检索结果-内蒙古大学图书馆

IST 2006 - 2006 IEEE International Workshop on Imagining systems and Techniques

作者： Karras, D.A. Mertzios, B.G. Chalkis Institute of Technology Dept. Automation Hellenic Open University Rodu 2 Ano Iliupolis Athens 16342 Greece Thessaloniki Institute of Technology Thessaloniki Hellas Democritus University Laboratory of Automatic Control Systems Xanthi Hellas

ISBN: (纸本)1424402514

The goal of this paper is to present the development of a new image restoration methodology for extracting Magnetic Resonance Images (MRI) from reduced scans in k-space. The proposed approach considers the combined use of Support Vector Machine (SVM) models in a time delayed Bayesian priors framework and Bayesian restoration, in the problem of MR image reconstruction from sparsely sampled k-space, following several different sampling schemes, including spiral and radial. Effective solutions to this problem are indispensable especially when dealing with MRI of dynamic phenomena since then, rapid sampling in k-space is required. The goal in such a case is to make measurement time smaller by reducing scanning trajectories as much as possible. In this way, however, underdetermined equations are introduced and poor image extraction follows. It is suggested here that significant improvements could be achieved, concerning quality of the extracted image, by judiciously applying time delayed SVM priors and Bayesian estimation methods to the k-space data. More specifically, it is demonstrated that SVM neural network techniques could construct efficient time delayed Bayesian priors and introduce them in the procedure of Bayesian restoration. These time delayed Bayesian Priors are independent of specific image properties and probability distributions. They are based on training SVM neural filters to estimate the missing samples of complex k-space and thus, to improve k-space information capacity. Such a neural filter based time delayed Bayesian prior is integrated to the maximum likelihood procedure involved in the Bayesian reconstruction. It is found that the proposed methodology leads to enhanced image extraction results favorably compared to the ones obtained by the Integrated Bayesian MRI reconstruction approach involving simple and non time delayed SVM models priors [7], by the traditional Bayesian MRI reconstruction approach [1] as well as by the pure Neural Network (NN)

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Improved MRI Restoration by Integrating Bayesian formalism and Support Vector Machines in a Time Delayed priors Framework

Improved MRI Restoration by Integrating Bayesian formalism a...

引用

IEEE International Workshop on Imaging systems and Techniques (IST)

作者： D.A. Karras B.G. Mertzios Department Automation and Hellenic Chalkis Institute of Technology Open University Athens Greece Thessaloniki Institute of Technology Laboratory of Automatic Control Systems Thessaloniki Hellas and Democritus University Xanthi Greece

来源：评论

学校读者我要写书评

暂无评论

NEURO-FUZZY ADAPTIVE control AND MODELING A THRUSTER OF AUTONOMOUS NDERWATER VEHICLES

引用

IFAC Proceedings Volumes 2006年第16期39卷 843-849页

作者： F. Palis Y. Tsepkovskiy V. Filaretov D. Ukhimets Magdeburg University Institute of Electric Energy Systems Universitätplatz2 / Postfach 4120 39106 Magdeburg Deutschland Tel:(+49)-391-67-18135. Fax:(+49)-391-67-12481 Robotic laboratory Institute for automation and control process Nekrasovskaya str 76 105 Vladivostok 690014 the Russian Federation Tel: + 7 4232 26694 +7 ***

The paper presents a direct adaptive fuzzy approach for parameter identification and control of unknown nonlinear systems. To prove the performance of the proposed method an autonomous underwater vehicle (AUV) is modeled using an Adaptive Neuro-Fuzzy Inference System (ANFIS). To guaranty high accuracy the model is adapted online using system input and output data. From the model the current process parameters are identified and utilized for controller design. Here, Single Step Ahead Direct Adaptive control and Multi Steps Ahead Direct control schemes are considered.

关键词： underwater vehicles adaptive filters back propagation algorithm predictive control fuzzy modelling

来源：评论

学校读者我要写书评

暂无评论

Dynamic compensation for the deadzone of the electrohydraulic proportional position servo control system

Dynamic compensation for the deadzone of the electrohydrauli...

引用

SICE Annual Conference 2005

作者： Peng, Xi-Wei Hou, Zeng-Guang Geng, Qing-Bo Department of Automatic Control Beijing Institute of Technology Beijing 100081 China Laboratory of Complex Systems and Intelligence Science Institute of Automation Chinese Academy of Sciences Beijing 100080 China

The valve-controlled motor eletrohydraulic proportional servo system driving static load torque has severe time-varying deadzone and gain nonlinearities in the presence of the load torque variations. An experimental comparison study was made between several control methods. Experimental results show that the high gain control causes instability;the integral control aggravates stick slip and introduces hunting;the fixed deadzone compensation greatly reduces the positioning error when the load torque is a constant, but it causes poor positioning accuracy when the load torque varies greatly. The dynamic deadzone compensation based on the error e(t) and error-in-change Δe(t) can significantly improve the positioning precision over a wide range of load torque. © 2005 SICE.

关键词： Servomechanisms

来源：评论

学校读者我要写书评

暂无评论

Quadratic stability of process systems in generalized Lotka-Volterra form 6

Quadratic stability of process systems in generalized Lotka-...

引用

6th IFAC Symposium on Nonlinear control systems, NOLCOS 2004

作者： Magyar, A. Szederkényi, G. Hangos, K.M. Process Control Research Group Systems and Control Laboratory Computer and Automation Research Institute HAS P.O. Box 63 BudapestH-1518 Hungary Department of Computer Science University of Veszprém Egyetem u. 10 VeszpremH-8200 Hungary

The global and local stability of process systems in generalized Lotka-Volterra form is studied in this paper using entropy-like and quadratic Lyapunov function candidates. The global stability check for LV models is performed by solving an LMI for a diagonal positive semi-definite matrix using singular perturbation technique. It is shown that a quadratic Lyapunov function can also be determined by solving linear matrix inequalities (LMIs). In addition, the quadratic stability neighborhood is convex in the space of the quasi-monomials and can be estimated by computing its comer points using LMIs. Furthermore, it is proved that quadratic stability with a diagonal weighting matrix enables to construct adissipative-Hamiltonian description of the system. The developed methods are illustrated on the model of a continuously stirred tank reactor with a nonlinear reaction system. Copyright © 2004 IFAC.

关键词： Linear matrix inequalities

来源：评论

学校读者我要写书评

暂无评论

Fault Detection and Diagnosis of a Linear Hydraulic Servo Axis

引用

IFAC Proceedings Volumes 2004年第6期37卷 861-866页

作者： Marco Muenchhof Rolf Isermann Institute of Automatic Control Laboratory of Control Engineering and Process Automation Darmstadt University of Technology Landgraf-GeorgrStr. 4 D-64283 Darmstadt Germany Phone: + 49 6151 16-7406 FAX: +49 6151 16-74 21

process model based methods to detect faults in a hydraulic linear servo axis by measurement of up to five variables are presented. The faults that are detected include gas enclosures in the hydraulic fluid, such as air, vapour and foam. The control edges and the valve spool are also monitored in order to detect faults such as erosion of the control edges and grooving of the valve spool. Furthermore, the leakage flow between the cylinder chambers and to the surroundings are monitored, which allows to detect damages of the sealing between the cylinder chambers and between the cylinder and the pushrod. The external mechanical load and the valve spool dynamics are also included in the fault detection and diagnosis approach.

关键词： Fault Detection Fault Diagnosis Hydraulic Actuator Parameter Estimation

来源：评论

学校读者我要写书评

暂无评论

Optimizing the structure of hierarchical mixture of experts using genetic algorithms

Optimizing the structure of hierarchical mixture of experts ...

引用

International IEEE Conference on Intelligent systems, IS

作者： D.A. Karras C.E. Vlitakis Y.S. Boutalis B.G. Mertzios Chalkis Institute of Technology Department Automation and Hellenic Open University Athens Greece Department Electrical and Computer Engineering Democritus University of Thrace Xanthi Greece Department of Automation Thessdoniki Institute of Technology Laboratory of Automatic Control Systems Thessaloniki Hellas and Democritus University Xanthi Greece

This study is an effort to give a practical solution in the problem of optimizing the structure of the hierarchical mixture of experts model, which is a natural extension of the associative Gaussian mixture of experts system. We present two novel methods for optimizing such structures using genetic algorithms. Special concern is taken for reducing the computational time so as to efficiently allow the structure to "grow" while it evolves with the genetic algorithm. The main contribution of the paper lies on the efficient, topologically oriented, representations of such architectures so as to be optimized through involving genetic algorithms.

关键词： Genetic algorithms Biological cells Optimization methods Neural networks Classification tree analysis automation Decision trees Expert systems Computer architecture Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Quadratic stability of process systems in generalized lotka-volterra form

引用

IFAC Proceedings Volumes 2004年第13期37卷 1217-1222页

作者： A. Magyar G. Szederkényi K.M. Hangos Process Control Research Group Systems and Control Laboratory Computer and Automation Research Institute HAS H-1518 P.O. Box 63 Budapest Hungary Tel.: +361 2796000； fax: +361 466 7503 Department of Computer Science University of Veszprém H-8200 Egyetem u. 10 Veszprém Hungary Tel.: +***； fax: +***

The global and local stability of process systems in generalized Lotka-Volterra form is studied in this paper using entropy-like and quadratic Lyapunov function candidates. The global stability check for LV models is performed by solving an LMI for a diagonal positive semi-definite matrix using singular perturbation technique. It is shown that a quadratic Lyapunov function can also be determined by solving linear matrix inequalities (LMIs). In addition, the quadratic stability neighborhood is convex in the space of the quasi-monomials and can be estimated by computing its corner points using LMIs. Furthermore, it is proved that quadratic stability with a diagonal weighting matrix enables to construct a dissipative-Hamiltonian description of the system. The developed methods are illustrated on the model of a continuously stirred tank reactor with a nonlinear reaction system.

关键词： quasi-polynomial systems Lotka-Volterra systems stability analysis linear matrix inequalities Hamiltonian systems

来源：评论

学校读者我要写书评

暂无评论

Two-stage neural network models for MR image reconstruction from sparsely sampled k-space

Two-stage neural network models for MR image reconstruction ...

引用

IEEE International Workshop on Imaging systems and Techniques (IST)

作者： D.A. Karras B.G. Mertzios D. Graveron-Demilly D. van Ormondt Chalkis Institute of Technology Department Automation Hellenic Open University Athens Greece Thessaloniki Institute of Technology Tbessaloniki Laboratory of Automatic Control Systems Hellas and Demoeritus University Xanthi Greece Laboratoire de RMN CNRS UPRESA 5012 Universite Lyon l France Applied Physics Department Delft University of Technnology Delft Netherlands

A novel approach for magnetic resonance imaging (MRI) reconstruction using a two-stage neural network model, involving regularization techniques, is herein presented. The MRI reconstruction problem is considered when the k-space is sparsely scanned. Effective solutions to this problem are indispensable especially when dealing with MRI of dynamic phenomena since then, rapid sampling in k-space is required. The goal in such a case is to reduce the measurement time by omitting as many scanning trajectories as possible. The proposed model involves a regularized Kohonen feature map (SOFM) in the first stage which aims at quantizing the input variable space into smaller regions representative of the input space probability distribution and preserving its original topology, while increasing, on the other hand, cluster distances. This is achieved through adapting not only the winning neuron and its neighboring neurons weights but, also, loosing neurons weights during map's convergence phase. During convergence phase of the map, a group of support vector machines (SVM), associated with its codebook vectors, is simultaneously trained in an online fashion so that each SVM learns to respond when the input data belong to the topological space represented by its corresponding codebook vector. Moreover, these SVMs follow a task specific regularization strategy which aims at incorporating additional information in their training process. It is found that such a model results in an improved image reconstruction performance very favourably compared to the one obtained by the trivial zero-filled k-space approach or traditional more sophisticated interpolation approaches.

关键词： Neural networks Image reconstruction Magnetic resonance imaging Neurons Support vector machines Convergence Sampling methods Time measurement Input variables Probability distribution

来源：评论

学校读者我要写书评

暂无评论

Model-based fault detection and diagnosis for hydraulic braking systems 5

Model-based fault detection and diagnosis for hydraulic brak...

引用

5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical processes, Safeprocess 2003

作者： Straky, Harald Muenchhof, Marco Isermann, Rolf Institute of Automatic Control Laboratory of Control Engineering and Process Automation Darmstadt University of Technology Landgraf-Georg-Str. 4 DarmstadtD-64283 Germany

This paper describes fault detection and diagnosis techniques for a hydraulic passenger car braking system. First, a model of the hydraulic braking system is derived in state space representation. This model is subsequentlyemployed for fault detection and diagnosis. Fault detection is based on the calculation of the loss volume, i.e. the amount of brake fluid consumed due to the presence of a fault. The type of fault is determined by means of a correlation analysis approach. Furthermore, the location of the fault can also be determined. This is accomplished by analyzing the master brake cylinder pressure signals. The paper also presents numerical results of data recorded at a braking system testbed which illustrates the feasibility of the proposed approaches. © 2003 International Federation of automatic control.

关键词： Fault detection

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：