检索结果-内蒙古大学图书馆

CES/IEEE 5th International Power Electronics and Motion Control Conference

作者： Liu Kaipei Zhang Junmin Wuhan Univ Sch Elect Engn Wuhan 430072 Peoples R China

ISBN: (纸本)1424404487

Based on the newton learning algorithm, a new algorithm for TLS estimation and its application in harmonic detection is presented in this paper. It is shown that the LMS algorithm and Constrained Anti-Hebbian algorithm are two examples of the proposed algorithm in this paper. To solve the contravention between the detecting precision and speediness, analog detecting architectures are studied. The simulation results show that its noise rejection capability is superior to those of the LMS and Constrained Anti-Hebbian algorithms and the new algorithm gives faster convergence and better precision than another two methods.

关键词： newton algorithm TLS estimation adaptive harmonic detection

来源：评论

学校读者我要写书评

暂无评论

Improved Least-Squares Methods for Source Localization: An Iterative Re-Weighting Approach

Improved Least-Squares Methods for Source Localization: An I...

引用

IEEE International Conference on Digital Signal Processing (DSP)

作者： Ismailova, Darya Lu, Wu-Sheng Univ Victoria Elect & Comp Engn Victoria BC Canada

ISBN: (纸本)9781479980581

Locating a radiating source from range or range-difference measurements in a passive sensor network has recently attracted an increasing amount of research interest as it finds applications in a wide range of network-based wireless systems. Striking results for these localization problems have emerged using squared range (SR-LS) or squared range-difference (SRD-LS) least-squares (LS) approaches. In this paper, we present improved LS methods that demonstrate improved localization performance when compared with the best known results from the literature.

关键词： improved globally optimal solution generalized trust region subproblems (GTRS) least squares non-convex iterative re-weighting source localization range measurements range-difference measurements newton algorithm

来源：评论

学校读者我要写书评

暂无评论

Statistical efficiency of adaptive algorithms

Statistical efficiency of adaptive algorithms

引用

INNS/IEEE International Joint Conference on Neural Networks (IJCNN 03)

作者： Widrow, B Kamenetsky, M Stanford Univ ISL Dept Elect Engn Stanford CA 94305 USA

The statistical efficiency of a learning algorithm applied to the adaptation of a given set of variable weights is defined as the ratio of the quality of the converged solution to the amount of data used in training the weights. Statistical efficiency is computed by averaging over an ensemble of learning experiences. A high quality solution is very close to optimal, while a low quality solution corresponds to noisy weights and less than optimal performance. In this work, two gradient descent adaptive algorithms are compared, the LMS algorithm and the LMS/newton algorithm. LMS is simple and practical. and is used in many applications worldwide. LMS/newton is based on newton's method and the LMS algorithm. LMS/newton is optimal in the least squares sense. It maximizes the quality of its adaptive solution while minimizing the use of training data. Many least squares adaptive algorithms have been devised over the years, but no other least squares algorithm can give better performance, on average, than LMS/newton. LMS is easily implemented, but LMS/newton, although of great mathematical interest, cannot be implemented in most practical applications. Because of its optimality, LMS/newton serves as a benchmark for all least squares adaptive algorithms. The performances of LMS and LMS/newton are compared, and it is found that under many circumstances, both algorithms provide equal performance. For example, when both algorithms are tested with statistically nonstationary input signals, their average performances are equal. When adapting with stationary input signals and with random initial conditions, their respective learning times are on average equal. However, under worst-case initial conditions, the learning time of LMS can be much greater than that of LMS/newton, and this is the principal disadvantage of the LMS algorithm. But the strong points of LMS are ease of implementation and optimal performance under important practical conditions. For these reasons, the LMS algo

关键词： LMS newton algorithm least squares efficiency

来源：评论

学校读者我要写书评

暂无评论

LQ multimodel optimal controller for wind turbine with Kalman filter wind speed estimation

LQ multimodel optimal controller for wind turbine with Kalma...

引用

International Conference on Communications and Electrical Engineering (ICCEE)

作者： Garmat, Abdelkader Azzouzi, Messaouda Bouchekima, Bachir Univ Ziane Achour Djelfa Dept Elect Engn Djelfa Algeria Univ Kasdi Merbah Ouargla Dept Renewable Energies Ouargla Algeria

ISBN: (纸本)9781728101125

In the present paper, a LQ multimodel optimal control approach for variable speed with pitch regulated (operating at high wind speeds) is proposed to allow the wind turbine to participate in frequency regulation and this by creating a reserve in power. This new control structure consisted of an optimal quadratic regulator combined with integral action and a reference model on the outputs, on one hand and with a wind estimator on the other hand, in order to allow a fast tracking of the power reference. These results show best performance for the control strategy developed with the wind estimator based on the Kalman filter.

关键词： Wind turbine variable-speed two-mass model pitch angle LQ multimodel optimal control approach wind estimator Kalman filter newton algorithm

来源：评论

学校读者我要写书评

暂无评论

A Smooth Hidden Space Support Vector Machine

A Smooth Hidden Space Support Vector Machine

引用

9th International Conference on Natural Computation (ICNC)

作者： Liang, Jinjin Wu, De Xian Shiyou Univ Sch Math Sci Xian Peoples R China Xidian Univ Sch Comp Sci Xian Peoples R China

ISBN: (纸本)9781467347143

Applying the smoothing techniques to the support vector machine in the hidden space, a smooth hidden space support vector machine (SHSSVM) is presented with some distinct mathematical features, such as the strong convexity and infinite differentiability. Beyond that, SHSSVM broadens the area of admissible kernel functions, where any real-valued symmetry function can be used as the hidden function, including the Mercer kernels and their combinations. Firstly, the input data are transformed to the hidden space by a hidden function. Secondly, the smoothing technique is utilized to derive the unconstrained smooth model. Finally, the newton algorithm is introduced to figure out the optimal solution. The numerical experiments on benchmark data demonstrate that SHSSVM has much higher training accuracies than HSSVM and SSVM, but with much lower training time.

关键词： smoothing technique hidden space symmetry function hidden function slack vector newton algorithm

来源：评论

学校读者我要写书评

暂无评论

LANGEVIN-BASED STRATEGY FOR EFFICIENT PROPOSAL ADAPTATION IN POPULATION MONTE CARLO 44

LANGEVIN-BASED STRATEGY FOR EFFICIENT PROPOSAL ADAPTATION IN...

引用

44th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Elvira, Victor Chouzenoux, Emilie IMT Lille Douai Villeneuve Dascq France CNRS CRIStAL UMR 9189 Villeneuve Dascq France Univ Paris Est LIGM CNRS UMR 8049 Champs Sur Marne France CentraleSupelec INRIA Saclay CVN Gif Sur Yvette France

ISBN: (纸本)9781479981311

Population Monte Carlo (PMC) algorithms are a family of adaptive importance sampling (AIS) methods for approximating integrals in Bayesian inference. In this paper, we propose a novel PMC algorithm that combines recent advances in the AIS and the optimization literatures. In such a way, the proposal densities are adapted according to the past weighted samples via a local resampling that preserves the diversity, but we also exploit the geometry of the targeted distribution. A scaled Langevin strategy with newton-based scaling metric is retained for this purpose, allowing to adapt jointly the means and the covariances of the proposals, without needing to tune any extra parameter. The performance of the proposed technique is clearly superior in two numerical examples at the cost of a reasonable computational complexity increment.

关键词： Importance sampling Monte Carlo methods population Monte Carlo Langevin dynamics newton algorithm stochastic optimization

来源：评论

学校读者我要写书评

暂无评论

Harmonic detection for three-phase circuit based on the TLS estimation algorithm

Harmonic detection for three-phase circuit based on the TLS ...

引用

5th International Conference on Wavelet Analysis and Pattern Recognition

作者： Zhang, Jun-Min Liu, Kai-Pei S Cent Univ Nationalities Wuhan 430074 Peoples R China Wuhan Univ Wuhan 430072 Peoples R China

ISBN: (纸本)9781424410651

Based on the newton learning algorithm, a new algorithm for TLS estimation and its application in harmonic detection is presented in this paper It is shown that the LMS algorithm and Constrained Anti-Hebbian algorithm are two examples of the proposed algorithm in this paper The dynamic performance of the TLS algorithm is better than that of LMS and anti-Hebbian algorithms. And the APF based on TLS algorithm can be used under the conditions of three-phase unbalanced and nonlinear loads, can be applicable to detect harmonic, reactive and negative sequence currents respectively or the sum of ones. At the same time the smaller error than that in single-phase can be obtained.

关键词： newton algorithm TLS estimation adaptive harmonic detection

来源：评论

学校读者我要写书评

暂无评论

Power System State Estimation Based on Nonlinear programming

Power System State Estimation Based on Nonlinear programming

引用

11th International Conference on Electrical Machines and Systems

作者： Yang Yude Deng Lijun Guangxi Univ Dept Elect Engn Nanning 530004 Peoples R China

ISBN: (纸本)9787506292214

State estimation base on a nonlinear programming model is presented (NLSE), which is applied the vectorization mode. We choose the L2 norm estimation as object. The nonlinear programming with equality constraint introduced slack variables, can guarantee the robust of the algorithm and dispose the convergence problem. The symmetric coefficient matrix of correction equation can be used by apply the AMD reordering algorithm and LDLT algorithm on the solution, which can speed up the calculation striking. The whole model of nonlinear state estimation applies vectorization form, so the complexity extent is simplified and both versatility and maintainability of code are improved. Numerical simulations use IEEE14, IEEE57, IEEE118, IEEE300, N1047 system to validate the correctness of the proposed model and method.

关键词： State estimation newton algorithm Sparse technology Reordering algorithm Vectorization mode Weighted least squares (WLS) state estimation

来源：评论

学校读者我要写书评

暂无评论

Smoothing techniques in solving non-Lipschitz absolute value equations

引用

INTERNATIONAL JOURNAL OF COMPUTER MATHEMATICS 2023年第4期100卷 867-879页

作者： Yilmaz, Nurullah Sahiner, Ahmet Suleyman Demirel Univ Dept Math Isparta Turkiye

In this study, we define for the first time the non-Lipschitz generalization of absolute value equations and concentrate on solving the problem of non-Lipschitz absolute value equations based on smoothing techniques. Two different types of smoothing techniques which are local and global ones are considered in smoothing process of the problem. With the help of these smoothing techniques, the non-Lipschitz absolute value equations are reformulated as a family of parametrized smooth equations. Two new algorithms are developed to solve the problem by the help of smoothing functions. Finally, the numerical experiments have been performed to illustrate the efficiency of the new algorithms.

关键词： Smoothing techniques smoothing algorithm absolute value equations non-Lipschitz functions newton algorithm

来源：评论

学校读者我要写书评

暂无评论

Numerical conductivity reconstruction from partial interior current density information in three dimensions

引用

INVERSE PROBLEMS 2021年第10期37卷 105010-105010页

作者： Yazdanian, Hassan Knudsen, Kim KN Toosi Univ Technol Dept Biomed Engn Tehran Iran Tech Univ Denmark DTU Compute DK-2800 Lyngby Denmark

This paper focuses on conductivity imaging of a three dimensional object using interior current density information (ICDI). Applications include the emerging hybrid tomographic methods known as magnetic resonance electrical impedance tomography and current density impedance imaging, which potentially have both high contrast and high resolution. For all possible forms of ICDI data, the Frechet derivative of the map between conductivity and ICDI is derived. Then, an iterative reconstruction method is formulated based on the newton scheme. The method is implemented numerically and its properties are investigated on simulated data obtained from two different phantoms. The method is also benchmarked against the J-substitution method. We systematically study the possibilities, challenges, shortcomings, and artifacts due the different forms of full and partial ICDI data and one or several boundary conditions. The results establish that at least two components of two non-parallel interior current densities are required to obtain good reconstructions;this is an important outcome for conductivity imaging methods which use only one component of the magnetic field. The results hold promise for the near real-time and high resolution conductivity reconstruction in practical applications.

关键词： conductivity imaging hybrid tomography newton algorithm partial data

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：