Position control of linear switched reluctance motors for high-precision applications

Most advanced manufacturing processes require precise motions for material transfer, packaging, assembly, and electrical wiring. To achieve precise linear motions, most of these high-performance manufacturing machines use X-Y sliding tables with permanent-magnet rotary motors and rotary to linear couplers. Though this method is the most widely used, it has disadvantages of low accuracy, complex mechanical adjustments, high cost, and low reliability. This paper describes the position control of a linear switched reluctance motor for high-performance motions in manufacturing automation. The proposed actuator has a very simple structure and it can be manufactured easily. There is no need for magnets and no limitation on the travel distance. The actuator is extremely robust and can be used in a hostile environment. A novel current-force-position lookup table is first developed to perform the force linearization. Then, a plug-in robust compensator using H/sub /spl infin// loop-shaping design is employed to improve the system robustness and the tracking performance. Experimental results of the motion system indicate that the system has fast tracking responses with good accuracy.     

模型库管理系统的设计和实现*

模型库管理系统是决策支持系统的核心，其基本功能包括模型的定义、组合、存储和调用.该文介绍一个支持客户/服务器模式的模型库管理系统的设计和实现，在这个系统中，模型被视为程序模块，通过运用面向对象的方法进行组织以及对模型的远程和动态的透明调用提供支持，使实现的系统具有良好的动态可扩充性.同时,通过提供模型定义语言和模型库管理，便于决策支持系统客户软件的开发.     

Delivery of video mail on the World Wide Web

This study was initiated to examine the feasibility of providing multimedia mail, in particular video mail, on the existing World Wide Web platform. In this instance, the Web browser will act as the user agent of the mailing system. With this, users will be able to compose, send and read video mail directly on the Web. This paper examines a number of possible system architectures that can be employed for the development of video mail. Based on the following key issues: portability of system, flexibility of system, Internet standard support, security and authentication, and message compression and storage, a final architecture was selected to build video mail. The approach taken is that the system will be interfaced to the Web using a standard HyperText Transfer Protocol (HTTP) server and a standard Web browser. This is achieved through the use of Common Gateway Interface (CGI) programs and forms which reside in a standard HTTP server to provide the functionality of video mail, and a separate mail server for the storage and management of messages. This will cause the HTTP server to behave as if it is an electronic mail (E-mail) system. Video mail has been successfully implemented at the School of Applied Science, Nanyang Technological University, Singapore. The development work has been carried out on a Sun SPARC5 workstation and the Typhoon relational database system was used to construct the mail server. All software codes were written in C or C++. A video capturing application used in conjunction with a SunVideo card has been developed for video and audio recording. The paper also addresses implementation issues pertaining to the use of CGI programs, mail server development, and video and audio management.     

Extended latent class models for collaborative recommendation

With the advent of the World Wide Web, providing just-in-time personalized product recommendations to customers now becomes possible. Collaborative recommender systems utilize correlation between customer preference ratings to identify "like-minded" customers and predict their product preference. One factor determining the success of the recommender systems is the prediction accuracy, which in many cases is limited by lacking adequate ratings (the sparsity problem). Recently, the use of latent class model (LCM) has been proposed to alleviate this problem. In this paper, we first study how the LCM can be extended to handle customers and products outside the training set. In addition, we propose the use of a pair of LCMs (called dual latent class model-DLCM), instead of a single LCM, to model customers' likes and dislikes separately for enhancing the prediction accuracy. Experimental results based on the EachMovie dataset show that DLCM outperforms both LCM and the conventional correlation-based method when the available ratings are sparse.     

Optimization of photo‐aligned ferroelectric liquid‐crystal display under passively addressed driving

Abstract— A ferroelectric liquid‐crystal (FLC) passively addressed 64 × 64 display based on the photo‐alignment technique has been developed. The display matrix has dimensions of 33 × 33 mm², and the FLC layer thickness is about 5 μm. Asymmetric boundary conditions, when only one of ITO surfaces of the display matrix is covered with the photo‐aligning layer while another one is not, have been used for providing both high contrast ratio and steady multiplex operation. The electro‐optical performance of the 5‐μm FLC display is presented, including bistable switching in static operation, optimization in multiplexing operation, and gray‐scale generation.     

The effect of pitch and collimation on radiation dose in spiral CT

1 Introduction Spiral computed tomography (CT) was intro- duced in clinical practice in 1980s. Its distinctive characteristic is continuous data acquisition and con- tinuous tube rotation with high speed while simulta- neously translating the patient through the gantry.[1] Owing to these functions the spiral CT possesses the ability to scan an entire organ volume within a single breathhold. Thus, the anatomical section is contiguous and therefore the reliable demonstration of small le- sions…     

Uncertainty Management in the Unit Commitment Problem

Uncertainty in power systems operations has been traditionally managed by multistage decision making and operating reserve requirements. A familiar example of multistage decisions is day-ahead unit commitment and real-time economic dispatch. An alternate approach for managing uncertainty is a stochastic formulation, which allows the explicit modeling of the sources of uncertainty. This paper compares stochastic and reserve methods and evaluates the benefits of a combined approach for the efficient management of uncertainty in the unit commitment problem. Numerical studies show that unit commitment solutions obtained for the combined approach are robust and superior with respect to the traditional approach in terms of both economics and reliability metrics.     

基于相关向量机的短期风速预测模型

通过对风速的时间序列进行分析,表明该序列具有混沌特性。在此基础上,利用相空间重构理论建立基于相关向量机（RVM）的短期风速预测模型,并对不同的核函数进行分析,选出最优的核函数。与现有的风速预测模型相比,该模型具有高稀疏性、核函数选择灵活等优点。仿真结果表明,与BP神经网络和支持向量机（SVM）模型相比,RVM模型预测精度更高。     

WEB INFORMATION MONITORING FOR COMPETITIVE INTELLIGENCE

The WWW has become one of the most important media for sharing information. Web information provides another emerging and important avenue and source of competitive intelligence (CI) for companies. CI is critical for companies to stay competitive in the marketplace. Apart from business users, there are other types of CI users such as technical users, casual users, news awareness users and others who would like to be kept informed on the latest development of their interested areas over the WWW. To discover web information, CI users need to constantly monitor certain web sites and web pages for related information. However, the dynamic nature of the web has made such monitoring task complicated and time-consuming. This paper proposes a web monitoring system, WebMon, to help users monitor specified web pages for latest changes and updates in information. Four monitoring functions including date monitoring, keywords monitoring, link monitoring and portion monitoring are supported by the system. The performance of these monitoring functions is also evaluated.     

Sparse and Truncated Nuclear Norm Based Tensor Completion

One of the main difficulties in tensor completion is the calculation of the tensor rank. Recently a tensor nuclear norm, which is equal to the weighted sum of matrix nuclear norms of all unfoldings of the tensor, was proposed to address this issue. However, in this approach, all the singular values are minimized simultaneously. Hence the tensor rank may not be well approximated. In addition, many existing algorithms ignore the structural information of the tensor. This paper presents a tensor completion algorithm based on the proposed tensor truncated nuclear norm, which is superior to the traditional tensor nuclear norm. Furthermore, to maintain the structural information, a sparse regularization term, defined in the transform domain, is added into the objective function. Experimental results showed that our proposed algorithm outperforms several state-of-the-art tensor completion schemes.