检索结果-内蒙古大学图书馆

Kernel matrix learning with a general regularized risk functional criterion

Journal of Systems engineering and Electronics 2010年第1期000卷

作者： Chengqun Wang1 Jiming Chen1 Chonghai Hu2 and Youxian Sun1 1. state key laboratory of industrial control technology department control science and engineering Zhejiang University Hangzhou 310027 P. R. China 2. department of Mathematics Zhejiang University Hangzhou 310027 P. R. China State Key Laboratory of Industrial Control Technology Department Control Science and EngineeringZhejiang UniversityHangzhou 310027P.R.China Department of Mathematics Zhejiang UniversityHangzhou 310027P. R. China

Kernel-based methods work by embedding the data into a feature space and then searching linear hypothesis among the embedding data points. The performance is mostly affected by which kernel is used. A promising way is to learn the kernel from the data automatically. A general regularized risk functional (RRF) criterion for kernel matrix learning is proposed. Compared with the RRF criterion, general RRF criterion takes into account the geometric distributions of the embedding data points. It is proven that the distance between different geometric distributions can be estimated by their centroid distance in the reproducing kernel Hilbert space. Using this criterion for kernel matrix learning leads to a convex quadratically constrained quadratic programming (QCQP) problem. For several commonly used loss functions, their mathematical formulations are given. Experiment results on a collection of benchmark data sets demonstrate the effectiveness of the proposed method.

关键词： kernel method support vector machine kernel matdx learning HKRS geometric distribution regularized risk functional criterion

来源：评论

学校读者我要写书评

暂无评论

Modeling of cortical signals using echo state networks

Modeling of cortical signals using echo state networks

引用

MIPPR 2009 - Pattern Recognition and Computer Vision: 6th International Symposium on Multispectral Image Processing and Pattern Recognition

作者： Zhou, Hanying Wang, Yongji Huang, Jiangshuai Key Lab. for Image Processing and Intelligent Control Department of Control Science and Engineering Huazhong University of Science and Technology China

ISBN: (纸本)9780819478078

Diverse modeling frameworks have been utilized with the ultimate goal of translating brain cortical signals into prediction of visible behavior. The inputs to these models are usually multidimensional neural recordings collected from relevant regions of a monkey's brain while the outputs are the associated behavior which is typically the 2-D or 3-D hand position of a primate. Here our task is to set up a proper model in order to figure out the move trajectories by input the neural signals which are simultaneously collected in the experiment. In this paper, we propose to use Echo state Networks (ESN) to map the neural firing activities into hand positions. ESN is a newly developed recurrent neural network(RNN) model. Besides its dynamic property and short term memory just as other recurrent neural networks have, it has a special echo state property which endows it with the ability to model nonlinear dynamic systems powerfully. What distinguished it from transitional recurrent neural networks most significantly is its special learning method. In this paper we train this net with a refined version of its typical training method and get a better model. © 2009 Copyright SPIE - The International Society for Optical engineering.

关键词： Brain

来源：评论

学校读者我要写书评

暂无评论

Modeling of Isomerization of C_8 Aromatics by Online Least Squares Support Vector Machine

引用

Chinese Journal of Chemical engineering 2009年第3期17卷 437-444页

作者：李丽娟苏宏业褚建 College of Automation and Electrical Engineering Nanjing University of Technology State Key Lab. of Industrial Control Technology Institute of Cyber-systems and ControlZhejiang University

The least squares support vector regression (LS-SVR) is usually used for the modeling of single output system, but it is not well suitable for the actual multi-input-multi-output system. The paper aims at the modeling of multi-output systems by LS-SVR. The multi-output LS-SVR is derived in detail. To avoid the inversion of large matrix, the recursive algorithm of the parameters is given, which makes the online algorithm of LS-SVR practical. Since the computing time increases with the number of training samples, the sparseness is studied based on the pro-jection of online LS-SVR. The residual of projection less than a threshold is omitted, so that a lot of samples are kept out of the training set and the sparseness is obtained. The standard LS-SVR, nonsparse online LS-SVR and sparse online LS-SVR with different threshold are used for modeling the isomerization of C8 aromatics. The root-mean-square-error (RMSE), number of support vectors and running time of three algorithms are compared and the result indicates that the performance of sparse online LS-SVR is more favorable.

关键词： least squares support vector machine multi-variable online sparseness isomerization

来源：评论

学校读者我要写书评

暂无评论

Soft Sensor for Distillation Column Feeds

Soft Sensor for Distillation Column Feeds

引用

2009年中国控制与决策会议(2009 Chinese control and Decision Conference)

作者： Lingyu Zhu Jixin Qian Zhijiang Shao Xi Chen State Key Laboratory of Industrial Control Technology Department of Control Science & EngineeringZ State Key Laboratory of Industrial Control Technology Department of Control Science & EngineeringZ

Distillation columns are normally controlled at a pre-designed condition, which is the optimal for a specified feed condition. In reality, however, the feeds may vary in their flow rates, temperatures, or compositions, which results in the distillation process operating at non-optimal values. Therefore, it is ideal to optimally adjust the operations in conjunction with the real feed variation. The online hardware measurement of the feed status, however, is basically impossible for most columns. A method using the temperature measurements along the distillation column to detect the feed variations is proposed in this paper. Verification is conducted through simulations under different conditions. Both derivative-based and derivative-free optimization methods are discussed for the soft-sensor establishment.

关键词： distillation column optimal operation feed variation temperature distribution soft-sensor.

来源：评论

学校读者我要写书评

暂无评论

Periodic Disturbance Rejection Method Using Iterative Learning Predictive Function control

Periodic Disturbance Rejection Method Using Iterative Learni...

引用

International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)

作者： Jing Wang Jun Zhao Zuhua Xu Jixin Qian State Key Lab of Industrial Control Technology Department of Control Science and Engineering University of Zhejiang Hangzhou China

In this paper, a robust control method is proposed for rejecting the periodic disturbance. Cooperating with predictive function control (PFC), iterative learning (IL) is used to deal with the period mismatch, that is, iterative learning control (ILC) is used to handle the periodic errors while PFC doing its job of system process control. Simulation results shows that the IL-PFC algorithm achieves significant improvement in rejecting periodic disturbance over the repetitive control scheme.

关键词： Iterative methods control systems Radio control Optical control Error correction Process control Signal generators Fuzzy systems industrial control Knowledge engineering

来源：评论

学校读者我要写书评

暂无评论

Multi-step prediction error approach for MPC performance monitoring

Multi-step prediction error approach for MPC performance mon...

引用

作者： Zhao, Yu Chu, Jian Su, Hongye Huang, Biao State Key Lab. of Industrial Control Technology Institute of Cyber Systems and Control Zhejiang University Hangzhou China Department of Chemical and Materials Engineering University of Alberta Edmonton AB T6G 2G6 Canada

ISBN: (纸本)9783902661548

Performance monitoring of model predictive control systems (MPC) has received a great interest from both academia and industry. In recent years some novel approaches for multivariate control performance monitoring have been developed without the requirement of process models or interactor matrices. Among them the prediction error approach has been shown to be a promising one, but it is k-step prediction based and may not be fully comparable with the MPC objective that is multi-step prediction based. This paper develops a multi-step prediction error approach for performance monitoring of model predictive control systems, and demonstrates its application in an industrial MPC performance monitoring and diagnosis problem.

关键词： Model predictive control

来源：评论

学校读者我要写书评

暂无评论

Flowrate measurement of air-water two-phase flow using an Electrical Resistance Tomography sensor and a Venturi meter

Flowrate measurement of air-water two-phase flow using an El...

引用

IEEE Instrumentation and Measurement technology Conference

作者： Zhenzhen Meng Zhiyao Huang Baoliang Wang Haifeng Ji Haiqing Li State Key Laboratory of Industrial Control Technology Department of Control Science and Engineering University of Zhejiang Hangzhou China

ISBN: (纸本)9781424433520

A flowrate measurement method for air-water two-phase flow is developed using an Electrical Resistance Tomography (ERT) sensor and a Venturi meter. With the conductance values measured by the ERT sensor, the real-time flow pattern is identified and the voidage is also calculated. The Butterworth correlation is applied to calculate the mass quality from the voidage value, and then six Venturi measurement correlations are introduced to predict the mass flowrate of two-phase flow. The flow pattern classifier is constructed by using the partial least square (PLS) method and the Least Squares Support Vector Machine (LS-SVM) classification method. The voidage measurement models are developed by using the PLS method and the LS-SVM regression method for different flow patterns. Experiment results verify that this measurement method is effective. The research work also shows that among the six Venturi measurement correlations the Collins correlation always performs the best for the flow condition in this work.

关键词： Electrical resistance measurement Fluid flow measurement Electric variables measurement Electric resistance Tomography Pressure measurement Electromagnetic measurements Least squares methods Electrodes Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Measurement system for characterization of gas-liquid two-phase flow in micro-channel

Measurement system for characterization of gas-liquid two-ph...

引用

IEEE Instrumentation and Measurement technology Conference

作者： Haifeng Ji Zhiyao Huang Baoliang Wang Haiqing Li State Key Laboratory of Industrial Control Technology Department of Control Science and Engineering University of Zhejiang Hangzhou China

ISBN: (纸本)9781424433520

A measurement system is developed for the parameter detection and flow characterization of gas-liquid two-phase flow in micro-channel. Electrical capacitance measurement system and optical measurement system are designed for the micro-channels with different inner diameters of 4.0 mm, 3.1 mm and 1.8 mm respectively. In experiment, bubble flow, stratified flow, slug flow and annular flow are observed. PSD (Power Spectrum Density) and PDF (Probability Density Function) methods are used to analyze the capacitance and optical signals. The results show that the stratified flow can not be observed in the pipe with inner diameter of 1.8 mm, and that the PSD and PDF curve can be used to characterize different flow regimes of gas-liquid two-phase flow in micro-channel. So the constructed measurement system is effective, and can be used to analyze flow characteristics and identify flow regime of gas-liquid two-phase flow in micro-channel.

关键词： Fluid flow measurement Capacitance measurement Image motion analysis Optical sensors Electric variables measurement Nitrogen Pressure measurement Optical design Microchannel Signal analysis

来源：评论

学校读者我要写书评

暂无评论

Quantitative determination of cyfluthrin in n-hexane by terahertz time-domain spectroscopy with chemometrics methods

Quantitative determination of cyfluthrin in n-hexane by tera...

引用

IEEE Instrumentation and Measurement technology Conference

作者： Yuefang Hua Hongjian Zhang Hongliang Zhou State Key Laboratory of Industrial Control Technology Department of Control Science and Engineering University of Zhejiang Hangzhou China

Terahertz time-domain spectroscopy (THz-TDS) was used for the detection of cyfluthrin content in N-hexane solvent with the concentration range of 0.5-10 μg/mL. The absorbance of the solution form cyfluthrin was obtained in the frequency range between 0.5-1.5 THz. Two kinds of multivariate linear regression models were then built between the absorbance and the concentration of 15 samples, using partial least squares (PLS) and principle component regression (PCR) methods. The number of factors leading to the least root-mean-square error of cross validation (RMSECV) was 2 for PCR and 9 for PLS respectively. Compared to the PCR model, the PLS model proves to be more effective, with the RMSECV as low as 0.565 μg/mL. Results show that the THz-TDS plus PLS method is very promising for the further pesticide residue detection in food safety control, with the characteristics of being nondestructive and laborsaving compared to other analytical tools.

关键词： Time domain analysis Spectroscopy Safety Least squares methods Chemical technology Submillimeter wave technology Solvents Linear regression Infrared spectra Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

Oil-water two-phase flow measurement based on a hybrid flowmeter and dominant phase identification

Oil-water two-phase flow measurement based on a hybrid flowm...

引用

IEEE Instrumentation and Measurement technology Conference

作者： Zhiyao Huang Xia Li Haifeng Ji Baoliang Wang Haiqing Li State Key Laboratory of Industrial Control Technology Department of Control Science and Engineering University of Zhejiang Hangzhou China

Based on a hybrid flowmeter and a dominant phase identifier, a new system is proposed for oil-water two-phase flow measurement. The hybrid flowmeter consists of an oval gear flowmeter and a Venturi meter. The dominant phase identifier is utilized to overcome the influence of oil fraction on the measurement results. In the practical measurement process, the total volume flowrate and the differential pressure of the two-phase flow are firstly obtained by the oval gear flowmeter and the Venturi meter, respectively. Meanwhile, the dominant phase of the oil-water two-phase flow is identified. Then the meter coefficient of the Venturi meter is selected according to the dominant phase identification result. Finally, the total volume flowrate and the total mass flowrate of the two-phase flow are obtained. The preliminary research results show that the proposed system is effective in measuring oil-water two-phase flow.

关键词： Fluid flow measurement Phase measurement Gears Fluid flow Petroleum Volume measurement Equations Chemical industry Instrumentation and measurement laboratories

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：