检索结果-内蒙古大学图书馆

Imbalance classification in a scaled-down wind turbine using radial basis function kernel and support vector machines

引用

ENERGY 2022年第PartC期238卷 122064-122064页

作者： Nogueira, Tiago de Oliveira Palacio, Gilderlanio Barbosa Alves Braga, Fabricio Damasceno Maia, Pedro Paulo Nunes Moura, Elineudo Pinho de Andrade, Carla Freitas de Rocha, Paulo Alexandre Costa Univ Fed Ceara Dept Engn Mecan BR-60455760 Fortaleza CE Brazil Univ Fed Ceara Dept Engn Met & Mat BR-60455760 Fortaleza CE Brazil

This work innovates by proposing the combination of DFA with the SVM and RBFK methods, two supervised algorithms that use the kernel-method, for the imbalance level classification in a scaled-down wind turbine. The results obtained were compared with other techniques proposed in previous works. The vibration signals analyzed were acquired under certain work conditions, and it is possible grouping them into 3 or 7 categories. It is worth mentioning that the dataset examined here is composed of the same signals used by previous works aiming at comparing results. The aforecited kernel-methods (Support Vector Machine and radial basis function kernel) classified, with a high success rate, the output of detrended fluctuation analysis (DFA) of vibration signals according to their respective working conditions. In the classification of three major classes, the performance achieved by both classifiers reduces with the increase in the rotation speed. The best average success rates reached at 900 rpm, 1200 rpm and 1500 rpm were, respectively, 99.96% by RBFK, 99.24% by RBFK and 98.73% by SVM. For seven imbalance levels, both classifiers showed the best performance at 900 rpm again. In this case, the best rates reached 98.83% by RBFK. At 1200 and 1500 rpm, the rates are slightly different. (c) 2021 Elsevier Ltd. All rights reserved.

关键词： Machine learning classification Vibration analysis Support vector machine radial basis function kernel

来源：评论

学校读者我要写书评

暂无评论

An Adaptive radial basis function kernel for Support Vector Data Description 1

引用

3rd International Workshop on Similarity-Based Pattern Recognition (SIMBAD)

作者： Lazzaretti, Andre E. Tax, David M. J. Inst Technol Dev Ave Comendador Franco 1341 Curitiba Parana Brazil Delft Univ Technol Pattern Recognit Lab NL-2628 CD Delft Netherlands

ISBN: (数字)9783319242613

ISBN: (纸本)9783319242613;9783319242606

For one-class classification or novelty detection, the metric of the feature space is essential for a good performance. Typically, it is assumed that the metric of the feature space is relatively isotropic, or flat, indicating that a distance of 1 can be interpreted in a similar way for every location and direction in the feature space. When this is not the case, thresholds on distances that are fitted in one part of the feature space will be suboptimal for other parts. To avoid this, the idea of this paper is to modify the width parameter in the radial basis function (RBF) kernel for the Support Vector Data Description (SVDD) classifier. Although there have been numerous approaches to learn the metric in a feature space for (supervised) classification problems, for one-class classification this is harder, because the metric cannot be optimized to improve a classification performance. Instead, here we propose to consider the local pairwise distances in the training set. The results obtained on both artificial and real datasets demonstrate the ability of the modified RBF kernel to identify local scales in the input data, extracting its general structure and improving the final classification performance for novelty detection problems.

关键词： Gaussian kernel kernel machines Metric learning Novelty detection radial basis function kernel Support Vector Data Description

来源：评论

学校读者我要写书评

暂无评论

An approach to facial expression recognition integrating radial basis function kernel and multidimensional scaling analysis

引用

SOFT COMPUTING 2014年第7期18卷 1363-1371页

作者： Wang, Shaowei Zhuo, Zhizheng Yang, Hongyu Li, Haiyun Capital Med Univ Sch Biomed Engn Beijing 100069 Peoples R China

To better deal with high dimensions and extract the essential feature of facial expression images in facial expression recognition task, a novel approach integrating radial basis function kernel and multidimensional scaling analysis is proposed in this paper. Firstly, the radial basis function kernel is invoked to map facial expression images to the Hilbert space. Then, Hilbert distance is substituted for the Euclidean distance and a neighbor graph is constructed to express the relationship between data points by employing k nearest neighbor method. Finally, we apply the modified MDS algorithm to reduce the dimension and extract features of facial expression images. Experiments results on the JAFFE database show that this proposed algorithm performs better than Isomap algorithm and supervised Isomap algorithm, and it is more feasible and effective.

关键词： Facial expression recognition Multidimensional scaling analysis radial basis function kernel Isometric mapping

来源：评论

学校读者我要写书评

暂无评论

Recent Advances in Data-Driven Modeling for Aerodynamic Applications using DLR's SMARTy Toolbox

Recent Advances in Data-Driven Modeling for Aerodynamic Appl...

引用

AIAA SciTech Forum

作者： Bekemeyer, Philipp Barklage, Alexander Chaves, Derrick Armando Hines Stradtner, Mario Goertz, Stefan German Aerosp Ctr DLR Inst Aerodynam & Flow Technol Lilienthalpl 7 D-38108 Braunschweig Germany German Aerosp Ctr DLR Inst Aerodynam & Flow Technol AIAA Lilienthalpl 7 D-38108 Braunschweig Germany

ISBN: (数字)9781624107115

ISBN: (纸本)9781624107115

From aircraft design to certification a huge amount of aerodynamic data is needed. In order to fulfil different requirements this data covers the entire flight envelope and includes pressure and shear stress distributions, global coefficients as well as derivatives. Moreover, data is typically gathered from different sources including flight tests, wind tunnel experiments or numerical simulations, and they are often available at various levels of fidelity, ranging from simple hand book methods to high-fidelity simulations. Within the past few years, the demand for efficient exploitation and exploration of these data sources became evident to further enhance existing designs, evaluate new technical capabilities and foster the availability of high-fidelity aerodynamic data in closely related disciplines. Driven by this, the aim of data-driven methods is to provide consistent aerodynamic data models based on various data-sources but with lower evaluation time and storage than the original models. Especially the increased availability of tools together with hardware in the field of machine learning in general and deep learning in particular has further accelerated demands as well as developments. In this paper we will show some recent advances in the field of data-driven modeling with a focus on applied aerodynamic challenges. This contains multi-fidelity modeling for aero-performance and stability and control databases, reduced-order modeling based on graph neural networks for the prediction of surface pressure distributions as well as data fusion to combine results from numerical and experimental analysis with different spatial resolutions. For all methods results will be shown for an industrial-relevant military configuration.

关键词： Data Driven Model Reduced Order Modelling Aerodynamic Shape Optimization Multilayer Perceptron Leading Edge Vortex Controller Data Fusion Transonic Wind Tunnel Flight Testing Top Level Aircraft Requirements radial basis function kernel

来源：评论

学校读者我要写书评

暂无评论

Analysis and Uncertainty Quantification of a Hybrid Laminar Flow Control System

引用

AIAA JOURNAL 2022年第10期60卷 5735-5749页

作者： Barklage, Alexander Roemer, Ulrich Bertram, Anna Bekemeyer, Philipp Himisch, Jan Radespiel, Rolf Badrya, Camli Tech Univ Carolo Wilhelmina Braunschweig Inst Fluid Mech Hermann Blenk Str 37 D-38108 Braunschweig Germany Tech Univ Carolo Wilhelmina Braunschweig Inst Dynam & Schwingungen Schleinitzstr 20 D-38106 Braunschweig Germany German Aerosp Ctr DLR Inst Aerodynam & Flow Technol Ctr Comp Applicat Aerosp Sci & Engn Dept Lilienthalpl 7 D-38108 Braunschweig Germany German Aerosp Ctr DLR Inst Aerodynam & Flow Technol Lilienthalpl 7 D-38108 Braunschweig ME Germany Tech Univ Carolo Wilhelmina Braunschweig Inst Aerodynam & Flow Technol Lilienthalpl 7 D-38108 Braunschweig Germany Tech Univ Carolo Wilhelmina Braunschweig Inst Inst Fluid Mech Hermann Blenk Str 37 D-38108 Braunschweig Germany

This study aims to quantify suction velocities for hybrid laminar flow control of a vertical fin by taking data and model uncertainties into account. Because suction velocities cannot be measured directly, they are calculated using the Darcy-Forchheimer model, which needs the pressure distribution and porosity characteristics as input. Uncertainties in the pressure distribution are characterized using a novel data fusion method incorporating wind-tunnel and numerical data. Furthermore, geometrical uncertainties of the airfoil that alter the pressure distribution are considered. Two methods are proposed to model the porosity characteristics of the surface: one of which uses a local representation, and the other a global representation. In addition, the modeling error of the Darcy-Forchheimer model is estimated. These input uncertainties are quantified and propagated by Monte Carlo sampling through the Darcy-Forchheimer model. It reveals that local information on the porosity characteristics is essential for accurate suction velocity estimates for this specific application. A sensitivity study shows that the most influential contribution to the variance of suction velocities is the uncertainty in the porosity characteristics, followed by uncertainties in the pressure distribution. The model error and uncertainties in the reference pressure are shown to have a negligible influence.

关键词： Hybrid Laminar Flow Control German Dutch Wind Tunnels Data Fusion Airfoil Reynolds Averaged Navier Stokes Cumulative Distribution function Courant Friedrichs Lewy Natural Laminar Flow Laminar to Turbulent Transition radial basis function kernel

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Vehicle-Based and Driver-Based Features for Driver Drowsiness Monitoring by Support Vector Machines

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2022年第12期23卷 23164-23178页

作者： Baccour, Mohamed Hedi Driewer, Frauke Schaeck, Tim Kasneci, Enkelejda Mercedes Benz AG Mercedes Benz Cars Dev Dept D-71059 Sindelfingen Germany Univ Tubingen Dept Human Comp Interact D-72076 Tubingen Germany

Driver drowsiness is a serious threat to road safety. Most driver monitoring systems (DMSs) already embedded in vehicles to detect drowsiness use vehicle-based features (i.e., measures) computed by outward-facing cameras for lane tracking or steering wheel angle sensors to analyze lane keeping and steering control behavior. Such DMSs are referred to as indirect DMSs as they monitor drowsiness indirectly through driving performance. In this work, we extend this classical technique by using a driver monitoring camera for tracking driverbased features associated with eye blinking behavior and head movements. We refer to DMSs based only on a driver monitoring camera as direct DMSs as they monitor drowsiness directly through observable driver-based behavioral cues. In this work, we conduct a comparative analysis between an indirect and direct DMS. We also combine vehicle-based and driver-based features to examine the potential of a so-called hybrid DMS. To this end, we use a database collected from 70 participants in driving simulator experiments. The comparative analysis is performed by means of the correlation-based feature selection technique and support vector machines independently. With a balanced accuracy of 87.1%, the direct DMS significantly outperforms the indirect DMS, which reaches a balanced accuracy of only 77.9%. The hybrid DMS achieves a slightly better balanced accuracy of 87.7%. This work motivates the development and use of direct or hybrid DMSs to detect driver drowsiness and increase road safety.

关键词： Drowsiness road safety driver monitoring system driver camera driving simulator sensor fusion machine learning support vector machines radial basis function kernel feature selection

来源：评论

学校读者我要写书评

暂无评论

Probabilistic REpresentatives Mining (PREM): A Clustering Method for Distributional Data Reduction

引用

AIAA JOURNAL 2022年第4期60卷 2580-2596页

作者： Gao, Zhenyu Puranik, Tejas G. Mavris, Dimitri N. Georgia Inst Technol Sch Aerosp Engn Atlanta GA 30332 USA

Complex computations and analyses on massive data sets can be impractical or infeasible. Data reduction is a crucial problem in the era of big data to obtain a reduced representation of the data set to facilitate more efficient yet accurate analyses. To best preserve the integrity of the original data set, a reduced representation aims to best maintain the same data distribution, referred to as a probabilistically representative subset. This paper considers the problem of reducing a large data set to very small such subsets at which random sampling does not perform well enough. We propose a data mining approach called Probabilistic Representatives Mining (PREM) to tackle this challenge. PREM uses balanced clustering to prevent undersampling and oversampling issues and multistage computing strategy to achieve better scalability and consistency. Numerical experiments on typical probability distributions and real-world data sets in the field of aeronautics and astronautics demonstrate PREM's superiority over the baselines. An uncertainty quantification case study from aviation environmental impact modeling further shows PREM's effectiveness and accuracy in generating probabilistically representative small samples for costly computations. Potential limitations and extensions of the method are also discussed in the paper.

关键词： Data Reduction Computing Strategies Continuous Probability Distributions Data Mining Algorithms Air Transportation System Astronautics Aeronautics Aviation Environmental Design Tool radial basis function kernel LANDSAT

来源：评论

学校读者我要写书评

暂无评论

DRBF-DS: Double RBF kernel-Based Deep Sampling with CNNs to Handle Complex Imbalanced Datasets

引用

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING 2022年第8期47卷 10043-10070页

作者： Rout, Subhashree Mallick, Pradeep Kumar Mishra, Debahuti Kalinga Inst Ind Technol KIIT Deemed Univ Sch Comp Engn Bhubaneswa Odisha India Siksha O Anusandhan Deemed Univ Dept Comp Sci & Engn Bhubaneswar Odisha India

The inappropriate distribution of samples is known as the data imbalance problem leading to minority and majority classes, often in the healthcare domain due to weakly labelled data. The challenge of working with these imbalanced datasets is to develop predictive models by augmenting the samples to minority class to equalize to the number of samples of the majority class. The successful integration of double radial basis function kernel with deep sampling and convolutional neural networks is experimented to propose a hybrid sampling method coined as DRBF-DS, which generates extended samples, and augments those to the minority class in this study. In this paper, the proposed DRBF-DS is compared with the variants of simple and widely used probability and non-probability-based sampling strategies. It confirms that the proposed approach outperforms the sampling strategies in terms of accuracy and other performance measures used for validation. Experimental validations are performed on five publicly available complex imbalanced datasets to demonstrate the effectiveness of the proposed DRBF-DS based on various evaluation metrics.

关键词： Fractal AI Complex imbalanced dataset radial basis function kernel Deep sampling Convolutional neural networks (CNN) Extreme learning machine

来源：评论

学校读者我要写书评

暂无评论

Efficient Acquisition functions for Bayesian Optimization in the Presence of Hidden Constraints

Efficient Acquisition Functions for Bayesian Optimization in...

引用

AIAA Aviation Forum

作者： Tfaily, Ali Kokkolaras, Michael Barto, Nathalie Diouane, Youssef McGill Univ Dept Mech Engn Montreal PQ Canada Bombardier Adv Prod Dev Dept Montreal PQ Canada Univ Toulouse ONERA DTIS Dept Informat Proc & Syst Toulouse France Polytech Montreal Dept Math & Ind Engn Montreal PQ Canada

ISBN: (数字)9781624107047

ISBN: (纸本)9781624107047

A common challenge in engineering design applications of Bayesian optimization is the presence of non-computable domains, which are often called hidden or unknown constraints. In these cases, the required function values cannot be computed in certain regions of the design space. In this work, we propose a novel method to handle hidden constraints by modifying a portion of the acquisition function of a typical Bayesian optimization algorithm using classifiers. Specifically, the proposed approach considers the exploitation part of the acquisition function, allowing the optimization algorithm to search unexplored regions of the design space. Convergence properties are maintained regardless of the type of classifier used. Results obtained for numerical experiments on analytical test problems confirm the usefulness of the proposed method of handling hidden constraints in the context of Bayesian optimization.

关键词： radial basis function kernel Optimization Algorithm Cumulative Distribution function Computational Modeling Software Development Aerospace Engineering Design of Experiments Design Optimization Aircraft Design

来源：评论

学校读者我要写书评

暂无评论

Entropy kernel Graph Cut Feature Space Enhancement with SqueezeNet Deep Neural Network for Textural Image Segmentation

引用

INTERNATIONAL JOURNAL OF IMAGE AND GRAPHICS 2024年第0期

作者： Niazi, Mehrnaz Rahbar, Kambiz Pishtazan Higher Educ Inst Dept Comp Engn Shiraz Iran Islamic Azad Univ Dept Comp Engn South Tehran Branch Tehran Iran

Recently, image segmentation based on graph cut methods has shown remarkable performance on a set of image data. Although the kernel graph cut method provides good performance, its performance is highly dependent on the data mapping to the transformation space and image features. The entropy-based kernel graph cut method is suitable for segmentation of textured images. Nonetheless, its segmentation quality remains significantly contingent on the accuracy and richness of feature space representation and kernel centers. This paper introduces an entropy-based kernel graph cut method, which leverages the discriminative feature space extracted from SqueezeNet, a deep neural network. The fusion of SqueezeNet's features enriches the segmentation process by capturing high-level semantic information. Moreover, the extraction of kernel centers is refined through a weighted k-means approach, contributing further to the segmentation's precision and effectiveness. The proposed method, while exploiting the benefits of suitable computational load of graph cut methods, will be a suitable alternative for segmenting textured images. Laboratory results have been taken on a set of well-known datasets that include textured shapes in order to evaluate the efficiency of the algorithm compared to other well-known methods in the field of kernel graph cut.

关键词： Image segmentation kernel graph cut radial basis function kernel textured images weighted clustering

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：