多变量加权多步预报控制*

对于线性多变量系统,本文给出了一种完全不同于以往的全状态反馈或观测器-控制器型的算法——多变量加权多步预报控制(MWLPC)算法。这种算法除引进了预测控制中的多步输出预报、滚动优化等机制外,最重要的是在二次型性能指标中引入了可调的多项式或有理分式矩阵权因子;适当选取这些权因子,便可按设计要求、仅用系统的输出信息反馈便能任意配置闭环系统的特征矩阵,从而保证闭环稳定性和其他优良性质。此外,该算法不改变原系统的零点,因而适用于非最小相位系统。     

专家系统应用于锅炉分气缸的设计和选型

基于专家系统的基本理论 ,根据锅炉分气缸设计、选型的特点和现场的实际条件 ,开发了锅炉分气缸设计、选型专家系统。该系统采用VisualC ++语言编程 ,模块结构 ,图形化用户界面等     

Characterization and catalytic gasification of the aqueous by-product from vacuum pyrolysis of biomass

Spruce wood residues were treated in a vacuum pyrolysis Process Development Unit with a throughput capacity of 28 kg/h. Two aqueous phase condensate samples with COD concentration varying between 190 and 255 g/L were produced and sequentially extracted with dichloromethane and ethylacetate solvents. The soluble organic matter was composed of acidic, phenolic, alcoholic and ketonic compounds. The insoluble fraction was sequentially distilled at 100 and 110°C under atmospheric pressure. Mainly water was recovered in the first distillate, while the second distillate contained 30.4% formic and acetic acids, 69.4% water and 0.2% residual organic compounds. The distillation residue was rich in oxygen and was essentially insoluble in any organic solvent. The two aqueous phase pyroligneous samples were treated in Bat-telle's Thermochemical Environmental Energy System (TEESr?), a registered service mark of Onsite*Ofsite, Inc. of Duarte, California, U.S.A. The results of the tests showed that similar results were obtained with either feedstock. In batch tests a COD reduction of 99% was achieved. The product gas composition was typically about 49% methane, 5% hydrogen, 1 % ethane and 45% carbon dioxide. Tests in a continuous stirred-tank reactor produced reproducible data which can be used for process scale-up. Catalyst lifetime was identified as needing further improvement. The preliminary results demonstrated the technical feasibility of the catalytic gasification process as a useful step in the recovery of energy from the secondary condensate stream and the cleanup of the by-product water from vacuum pyrolysis of wood.     

线性规划模型的自动生成与求解

本文描述了《有色企业经济活动分析决策支持系统》模型库中线性规划模型的自动生成与求解系统的设计思想、方法及实现     

预测控制算法的统一格式及其在电加热炉的应用

本文根据常用的两类预测控制算法推出一种新型预测控制算法。并将其应用于一双输入/双输出的力学持久机电加热炉的温度控制系统上,获得了满意的结果:控温精度为±1.5℃/300℃,且适用于非最小相位系统。     

Machine translation of non-literary texts: Some Canadian experiences

A survey was carried out of past and current Canadian experiences with machine-aided translation (MAT) systems of non-literary texts in one government institution and three industrial settings. Each organization requires a great deal of translation and had conducted several experiments in the use of MAT. The survey supports positive conclusions about the prospects for expanded use of computers in translation.The authors express their thanks to all those who helped in the preparation of this paper. We are particularly grateful to the staff of the translation departments of the organizations mentioned in the paper for their kind cooperation and their willingness to answer our questions: Chantal Delorme ofIBM Canada, Lorraine Desmarais and Mr. Plamondon of Bell Canada, Pierre Tellier of the Department of the Secretary of State, Maria Russo and Raymond Lapointe of the Xerox Corporation. Furthermore we thank Prof. Graeme Hirst for helpful discussion.     

利用增量矩阵辨识确定型ARMA模型的新算法

本文提出了一种无偏辨识确定型ARMA模型参数及阶次的新方法。和其他辨识算法不同的是在这里我们构造了关于系统开环阶跃响应的增量矩阵,并且通过分析该增量矩阵子矩阵的代数性质而得到了辨识系统阶次的一种简单方法。仿真的结果说明了本文理论的正确性。     

高精度节能型电力稳压器

分析了现有交流稳压器的特点和不足,提出了新一代数控稳压器的理论和实际无均差系统的方法,给出了超高精度无均差稳压系统的实现方法.经全乐科技公司模拟试验,获得了很好的结果.市电220.4 V和220.5 V,试验样机采用3 V、9 V和27 V这3个变压器,已达到优于±1%的精度,符合大部分负载的实际需要,经用户单位试用半年,效果非常满意.     

基于混合遗传算法的树状注水管网拓扑优化

以投资最小为目标函数,以井间隶属关系的惟一性、连接注水井数量等限制为约束条件,建立了树状注水管网拓扑优化数学模型.根据模型特点,将优化问题分为分配层和几何层,分别采用遗传算法和非线性优化方法进行求解.并对遗传算法的操作过程进行了改进,调整了适应函数,采用随机多父辈适应函数值加权交叉和多种变异操作,结合了模拟退火算法,并给出了初温的确定方法,改进了选择复制操作,从而形成了混合遗传算法.该算法能够有效地提高收敛速度,避免早熟收敛.利用算例说明了该方法的有效性.     

单输入中立型时滞系统的镇定

本文考虑了单输人中立型时滞系统的镇定问题。利用Byrnes等人提出的中立型时滞系统的模型,我们证明:如果系统是非奇异的,则系统可镇定的充要条件是其谱可镇定。