可编程控制器在包装机上的应用

本文介绍了可编程控制器（PLC）在包装机上的应用．介绍了系统的硬件设计与软件开发．     

海洋机器人用水下电机的深水密封研究

本文研究了海洋机器人用水下电机各个部分的密封方法，同时介绍了一个先进的动密封装置以及检验密封性能的水压试验方法．经测试及实际使用表明该密封方法和动密封装置完全满足水下机器人在深水中运行的密封要求     

红外遥感用低温制冷器的进展

低温制冷器与红外探测器配套，是军事现代化、高新技术等急需发展的关键产品。国内研制技术也已接近国际先进水平，基本掌握了小型化、低振动、长寿命、细方形漆包线制作、间隙密封、消除电磁干扰等相关关键技术。本文列出了已研制过的直线电机谐振压缩机的主要技术性能。希望主管、用户、研制、生产各方通力合作，应用－改进－再应用－再改进，实现我国低温制冷器的产业化。     

An automatic torque ripple compensation system for direct drive motors

A serious problem in motion control is the occurrence of torque ripple. Since a direct drive (DD) motor with a magnet rotor uses a rare-earth magnet, the torque ripple is more serious than that of conventional servomotors. This paper presents an auto-compensation of torque ripple using a software-implemented torque observer. The frequency and amplitude of torque ripple can be obtained in the controller by using the estimated torque from an observer, and the autonomous generation of a compensating current component for a torque ripple is possible. The proposed algorithm has been verified by experiments.     

我国车用汽油的发展概况及发展方向探讨

介绍了我国车用汽油的发展轨迹 ,通过与国外先进标准的对比分析了我国车用汽油存在的不足 .指出我国在加入WTO后 ,更应不断升级车用汽油质量标准 ,需要从标准制定、炼油技术等多方面追赶世界先进水平 ,积极研制生产清洁型无污染汽油 ,从根本上消除汽车尾气对环境的污染 .     

提高大直径井眼机械钻速的几项措施

随着我国西部地区石油勘探开发的进展，5000－7000m的深探井所钻数量不断增多，套管程序增加，特别是大尺寸井眼所钻深度明显加大。由于国产大尺寸钻头结构单一、型号少，破岩机械能量和水力能量不足，在易斜井段目前尚无可靠的防斜保直措施，以及钻井液性能不适宜等原因，大尺寸井眼的钻这远远低于215．9mm井眼的钻速。据此，提出了几项可以提高大尺寸井眼钻速的措施：增加大尺寸钻头的种类，采用阶梯式扩眼钻头，强化钻井参数和水力参数，选用中速动力钻具，改善钻井液性能以及采用冲击旋转钻具等。     

二阶A型Cooper变换的增强型

对二阶Ａ型Ｃｏｏｐｅｒ变换进行增强，放宽可用性条件，并证明了其正确性，最后举例说明其应用。     

A sensorless brushless dc motor system

This paper describes a brushless dc motor system without position or speed sensor. The brushless motor consists of a permanent magnet synchronous motor and a voltage-source inverter capable of controlling the amplitude and frequency of voltage. The rectangular-shaped stator current with a conducting interval of 120° (electrical) is controlled to be in phase with the trapezoidal back electromotive force. This results in producing maximum torque. Variable speed is achieved by adjusting the average motor voltage similarly to chopper control of dc motors. In this paper, two sensorless position detecting methods, i.e., an “indirect method” suited for the lower-speed range and a “direct method” suited for the higher-speed range are proposed. The combination of the two makes it possible to detect the rotor position over a wide-speed range. Furthermore, a speed-sen-sorless PLL control is proposed in applying the principle of the direct method. Experimental results obtained from a prototype brushless dc motor are shown to confirm the validity of the sensorless drive. The starting procedure of the motor also is discussed because it is impossible to detect the rotor position at a standstill.     

Numerical solution of the kinematic dynamo problem for Beltrami flows in a sphere

The article discusses computational aspects of the kinematic problem of magnetic field generation by a Beltrami flow in a sphere. Galerkin's method is applied with a functional basis consisting of Laplace operator eigenfunctions. Dominant eigenvalues of the magnetic induction operator and associated magnetic eigenmodes are obtained numerically for a certain Beltrami flow for magnetic Reynolds numbers up to 100. The eigenvalue problem is solved by a highly optimized iterative procedure, which is quite general and can be applied to numerical treatment of arbitrary linear stability problems.     

Post-Supervised Template Induction for Information Extraction from Lists and Tables in Dynamic Web Sources

Dynamic web sites commonly return information in the form of lists and tables. Although hand crafting an extraction program for a specific template is time-consuming but straightforward, it is desirable to automatically generate template extraction programs from examples of lists and tables in html documents. Supervised approaches have been shown to achieve high accuracy, but they require manual labelling of training examples, which is also time consuming. Fully unsupervised approaches, which extract rows and columns by detecting regularities in the data, cannot provide sufficient accuracy for practical domains. We describe a novel technique, Post-supervised Learning, which exploits unsupervised learning to avoid the need for training examples, while minimally involving the user to achieve high accuracy. We have developed unsupervised algorithms to extract the number of rows and adopted a dynamic programming algorithm for extracting columns. Our method achieves high performance with minimal user input compared to fully supervised techniques.