检索结果-内蒙古大学图书馆

Efficient Detection of Hessian Matrix Sparsity Pattern

ACM COMMUNICATIONS IN COMPUTER ALGEBRA 2016年第4期50卷 151-154页

作者： Carter, Richard G. Hossain, Shahadat Sultana, Marzia DNV GL Katy TX 77449 USA Univ Lethbridge Dept Math & Comp Sci Lethbridge AB Canada

Evaluation of the Hessian matrix of a scalar function is a subproblem in many numerical optimization algorithms. For large-scale problems often the Hessian matrix is sparse and structured, and it is preferable to exploit such information when available. Using symmetry in the second derivative values of the components it is possible to detect the sparsity pattern of the Hessian via products of the Hessian matrix with specially chosen direction vectors. We use graph coloring methods and employ efficient sparse data structures to implement the sparsity pattern detection algorithms. Results from preliminary numerical testings are highly promising.

关键词： Efficient Detection Pattern optimization algorithms Hessian Hessian matrix SCALAR FUNCTIONS

来源：评论

学校读者我要写书评

暂无评论

A Review Paper on the Elimination of Harmonics in Multilevel Inverters using Bioinspired algorithms

A Review Paper on the Elimination of Harmonics in Multilevel...

引用

IEEE International Conference on Circuit, Power and Computing Technologies (ICCPCT)

作者： Jacob, Teena Suresh, L. Padma Noorul Islam Univ Dept EEE Thuckalay Tamil Nadu India

ISBN: (纸本)9781509012770

Multilevel inverters are highly capable of achieving higher quality output voltage waveforms and higher power ratings with the help of their multilevel structure. They have been of great interest in the field of power industry and are best suited for reactive power compensation. Multilevel voltage source inverters are capable of operating at high voltagewith less electromagnetic interference and results in higher efficiency. The harmonic elimination in a multilevel voltage source inverter is of atmost importance and different types of modulation strategies can be applied to the inverters to eliminate these harmonics. Among these modulation techniques, Selective harmonic elimination PWM is asignificant switching strategy that can be applied to the output voltage waveform of multilevel inverters for lower order harmonic elimination. This paper gives a review on the various optimization algorithms that is been used for the SHEPWM technique. Performance comparisons of these optimization algorithms in SHEPWM are discussed.

关键词： Multilevel voltage source inverter Selective Harmonic Elimination Pulse Width Modulation (SHEPWM) optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

A SEQUENTIAL SAMPLING ALGORITHM FOR MULTI-STAGE STATIC COVERAGE PROBLEMS

A SEQUENTIAL SAMPLING ALGORITHM FOR MULTI-STAGE STATIC COVER...

引用

ASME International Design Engineering Technical Conference / Computer and Information in Engineering Conference (IDETC/CIE)

作者： Manjunatha, Hemanth Huang, Jida Zhang, Binbin Rai, Rahul SUNY Buffalo Mech & Aeronaut Engn Buffalo NY 14260 USA SUNY Buffalo Ind & Syst Engn Buffalo NY 14260 USA SUNY Buffalo Mech & Aeronaut Engn MAD LAB Buffalo NY 14260 USA

ISBN: (纸本)9780791850114

It is critical in many system-engineering problems (such as surveillance, environmental monitoring, and cooperative task performance) to optimally allocate resources in the presence of limited resources. Static coverage problem is an important class of the resource allocation problems that focuses on covering an area of interest so that the activities in the area of interest can be detected/monitored with higher probability. In many practical settings (primarily due to financial constraints) a system designer has to allocate resources in multiple stages. In each stage, the system designer can assign a fixed number of resources (agents). In the multi-stage formulation, the agents locations for the next stage are dependent on all the agents location in the previous stages. Such multi-stage static coverage problems are non-trivial to solve. In this paper, we propose a robust and efficient sequential sampling algorithm to solve the multi-stage static coverage problem in the presence of probabilistic resource intensity allocation maps (RIAMs). The agents locations are determined by formulating this problem as an optimization problem in the successive stage. Three different objective functions are compared and discussed from the aspects of decreasing L2 difference and Sequential Minimum Energy Design (SMED). It is shown. that utilizing SMED objective function leads to a better approximation of the RIAMs. Two heuristic algorithms, i.e. cuckoo search, and pattern search, are used as optimization algorithms. Numerical functions and real-life applications are provided to demonstrate the robustness and efficiency of the proposed approach.

关键词： Approximation optimization Robustness Probability Systems engineering Design algorithms optimization algorithms Surveillance Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Analysis and Application of Compressive Sensing Technique for Rapid Magnetic Resonance Imaging 1

Analysis and Application of Compressive Sensing Technique fo...

引用

IEEE International Conference on Recent Trends in Electronics, Information and Communication Technology (RTEICT)

作者： Kavitha, S. Shrividya, G. Bharathi, S. H. NMAMIT Dept ECE Nitte India Reva Univ Sch E&C Engn Bengaluru India

ISBN: (纸本)9781509007745

Shannon/Nyquist sampling theorem states that signal should be sampled at least two times faster than its frequency to avoid any loss in the information, when it is captured. Too high sampling rate results in too many samples. So data compression is required before the signal is stored or transmitted. In some applications such as medical scanners, it is very expensive to increase the sampling rate. An alternative sampling theory known as compressive sensing or compressed sensing (CS) is introduced recently, which facilitates the capturing of a signal or image at a rate considerably below the Nyquist rate. Magnetic resonance imaging (MRI) is an important tool used in medical imaging but has very low data acquisition speed. This makes the patient to be exposed to a scanning machine for a long time. In this work, it is analyzed that how compressive sensing can be applied to MRI without the loss of resolution using number of optimization algorithms and also different optimization algorithms are compared using the parameters such as signal to noise ratio (SNR) and CPU time. This enables the significant reduction in scan time, which benefits both the patient and health care economics.

关键词： Shannon/Nyquist sampling theorem CS MRI optimization algorithms SNR CPU time

来源：评论

学校读者我要写书评

暂无评论

Application of information processes applicative modelling to virtual machines auto configuration 20th

Application of information processes applicative modelling t...

引用

20th International Conference on Knowledge - Based and Intelligent Information and Engineering Systems (KES)

作者： Zykov, Sergey Shumsky, Leonid MEPhI Natl Res Nucl Univ 31 Kashirskoe Shosse Moscow 115409 Russia Innopolis Univ 1 Univ Skaya St Innopolis 420500 Russia

The paper discusses application of the algorithm based information processes modeling and Q-machine learning to the task of virtual machines auto configuration. Usage of virtual machines is one the most common solution nowadays for almost every companies. Although using of virtual machines simplifies settings of hardware landscape and allows physical servers decentralization they may cause performance decrease. To overcome this issue, we propose using method of virtual machines auto configuration. The method is based on the following premises: tracking real business process held by virtual machines, using algorithms of machine learning to investigate optimal configuration, and virtual machine configuration by independent process. (C) 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/). Peer-review under responsibility of KES International

关键词： Virtual machines Process modeling optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Determination of Force Parameters for Milling Simulations by Combining optimization and Simulation Techniques

引用

JOURNAL OF MANUFACTURING SCIENCE AND ENGINEERING-TRANSACTIONS OF THE ASME 2016年第4期138卷 044502-044502页

作者： Freiburg, Dennis Hense, Rouven Kersting, Petra Biermann, Dirk TU Dortmund Univ Inst Machining Technol Baroper Str 303 D-44227 Dortmund Germany Univ Magdeburg Inst Proc Technol & Qual Management D-39106 Magdeburg Germany

Milling is a machining process in which material removal occurs due to the rotary motion of a cutting tool relative to a typically stationary workpiece. In modern machining centers, up to and exceeding six degrees of freedom for motion relative to the tool and workpiece are possible, which results in a very complex chip and force formation. For the process layout, simulations can be used to calculate the occurring process forces, which are needed, e.g., for the prediction of surface errors of the workpiece, or for tool wear and process optimization examinations. One limiting factor for the quality of simulation results is the parametrization of the models. The most important parameters for milling simulations are the ones that calibrate the force model, as nearly every modeled process characteristic depends on the forces. This article presents the combination of a milling simulation with the Broyden-Fletcher-Goldfarb-Shanno (BFGS) optimization algorithm for the fast determination of force parameters that are valid for a wide range of process parameters. Experiments were conducted to measure the process forces during milling with different process parameters. The measured forces serve as basis for tests regarding the quality of the determined force parameters. The effect of the tool runout on the optimization result is also discussed, as this may have significant influence on the forces when using tools with more than one tooth. The article ends with a conclusion, in which some notes about the practical application of the algorithm are given.

关键词： Simulation optimization Milling Errors optimization algorithms Cutting

来源：评论

学校读者我要写书评

暂无评论

OPTIMAL DESIGN OF 3D PRINTED SPIRAL TORSION SPRINGS

引用

ASME Conference on Smart Materials, Adaptive Structures and Intelligent Systems

作者： Scarcia, Umberto Berselli, Giovanni Melchiorri, Claudio Ghinelli, Manuele Palli, Gianluca Univ Bologna Dept Elect Elect & Informat Engn Guglielmo Marcon I-40126 Bologna Italy Univ Genoa Dept Mech Energet Management & Transportat Genoa Italy

ISBN: (纸本)9780791850497

Spiral Torsion Springs (STS) are generally manufactured employing medium/high-carbon steel alloys shaped as thin rods with rectangular cross section. Meanwhile, plastic materials (e.g. ABS or PM), currently used in freeform manufacturing processes, may not be suited for several applications, owing to the low material yield strength and the rather poor fatigue life. Despite the above-mentioned limitations, the main advantages of a 3D printing process, as compared to more traditional manufacturing techniques, are the design flexibility and the possibility to directly integrate elastic components within a joint mechanism produced as a single (monolithic) part. In particular, provided that the external forces acting on the spring coils are maintained within a certain threshold and that the spring geometry is suitably optimized, a reliable 3D printed STS alternative to traditional steel springs is actually feasible. Given these premises, the main purpose of the present paper is to propose a model based optimization algorithm that allows to optimally size STS for user-specified torque-deflection characteristics. Optimal STS geometries are then realized in ABS via Fused Deposition Manufacturing, and subsequently tested with a purposely-designed experimental set-up. Furthermore, the behavior of each STS sample (in terms of stiffness and equivalent Von Mises stress) is evaluated by means of non-linear finite elements analysis, in order to check the correspondence with the expected behavior. Finally, numerical and experimental results are provided, which demonstrate the prediction capabilities of the proposed modeling/optimization techniques, and confirm that well-behaved STS can be conceived and produced. Envisaged applications concern the development of smart structures for robot design, such as multi-articulated compliant robotic chains that can be used as low-cost manipulators (i.e. arm) or as mini-manipulators (i.e. fingers). The proposed approach effectively simplifies

关键词： Fatigue life Robotics Robots Smart structures Manipulators Stiffness Design Springs Geometry Torsion Finite element analysis Deflection Modeling optimization algorithms Rods Carbon Steel Additive manufacturing Manufacturing Alloy steel Plastics optimization Torque Chain Yield strength Stress

来源：评论

学校读者我要写书评

暂无评论

AUTOMATED ITERATIVE TOLERANCE VALUE ALLOCATION AND ANALYSIS

AUTOMATED ITERATIVE TOLERANCE VALUE ALLOCATION AND ANALYSIS

引用

ASME International Design Engineering Technical Conference / Computer and Information in Engineering Conference (IDETC/CIE)

作者： Biswas, Deepanjan Venkiteswaran, Adarsh Hejazi, Sayed Mohammad Shah, Jami J. Davidson, Joseph K. Arizona State Univ Design Automat Lab Tempe AZ 85287 USA

ISBN: (纸本)9780791850114

Dimensional variation is inherent in manufacturing and it is impossible to attain exact nominal dimensions. Ideally, designer should accommodate this variation likelihood in design stage and define an allowable variation. This allowable variation is represented as tolerances and is either considered a bounded zone or a scaled allowable process standard deviation (e.g. 6 times standard deviation). The allowable tolerances are usually constrained by assemblability, functionality and manufacturing economics. Meeting these constraints simultaneously becomes a paramount task and designers usually never consider production economics when specifying tolerances. As a consequence rarely do the tolerances specified by product designer match that of the process designer. Automated tolerance value allocation can empower the product designer to include all the constraints at the design phase and reduce the overall time line between development to production. Automated tolerance allocation method described in this paper is intended for use by the designer and encompasses only 1st order tolerancing. The Critical stack detection/loop detection tool extracts the assembly level stacks that dictate assemblability. The allocation tool utilizes these stacks to distribute tolerance budget by a rule of thumb. These stacks are subjected to variation analysis to compute acceptance rates and used as feedback for iterative reallocation using a hill climbing optimization algorithm till statistical fit requirements are satisfied or allocation gets exhausted. The algorithm is tested on some case studies and presented in the paper.

关键词： algorithms Manufacturing Dimensions Design optimization algorithms Economics Feedback

来源：评论

学校读者我要写书评

暂无评论

Exploiting Sample Diversity in Distributed Machine Learning Systems 16

Exploiting Sample Diversity in Distributed Machine Learning ...

引用

16th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid)

作者： Liu, Zhiqiang Shi, Xuanhua Jin, Hai Huazhong Univ Sci & Technol Sch Comp Sci & Technol Big Data Technol & Syst Lab Serv Comp Technol & Syst LabCluster & Grid Comp Wuhan 430074 Peoples R China

ISBN: (纸本)9781509024537

With the increase of machine learning scalability, there is a growing need for distributed systems which can execute machine learning algorithms on large clusters. Currently, most distributed machine learning systems are developed based on iterative optimization algorithm and parameter server framework. However, most systems compute on all samples in every iteration and this method consumes too much computing resources since the amount of samples is always too large. In this paper, we study on the sample diversity and find that most samples contribute little to model updating during most iterations. Based on these findings, we propose a new iterative optimization algorithm to reduce the computation load by reusing the iterative computing results. The experiment demonstrates that, compared to the current methods, the algorithm proposed in this paper can reduce about 23% of the whole computation load without increasing of communications.

关键词： optimisation distributed processing iterative methods learning (artificial intelligence) iterative methods Distributed processing optimization algorithms Artificial Intelligence Machine Learning Model updating Iterative Distributed Systems Communication Research

来源：评论

学校读者我要写书评

暂无评论

Analytical and numerical studies of thalamo-cortical neural population models during general anesthesia

Analytical and numerical studies of thalamo-cortical neural ...

引用

作者： Meysam HASHEMI UNIVERSITE DE LORRAINE

学位级别：博士

Although general anesthesia is an indispensable tool in today's medical surgery, its precise underlying mechanisms are still unknown. During the propofol-induced sedation, the anesthetic actions on the microscopic single neuron scale lead to specific changes in macroscopic-scale ob- servables such as electroencephalogram (EEG) signals. For low concentration of propofol these characteristic changes comprised increased activity in the δ− (0.5-4 Hz) and α− (8-13 Hz) fre- quency bands over the frontal head region, but increased δ− and decreased α−power activity over the occipital region. In this thesis, we employ thalamo-cortical neural population models, and based on the experimental data, the propofol effects on the synaptic and extra-synaptic GABAergic receptors located in the cortex and thalamus are modelized to understand the mech- anisms underlying the observed certain changes in EEG-spectral power. It is shown that the models reproduce well the characteristic spectral features observed experimentally. A key find- ing of this work is that the origin of δ−rhythm is fundamentally different from the α−rhythm. Our results indicate that dependent on the mean potential values of the system resting states, an increase or decrease in the thalamo-cortical gain functions results in an increase or decrease in the α−power, respectively. In contrast, the evolution of the δ−power is rather independent of the system resting states; the enhancement of spectral power in the δ−band results from the increased synaptic or extra-synaptic GABAergic inhibition for both increasing and decreasing nonlinear gain functions. Furthermore, we aim to identify the parameters of a thalamo-cortical model by fitting the model power spectrum to the EEG recordings. To this end, we address the task of parameter estimation in the models that are described by a set of stochastic ordi- nary or delay differential equations. Two case studies dealing with noisy pseudo-experimental data are first carried out

关键词： general anesthesia thalamo-cortical neural population models propofol EEG-spectral power optimization algorithms

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：