the classification. The analysis frame was restricted to only 8 msec. In addition, commonly-used linear and nonlinear principal component analysis dimensionality reduction techniques that project a high-dimensional feature vector onto a lower dimensional space were examined.ResultsWith 13 MFCC coefficients, 14 or 24 Gammatone filter outputs, classification performance was greater than or equal to 85% in quiet and at +10 dB SNR. Using 14 Gammatone filter outputs above 1 kHz, classification accuracy remained high (greater than 80%) for a wide range of SNRs from +20 to +5 dB SNR.ConclusionsHigh levels of classification accuracy for fricative consonants in quiet and in noise could be achieved using only spectral features extracted from a short time window. Results of this work have a direct impact on the development of speech enhancement algorithms for hearing devices.

The fountain of age: a remarkable 3D shape that portrays health and functional differences among the European elderly.

There are very few norms to evaluate and monitor the health and functioning of the elderly. This paper proposes a compact spatial representation of 25 health measurements of European citizens older than 50 years. Data from 44,285 unique individuals were obtained from the EU-wide Survey of Health, Ageing and Retirement in Europe(SHARE) data collected in 2004–2007 and were analyzed by homogeneity analysis, a form of non-linear principal components analysis. The resulting configuration of persons shows a remarkable three-dimensional shape that resembles a fountain. The three components explain 13.7, 5.8 and 4.6 percent of the total variation, respectively. Component 1 is driven by age and by the disabilities that come with old age. Component 2 portrays differences in health that are independent of age, with the high scores in relatively good health, given age. Component 3 distinguishes specific types of functional decline from general complaints that impact on daily life. The shape suggests that the elderly keep on maturing as they grow older, actually becoming more diverse as a group. We show how the solution may be used to develop and support profiles for the elderly. Another potential application is to track the individual development of the elderly, thereby objectifying personalized medicine.

Developing anti-tobacco messages for Australian Aboriginal and Torres Strait Islander peoples: evidence from a national cross-sectional survey

BackgroundSmoking rates in Australian Aboriginal and Torres Strait Islander peoples remain high, with limited impact of government measures for many subgroups. The aim of this cross-sectional study was to investigate differences in organisational practice for developing anti-tobacco messages for these target populations.MethodsTelephone interviews were conducted with 47 organisation representatives using a structured questionnaire based on health communication and health promotion frameworks. Responses were coded into phases of message development, message types (educational, threat, positive or advocacy), target groups, message recommendations, and evaluations undertaken. Cultural sensitivity for message development was divided into surface structure (use of images, language, demographics) and deep structure (use of socio-cultural values). A categorical principal component analysis explored the key dimensions of the findings and their component relationships.ResultsAmong organisations interviewed, a community-orientated, bottom-up approach for developing anti-tobacco messages was reported by 47% (n = 24); 55% based message development on a theoretical framework; 87% used a positive benefit appeal; 38% used threat messages. More Aboriginal Medical Services (AMSs) targeted youth (p < 0.005) and advised smokers to quit (p < 0.05) than other types of organisations. AMSs were significantly more likely to report using deep structure in tailoring messages compared with non-government (p < 0.05) and government organisations (p < 0.05). Organisations that were oriented to the general population were more likely to evaluate their programs (p < 0.05). A two-dimensional non-linear principal component analysis extracted components interpreted as “cultural understanding” (bottom-up, community-based approaches, deep structures) and “rigour” (theoretical frameworks, and planned/completed evaluations), and accounted for 53% of the variability in the data.ConclusionMessage features, associated with successful campaigns in other populations, are starting to be used for Aboriginal and Torres Strait Islander peoples. A model is proposed to facilitate the development of targeted anti-tobacco messages for Aboriginal and Torres Strait Islander peoples. Organisations could consider incorporating both components of cultural understanding-rigour to enable the growth of evidence-based practice.

Probabilistic PCA of censored data: accounting for uncertainties in the visualization of high-throughput single-cell qPCR data

Motivation: High-throughput single-cell quantitative real-time polymerase chain reaction (qPCR) is a promising technique allowing for new insights in complex cellular processes. However, the PCR reaction can be detected only up to a certain detection limit, whereas failed reactions could be due to low or absent expression, and the true expression level is unknown. Because this censoring can occur for high proportions of the data, it is one of the main challenges when dealing with single-cell qPCR data. Principal component analysis (PCA) is an important tool for visualizing the structure of high-dimensional data as well as for identifying subpopulations of cells. However, to date it is not clear how to perform a PCA of censored data. We present a probabilistic approach that accounts for the censoring and evaluate it for two typical datasets containing single-cell qPCR data.Results: We use the Gaussian process latent variable model framework to account for censoring by introducing an appropriate noise model and allowing a different kernel for each dimension. We evaluate this new approach for two typical qPCR datasets (of mouse embryonic stem cells and blood stem/progenitor cells, respectively) by performing linear and non-linear probabilistic PCA. Taking the censoring into account results in a 2D representation of the data, which better reflects its known structure: in both datasets, our new approach results in a better separation of known cell types and is able to reveal subpopulations in one dataset that could not be resolved using standard PCA.Availability and implementation: The implementation was based on the existing Gaussian process latent variable model toolbox (https://github.com/SheffieldML/GPmat); extensions for noise models and kernels accounting for censoring are available at http://icb.helmholtz-muenchen.de/censgplvm.Contact: fbuettner.phys@gmail.comSupplementary information: Supplementary data are available at Bioinformatics online.

Prediction-Focused Subsurface Modeling: Investigating the Need for Accuracy in Flow-Based Inverse Modeling

The objective of most formulations of inverse modeling in the Earth Sci- ences is to estimate the model parameters given the observation data. In this paper, an additional element to these formulations is considered, namely, the prediction for which the models are built. An example of such modeling is the prediction of solute transport using geological models of the subsurface constrained to existing geophys- ical, flow dynamic and solute observations. The paper then illustrates and addresses a fundamental question relating data, model and prediction: does model inversion reduce uncertainty in prediction variables given the observed data? To investigate this question, a diagnostic tool is proposed to assess whether matching the observed data significantly reduces uncertainty in the prediction variables. In addition, for some cases, a quick estimate of uncertainty can be obtained without applying inverse mod- eling. It relies on a dimensionality reduction method using non-linear principal com- ponent analysis (NLPCA) calibrated from evaluating the prediction and data response variables on a few prior Earth models. The proposed diagnostic tool is applied on a simple example of tracer flow and, for the cases investigated, the NLPCA provided an accurate diagnostic and a satisfactory pre-estimation of uncertainty in the prediction variables.

A nonlinear principal component analysis approach for turbulent combustion composition space

An approach for the determination of principal components using nonlinear principal component analysis (NLPCA) is proposed in the context of turbulent combustion. NLPCA addresses complex data sets where the contours of the inherent principal directions are curved in the original manifold. Thermo-chemical scalars' statistics are reconstructed from the optimally derived moments. The tabulation of the scalars is then implemented, using artificial neural networks (ANN). The approach is implemented on numerical data generated for the stand-alone one-dimensional turbulence (ODT) simulation of hydrogen autoignition in a turbulent jet with preheated air. It is found that 2 nonlinear principal components are sufficient to capture thermo-chemical scalars' profiles. For some of the scalars, a single principal component reasonably captures the scalars' profiles as well.

Partial Discharge Pattern Recognition of Cast Resin Current Transformers Using Radial Basis Function Neural Network

This paper proposes a novel pattern recognition approach based on the radial basis function (RBF) neural network for identifying insulation defects of high-voltage electrical apparatus arising from partial discharge (PD). Pattern recognition of PD is used for identifying defects causing the PD, such as internal discharge, external discharge, corona, etc. This information is vital for estimating the harmfulness of the discharge in the insulation. Since an insulation defect, such as one resulting from PD, would have a corresponding particular pattern, pattern recognition of PD is significant means to discriminate insulation conditions of high-voltage electrical apparatus. To verify the proposed approach, experiments were conducted to demonstrate the field-test PD pattern recognition of cast resin current transformer (CRCT) models. These tests used artificial defects created in order to produce the common PD activities of CRCTs by using feature vectors of field-test PD patterns. The significant features are extracted by using nonlinear principal component analysis (NLPCA) method. The experimental data are found to be in close agreement with the recognized data. The test results show that the proposed approach is efficient and reliable.

Online classification for spalling detection and vibratory behavior monitoring

Vibration analysis is the most widely used tool in industrial application for machine’s health condition assessment. Bearings, however, are very sensitive and require special attention. Vibration analysis employs various signal processing methods such as spectral analysis, time-scale, time frequency analysis, etc. these methods are used to analyze bearings’ vibratory behavior by monitoring the evolution of statistical indicators.However, diagnosing the bearing depending on traditional features only isn’t sufficient to assure effective or reliable assessment of the component’ health condition. This paper proposes a multi-features online dynamic classification as a new method for fault detection and health condition monitoring for bearings; this technique uses multiple features, including traditional features extracted from the raw signal, two special features extracted by wavelet analysis, the spectral kurtosis, coupled with a nonlinear principal component analysis and a dynamic classification to capitalize on the hidden information in the time evolution of the features.Through this article, we introduce different measures and techniques used to characterize the health state of rolling, then we deploy a methodology using dynamic classification to detect early defect. To ensure an almost continuous surveillance, this methodology is based on a real-time analysis, and uses specific statistical indicators adapted to the experimental bench. Then, the monitoring of the degradation is achieved through the resulting class of the state of degradation. New parameters such as the speed of the class, the position of the class, the shape of the class will be discussed to inform the state of damage. The suggested methodology is validated by analyzing several fatigue tests from a fatigue bench bearing thrust ball referenced SNR51207.

Nonlinear principal component based fuzzy clustering: A case study of lentil genotypes

Cluster analysis is frequently used by the plant breeders in grouping germplasm collections into a few homogeneous groups in order to identify accessions with specific property of potential relevance for their plant improvement programs. The set of descriptors for the germplasm accessions consists of both numerical and categorical descriptors. In such situations, the standard principal component analysis will not be appropriate for feature extraction of data using all descriptors because it deals with only numeric variables. In this paper, nonlinear principal component analysis was used to analyse the descriptors of lentil accessions which can handle mixture of measurement types. The first two nonlinear principal components were used as input to fuzzy c-means algorithm in grouping 518 lentil genotypes into four clusters based on their agronomic and morphological traits. The study demonstrated that the proposed nonlinear principal component based fuzzy clustering has a promising potential in agriculture as a tool for evaluation and efficient grouping of germplasm collections.

Fault diagnosis based on graphical tools for multi-energy processes

To guarantee the safe operation of systems, it is necessary to use systematic techniques to detect and isolate faults for the purpose of diagnosis. The Fault Detection and Isolation (FDI) of nonlinear systems with coupling multiple energies became a difficult task. This why, we propose in this paper, the exploitation of the behavioral and structural properties of graphs (Bond Graph and Signed Directed Graph) combining with a Principal Component Analysis method (PCA). Therein, a coupled Bond Graph model is used for modeling methodology. A Signed Directed Graph (SDG) is then deduced. Fault detection is later carried out by graph coloring. The localization of the actual fault is performed based on a nonlinear PCA (NLPCA) and back/forward propagations on the SDG.The proposed approach is tested on the thermofluid case study and some simulations are provided.

Long-term transient thermal analysis using compact models for data center applications

Reduced-order thermal models are necessary to enable real-time assessment of the optimum operating and control conditions to improve data center energy efficiency. A 3D computational fluid dynamics (CFD) and heat transfer model of a basic raised floor, air cooled, hot aisle/cold aisle data center configuration was developed and simulation results were used to generate snapshots for initializing compact models based on the Proper Orthogonal Decomposition (POD) and the Nonlinear Principle Component Analysis (NLPCA) methods. The specific focus of the study was to numerically investigate how the thermal trends can be affected by the long-term transient flow patterns associated with leakage and how the compact models (e.g. POD and NLPCA) can be utilized for much faster implementations and characterizations of the transient flows. Using both the POD and the NLPCA method, good agreement was achieved between the full simulation results and the compact models for predicting the dynamically developed local flow structure over a range of transient cooling air supply operating conditions. In addition, the NLPCA method was implemented to better characterize the nonlinear aspects of the CFD results. The benefits of using both the POD and NLPCA methods are discussed in relation to constructing compact models as real-time predictive tools. A systematic use of the compact models has also been proposed, to enable more robust thermal management and control of data centers.