Standard Stacking Research Articles

Progressive failure analysis of laminates with an open hole subjected to compressive loading (OHC) using the enhanced semi-discrete modeling framework

The Open-hole compressive (OHC) strength of a fiber reinforced laminate is one of the most critical allowables for design of aerostructures. In this paper, results for predicting the OHC strength using the semi-discrete damage modeling framework are presented. The predictions are seen to capture experimentally observed failure mechanisms and measured failure loads. The constitutive model includes local axial compressive failure (and subsequent load bearing at a reduced plateau stress) while maintaining numerical robustness. Standard stacking sequences (quasi-isotropic, “hard” and “soft”) have been analyzed and compared to publicly available databases. Additionally, the model is challenged to predict delamination-dominated failure due to ply-scaling. Overall, the model is able to capture relevant failure mechanisms, while the ultimate loads are predicted with good accuracy. Therefore, this framework can be used with confidence in predicting OHC strengths of laminates.

A Novel Stacked Ensemble for Hate Speech Recognition

Detecting harmful content or hate speech on social media is a significant challenge due to the high throughput and large volume of content production on these platforms. Identifying hate speech in a timely manner is crucial in preventing its dissemination. We propose a novel stacked ensemble approach for detecting hate speech in English tweets. The proposed architecture employs an ensemble of three classifiers, namely support vector machine (SVM), logistic regression (LR), and XGBoost classifier (XGB), trained using word2vec and universal encoding features. The meta classifier, LR, combines the outputs of the three base classifiers and the features employed by the base classifiers to produce the final output. It is shown that the proposed architecture improves the performance of the widely used single classifiers as well as the standard stacking and classifier ensemble using majority voting. We also present results on the use of various combinations of machine learning classifiers as base classifiers. The experimental results from the proposed architecture indicated an improvement in the performance on all four datasets compared with the standard stacking, base classifiers, and majority voting. Furthermore, on three of these datasets, the proposed architecture outperformed all state-of-the-art systems.

Multiwindow weighted stacking of surface-wave dispersion

Surface-wave methods are classically used to characterize shear (S-) wave velocities ([Formula: see text]) of the shallow subsurface through the inversion of dispersion curves. When targeting 2D shallow structures with sharp lateral heterogeneity, windowing and stacking techniques can be implemented to provide a better description of [Formula: see text] lateral variations. These techniques, however, suffer from the trade-off between lateral resolution and depth of investigation (DOI), which is well-known when using the multichannel analysis of surface waves (MASW) method. We have adopted a novel methodology aimed at enhancing lateral resolution and DOI of MASW results through the use of multiwindow weighted stacking of surface waves (MW-WSSW). MW-WSSW consists of stacking dispersion images obtained from data segments of different sizes, with a wavelength-based weight that depends on the aperture of the data selection window. In that sense, MW-WSSW provides additional weight to short wavelengths in smaller windows so as to better inform shallow parts of the subsurface, and vice versa for deeper velocities. Using multiple windows improves the DOI, whereas applying wavelength-based weights enhances the shallow lateral resolution. MW-WSSW was implemented within the open-source package SWIP and applied to the processing of synthetic and real data sets. In both cases, we compared it with standard windowing and stacking procedures that are already implemented in SWIP. MW-WSSW provided convincing results with optimized lateral extent, improved shallow resolution, and increased DOI.

Deep ensemble learning for Alzheimer's disease classification.

Ensemble learning uses multiple algorithms to obtain better predictive performance than any single one of its constituent algorithms could. With the growing popularity of deep learning technologies, researchers have started to ensemble these technologies for various purposes. Few, if any, however, have used the deep learning approach as a means to ensemble Alzheimer's disease classification algorithms. This paper presents a deep ensemble learning framework that aims to harness deep learning algorithms to integrate multisource data and tap the 'wisdom of experts'. At the voting layer, two sparse autoencoders are trained for feature learning to reduce the correlation of attributes and diversify the base classifiers ultimately. At the stacking layer, a nonlinear feature-weighted method based on a deep belief network is proposed to rank the base classifiers, which may violate the conditional independence. The neural network is used as a meta classifier. At the optimizing layer, over-sampling and threshold-moving are used to cope with the cost-sensitive problem. Optimized predictions are obtained based on an ensemble of probabilistic predictions by similarity calculation. The proposed deep ensemble learning framework is used for Alzheimer's disease classification. Experiments with the clinical dataset from National Alzheimer's Coordinating Center demonstrate that the classification accuracy of our proposed framework is 4% better than six well-known ensemble approaches, including the standard stacking algorithm as well. Adequate coverage of more accurate diagnostic services can be provided by utilizing the wisdom of averaged physicians. This paper points out a new way to boost the primary care of Alzheimer's disease from the view of machine learning.

Classifiers ensemble and synthetic minority oversampling techniques for academic performance prediction

The increasing need for data driven decision making recently has resulted in the application of data mining in various fields including the educational sector which is referred to as educational data mining. The need for improving the performance of data mining models has also been identified as a gap for future researcher. In Nigeria, higher educational institutions collect various students’ data, but these data are rarely used in any decision or policy making to improve the academic performance of students. This research work, attempts to improve the performance of data mining models for predicting students’ academic performance using stacking classifiers ensemble and synthetic minority over-sampling techniques. The research was conducted by adopting and evaluating the performance of J48, IBK and SMO classifiers. The individual classifiers models, standard stacking classifier ensemble model and stacking classifiers ensemble model were trained and tested on 206 students’ data set from the faculty of science federal university Dutse. Students’ specific previous academic performance records at Unified Tertiary Matriculation Examination, Senior Secondary Certificate Examination and first year Cumulative Grade Point Average of students are used as data inputs in WEKA 3.9.1 data mining tool to predict students’ graduation classes of degrees at undergraduate level. The result shows that application of synthetic minority over-sampling technique for class balancing improves all the various models performance with the proposed modified stacking classifiers ensemble model outperforming the various classifiers models in both performance accuracy and RSME values making it the best model.<strong></strong>

Comprehensive comparison of different structures of passive permanent magnet bearings

Here, different feasible configurations of passive permanent magnet bearings (PMBs) are presented and compared with each other. In these configurations, stiffness is improved by adding iron cores and air intervals between magnets. Bearings with opposite magnetisation stacking (standard) and rotating magnetisation stacking (Halbach) are studied here. A two-dimensional planner model is used to calculate the magnetic field and stiffness. The optimisation will be performed for obtaining maximum stiffness per magnet volume ratio. The results show that standard stacking with radially magnetised PMs and having back iron and air intervals between magnets is the most optimum structure. For larger dimensions of magnets, other structures will be useful too.

Virtual multiphase flow metering using diverse neural network ensemble and adaptive simulated annealing

Real-time production monitoring in oil and gas industry has become very significant particularly as fields become economically marginal and reservoirs deplete. Virtual flow meters (VFMs) are intelligent systems that infer multiphase flow rates from ancillary measurements and are attractive and cost-effective solutions to meet monitoring demands, reduce operational costs, and improve oil recovery efficiency. Current VFMs are very challenging to develop and very expensive to maintain, most of which were developed for wells with dedicated physical meters where there exists an abundance of well test data. This study proposes a VFM system based on ensemble learning for fields with common metering infrastructure where data generated is very limited. The proposed method generates diverse neural network (NN) learners by manipulating training data, NN architecture and learning trajectory. Adaptive simulated annealing optimization is proposed to select the best subset of learners and the optimal combining strategy. The proposed method was evaluated using actual well test data and managed to achieve a remarkable performance with average errors of 4.7% and 2.4% for liquid and gas flow rates respectively. The accuracy of the developed VFM was also analyzed using cumulative deviation plot where the predictions are within a maximum deviation of ± 15%. Furthermore, the proposed ensemble method was compared to standard bagging and stacking and remarkable improvements have been observed in both accuracy and ensemble size. The proposed VFM is expected to be easier to develop and maintain than model-driven VFMs since only well test samples are required to tune the model. It is hoped that the developed VFM can augment and backup physical meters, improve data reconciliation, and assist in reservoir management and flow assurance ultimately leading to a more efficient oil recovery and less operating and maintenance costs.

StackSplit - a plugin for multi-event shear wave splitting analyses in SplitLab

SplitLab is a powerful and widely used tool for analysing seismological shear wave splitting of single event measurements. However, in many cases, especially temporary station deployments close to the noisy seaside, ocean bottom or for recordings affected by strong anthropogenic noise, only multi-event approaches provide stable and reliable splitting results. In order to extend the original SplitLab environment for such analyses, I present the StackSplit plugin that can easily be implemented within the well accepted main program. StackSplit grants easy access to several different analysis approaches within SplitLab, including a new multiple waveform based inversion method as well as the most established standard stacking procedures. The possibility to switch between different analysis approaches at any time allows the user for the most flexible processing of individual multi-event splitting measurements for a single recording station. Besides the provided functions of the plugin, no other external program is needed for the multi-event analyses since StackSplit performs within the available SplitLab structure which is based on MATLAB. The effectiveness and use of this plugin is demonstrated with data examples of a long running seismological recording station in Finland.

Estimation of acoustic macro models using a genetic full-waveform inversion: Applications to the Marmousi model

We have developed a stochastic full-waveform inversion that uses genetic algorithms (GA FWI) to estimate acoustic macro models of the P-wave velocity field. Stochastic methods such as GA severely suffer the curse of dimensionality, meaning that they require unaffordable computer resources for inverse problems with many unknowns and expensive forward modeling. To mitigate this issue, we have proposed a two-grid technique with a coarse grid to represent the subsurface for the GA inversion and a finer grid for the forward modeling. We have applied this procedure to invert synthetic acoustic data of the Marmousi model, and we have developed three different tests. The first two tests use a velocity model derived from standard stacking velocity analysis as prior information and differ only for the parameterization of the coarse grid. Their comparison indicates that a smart parameterization of the coarse grid may significantly improve the final result. The third test uses a linearly increasing 1D velocity model as prior information, a layer-stripping procedure, and a large number of model evaluations. All three tests return velocity models that fairly reproduce the long-wavelength structures of the Marmousi. First-break cycle skipping related to the seismograms of the final GA-FWI models is significantly reduced compared with that computed on the models used as prior information. Descent-based FWIs starting from final GA-FWI models yield velocity models with low and comparable model misfits and with an improved reconstruction of the structural details. The quality of the models obtained by GA FWI plus descent-based FWI is benchmarked against the models obtained by descent-based FWI started from a smoothed version of the Marmousi and started directly from the prior information models. Our results are promising and demonstrate the ability of the two-grid GA FWI to yield velocity models suitable as input to descent-based FWI.

Classifier Subset Selection for the Stacked Generalization Method Applied to Emotion Recognition in Speech.

In this paper, a new supervised classification paradigm, called classifier subset selection for stacked generalization (CSS stacking), is presented to deal with speech emotion recognition. The new approach consists of an improvement of a bi-level multi-classifier system known as stacking generalization by means of an integration of an estimation of distribution algorithm (EDA) in the first layer to select the optimal subset from the standard base classifiers. The good performance of the proposed new paradigm was demonstrated over different configurations and datasets. First, several CSS stacking classifiers were constructed on the RekEmozio dataset, using some specific standard base classifiers and a total of 123 spectral, quality and prosodic features computed using in-house feature extraction algorithms. These initial CSS stacking classifiers were compared to other multi-classifier systems and the employed standard classifiers built on the same set of speech features. Then, new CSS stacking classifiers were built on RekEmozio using a different set of both acoustic parameters (extended version of the Geneva Minimalistic Acoustic Parameter Set (eGeMAPS)) and standard classifiers and employing the best meta-classifier of the initial experiments. The performance of these two CSS stacking classifiers was evaluated and compared. Finally, the new paradigm was tested on the well-known Berlin Emotional Speech database. We compared the performance of single, standard stacking and CSS stacking systems using the same parametrization of the second phase. All of the classifications were performed at the categorical level, including the six primary emotions plus the neutral one.

Phase-Weighted Stacking Applied to Low-Frequency Earthquakes

Abstract We apply phase‐weighted stacking (PWS) to the analysis of low‐frequency earthquakes (LFEs) in the Parkfield, California, region and central Cascadia. The technique uses the coherence of the instantaneous phase among the stacked signals to enhance the signal‐to‐noise ratio (SNR) of the stack. We find that for picking LFE arrivals for the Parkfield, California, region and for LFE template formation in central Cascadia, PWS is extremely effective. For LFEs in the Parkfield, California, region, PWS yields many more usable phases than standard linear stacking; and, for LFE detection in Cascadia, PWS produces templates with much higher SNR than linear stacking.

Construction of invisibility cloaks of arbitrary shape and size using planar layers of metamaterials

Transformation optics (TO) is a powerful tool for the design of electromagnetic and optical devices with novel functionality derived from the unusual properties of the transformation media. In general, the fabrication of TO media is challenging, requiring spatially varying material properties with both anisotropic electric and magnetic responses. Though metamaterials have been proposed as a path for achieving such complex media, the required properties arising from the most general transformations remain elusive, and cannot implemented by state-of-the-art fabrication techniques. Here, we propose faceted approximations of TO media of arbitrary shape in which the volume of the TO device is divided into flat metamaterial layers. These layers can be readily implemented by standard fabrication and stacking techniques. We illustrate our approximation approach for the specific example of a two-dimensional, omnidirectional “invisibility cloak,” and quantify its performance using the total scattering cross section as a practical figure of merit.

Vector Evaluated and Objective Switching Approaches of Artificial Bee Colony Algorithm (ABC) for Multi-Objective Design Optimization of Composite Plate Structures

In this paper, a generic methodology based on swarm algorithms using Artificial Bee Colony (ABC) algorithm is proposed for combined cost and weight optimization of laminated composite structures. Two approaches, namely Vector Evaluated Design Optimization (VEDO) and Objective Switching Design Optimization (OSDO), have been used for solving constrained multi-objective optimization problems. The ply orientations, number of layers, and thickness of each lamina are chosen as the primary optimization variables. Classical lamination theory is used to obtain the global and local stresses for a plate subjected to transverse loading configurations, such as line load and hydrostatic load. Strength of the composite plate is validated using different failure criteria—Failure Mechanism based failure criterion, Maximum stress failure criterion, Tsai-Hill Failure criterion and the Tsai-Wu failure criterion. The design optimization is carried for both variable stacking sequences as well as standard stacking schemes and a comparative study of the different design configurations evolved is presented. Performance of Artificial Bee Colony (ABC) is compared with Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) for both VEDO and OSDO approaches. The results show ABC yielding a better optimal design than PSO and GA.

Implementation of the CMOS MEMS Condenser Microphone with Corrugated Metal Diaphragm and Silicon Back-Plate

This study reports a CMOS-MEMS condenser microphone implemented using the standard thin film stacking of 0.35 μm UMC CMOS 3.3/5.0 V logic process, and followed by post-CMOS micromachining steps without introducing any special materials. The corrugated diaphragm for the microphone is designed and implemented using the metal layer to reduce the influence of thin film residual stresses. Moreover, a silicon substrate is employed to increase the stiffness of the back-plate. Measurements show the sensitivity of microphone is −42 ± 3 dBV/Pa at 1 kHz (the reference sound-level is 94 dB) under 6 V pumping voltage, the frequency response is 100 Hz–10 kHz, and the S/N ratio >55 dB. It also has low power consumption of less than 200 μA, and low distortion of less than 1% (referred to 100 dB).

Topographic Correction for ALOS PALSAR Interferometry

L-band synthetic aperture radar (SAR) interferometry is very successful for mapping ground deformation in densely vegetated regions. However, due to its larger wavelength, the capacity to detect slow deformation over a short period of time is limited. Stacking and small baseline subset (SBAS) techniques are routinely used to produce time series of deformation and average deformation rates by reducing the contribution of topographic and atmospheric noise. For large sets of images that are presently available from C-band European Remote Sensing Satellites (ERS-1/2) and Environmental Satellite (ENVISAT), the standard stacking and SBAS algorithms are accurate. However, the same algorithms are often inaccurate when used for processing of interferograms from L-band Advanced Land Observing Satellite Phased Array type L-band SAR (ALOS PALSAR). This happens because only a limited number of interferograms is acquired and also because of large spatial baselines often correlated with the time of acquisition. In this paper two techniques are suggested that can be used for removing the residual topographic component from stacking and SBAS results, thereby increasing their accuracy.

Artificial Bee Colony (ABC) for multi-objective design optimization of composite structures

In this paper, we present a generic method/model for multi-objective design optimization of laminated composite components, based on Vector Evaluated Artificial Bee Colony (VEABC) algorithm. VEABC is a parallel vector evaluated type, swarm intelligence multi-objective variant of the Artificial Bee Colony algorithm (ABC). In the current work a modified version of VEABC algorithm for discrete variables has been developed and implemented successfully for the multi-objective design optimization of composites. The problem is formulated with multiple objectives of minimizing weight and the total cost of the composite component to achieve a specified strength. The primary optimization variables are the number of layers, its stacking sequence (the orientation of the layers) and thickness of each layer. The classical lamination theory is utilized to determine the stresses in the component and the design is evaluated based on three failure criteria: failure mechanism based failure criteria, maximum stress failure criteria and the tsai-wu failure criteria. The optimization method is validated for a number of different loading configurations—uniaxial, biaxial and bending loads. The design optimization has been carried for both variable stacking sequences, as well fixed standard stacking schemes and a comparative study of the different design configurations evolved has been presented. Finally the performance is evaluated in comparison with other nature inspired techniques which includes Particle Swarm Optimization (PSO), Artificial Immune System (AIS) and Genetic Algorithm (GA). The performance of ABC is at par with that of PSO, AIS and GA for all the loading configurations.

Quantum behaved Particle Swarm Optimization (QPSO) for multi-objective design optimization of composite structures

We present a new, generic method/model for multi-objective design optimization of laminated composite components using a novel multi-objective optimization algorithm developed on the basis of the Quantum behaved Particle Swarm Optimization (QPSO) paradigm. QPSO is a co-variant of the popular Particle Swarm Optimization (PSO) and has been developed and implemented successfully for the multi-objective design optimization of composites. The problem is formulated with multiple objectives of minimizing weight and the total cost of the composite component to achieve a specified strength. The primary optimization variables are – the number of layers, its stacking sequence (the orientation of the layers) and thickness of each layer. The classical lamination theory is utilized to determine the stresses in the component and the design is evaluated based on three failure criteria; Failure Mechanism based Failure criteria, Maximum stress failure criteria and the Tsai–Wu Failure criteria. The optimization method is validated for a number of different loading configurations – uniaxial, biaxial and bending loads. The design optimization has been carried for both variable stacking sequences as well as fixed standard stacking schemes and a comparative study of the different design configurations evolved has been presented. Also, the performance of QPSO is compared with the conventional PSO.

Optimal seismic-data acquisition in very shallow waters: Surveys in the Venice lagoon

Acquiring large amounts of data in very shallow waters of the immediate subtidal zone using a multichannel technique is unaffordable because moveout correction and a standard stacking procedure are required. The conventional inline longitudinal deployment of source and streamer leads to a phase difference along the hydrophone array for nonvertical arrivals. The array in this system is hard wired, so these phase differences cannot be accounted for and the summed output is attenuated as a result. We propose a transverse configuration, whereby the variation in phase along the array is smaller. The summed output improves because pairs of hydrophones with equivalent absolute offset are stacked in phase as a result of the symmetric configuration; however, all of the hydrophones are out of phase in the conventional geometry. Even at depths of [Formula: see text], a better image is obtained. The technique has been used during surveys in the Venice lagoon, where the water depth ranges from [Formula: see text].

Artificial immune system for multi-objective design optimization of composite structures

We present a new, generic method/model for multi-objective design optimization of laminated composite components using a novel multi-objective optimization algorithm developed on the basis of the artificial immune system (AIS) paradigm. A co-variant of the popular clonal selection principle called as the Objective Switching Clonal Selection Algorithm (OSCSA) has been developed and implemented successfully for the multi-objective design optimization of composites. The problem is formulated with multiple objectives of minimizing weight and the total cost of the composite component to achieve a specified strength. The primary optimization variables are—the number of layers, its stacking sequence (the orientation of the layers) and thickness of each layer. The classical lamination theory is utilized to determine the stresses in the component and the design is evaluated based on three failure criteria; failure-mechanism-based failure criteria, maximum stress failure criteria and the Tsai-Wu failure criteria. The optimization method is validated for a number of different loading configurations—uniaxial, biaxial and bending loads. The design optimization has been carried for both variable stacking sequences as well as fixed standard stacking schemes and a comparative study of the different design configurations evolved has been presented.

Vector evaluated particle swarm optimization (VEPSO) for multi-objective design optimization of composite structures

We present a generic method/model for multi-objective design optimization of laminated composite components, based on vector evaluated particle swarm optimization (VEPSO) algorithm. VEPSO is a novel, co-evolutionary multi-objective variant of the popular particle swarm optimization algorithm (PSO). In the current work a modified version of VEPSO algorithm for discrete variables has been developed and implemented successfully for the, multi-objective design optimization of composites. The problem is formulated with multiple objectives of minimizing weight and the total cost of the composite component to achieve a specified strength. The primary optimization variables are – the number of layers, its stacking sequence (the orientation of the layers) and thickness of each layer. The classical lamination theory is utilized to determine the stresses in the component and the design is evaluated based on three failure criteria; failure mechanism based failure criteria, Maximum stress failure criteria and the Tsai–Wu failure criteria. The optimization method is validated for a number of different loading configurations – uniaxial, biaxial and bending loads. The design optimization has been carried for both variable stacking sequences, as well fixed standard stacking schemes and a comparative study of the different design configurations evolved has been presented.