一类多测点系统的智能化决策方法

本文以人工智能原理和模糊推为基础，分析了一类多测点系统中决策测点的智能化方法，并给出实现测试系统的智能化决定策算法，实验结果表明，本文所述方法是成功的。     

一类大系统的稳定性

In this paper, We consider the stability of the equalibrium x = 0 of composite systemx = h(x, t)(3)with decomposition Xi= gi (xi,t) fi(x1,..., xr, t)(i = 1,..., r)( 4 )wherext = col (x1(i),...., x), i=1,..., r, n1 ... nr = n,under suitable assumptions for fi and gi, using the method of scalar Lyapunov function, we obtain the conditions under which the equilibrium x = 0 of Eq (3) is asymptotically stable or is unstable respectively.     

一类最小代价模糊决策系统及其算法

针对模糊决策系统在应用中的实际问题,提出一类最小代价模糊决策系统模型,定义了最优决策约简和最优决策代价,并对其性质进行分析。求解最优决策约简和最优决策代价是NP完全问题,为此给出基本算法、贪婪算法和基于拉格朗日松弛的子梯度优化算法,并进行实验分析。     

基于神经网络的一类大系统动态递阶优化方法

提出了一种用于求解离散时间大系统动态递阶优化问题的神经网络模型(LHONN),该网络以全集成化为特征:1)各子系统的动态方程嵌入相应的局部优化网络中,使得网络结构具有较低的维数,易于硬件实现;2)其上级协调网络和局部优化网络的求解过程同时进行,优化求解速度高,适宜于实时系统优化.     

一类非线性大系统递阶优化的神经网络方法*

本文给出了一种求解一类非线性大系统递阶优化问题的神经网络模型，克服了非线性大系统优化中的对偶间隙与不可分性问题，并且该神经网络具有全集成化的特点，易于硬件实现，其协调网络和局部优化网络同步工作，具有很高的求解效率，适宜于系统实时优化应用。     

一类非线性组合大系统的稳定性

本文研究一类比较广泛的非线性组合大系统的稳定性问题.基于矩阵代数Riccati方程存在对称正定解的新判据(引理2),给出了检验这类非线性组合大系统稳定性的新准则,所给出的稳定性准则只涉及子系统的稳定性以及子系统之间的互联项,因此,当应用该准则检验这类非线性组合大系统的稳定性时,避免了构造子系统以及大系统的Lyapunov函数的困难.     

一类动态大系统的协调控制模型

讨论了一类离散动态大系统的协调控制问题，提出了相应的模型辨识方法及自适应协调控制算法。通过仿真实例，说明效果是良好的。     

一类大时间滞后系统的预估

本文研究一类具有大纯时间滞后系统的预估.所述系统的动态特性能用包含中间测量的链形模型描述.在这种模型的基础上,作者构造了充分利用测量信息的预估算法.该算法已在一个化工过程控制中成功应用.     

一类时变离散大系统的稳定性

本文利用标量函数法研究了一类线性、非线性时变离散大系统的渐近稳定性及不稳定性的分解。同时得到了分解系数和非线性项界限的估计公式。     

关于大系统理论的几个问题

本文讨论关于大系统理论的几个问题,即:什么是大系统?大系统理论与控制理论的关系;大系统理论与系统工程的关系;大系统理论与控制论的关系;大系统理论的特殊问题. 作者认为,大系统理论是第三代控制理论的重要内容,是综合自动化的理论基础.另一方面,大系统理论也是系统工程发展的新阶段--动态系统工程.也可以说,大系统理论是控制论的新分支--大系统控制论.大系统理论不是关于系统大小的分类学,而是大规模复杂控制与信息系统分析和综合的方法学,是研究大系统的系统科学.     

Some Thoughts on Large Scale Systems in the Role of Cybernetics

One of the branches of research which has interested me recently is modeling global systems: how to develop computer representations of the world system and how to develop a planning and decision tool which decision makers, politicians and administrators can really use and believe in so that they can do something about forthcoming world problems. I am interested in this for two reasons:     

Rationality in Information Systems Support to Decision Making

Management information systems serve business organizations by providing information for decision making. Various types of systems serve different types of decision contexts. The philosophic basis of information system support is discussed. The rational (or normative) philosophy is widely used, and appears in business theory in the form of agency theory and transaction cost analysis. While this approach has been valuable in some contexts, there are other contexts where the rational approach has limited in utility for real business decision making. Decision makers need to consider subjective factors to enable them to cope with the high levels of uncertainty, incomplete understanding, and imperfect data typical of dynamic open systems. There are alternative philosophies upon which to base decision-making that are appropriate for specific decision contexts. Churchman identified empirical, multi-perspective frameworks, dialectic, and cause-and-effect inquiring systems as alternatives to the rational (normative) system. A number of information system tools, such as decision support systems, expert systems, and group support systems can be supported by models based on philosophies other than rational models. A more empirically based philosophy, with decision-makers balancing hypothesis generation and observations of performance, is often more appropriate. The relationship between Churchman's inquiring systems and information system types are discussed.     

不协调广义多尺度决策系统的尺度组合

针对不协调广义多尺度决策系统的知识获取问题,首先回顾广义多尺度信息系统中尺度组合的概念,给出在不同尺度组合下信息粒的表示及其相互关系.然后进一步定义在不同尺度组合下集合的下、上近似概念,给出近似集的性质.最后讨论不协调广义多尺度决策系统中的最优尺度组合的选择,并使用证据理论中的信任函数和似然函数刻画不协调广义多尺度决策系统中的最优尺度组合特征.     

大系统的分布式控制

本文根据分布式计算机系统的几种典型结构的特点,定义了几种新的大系统分布式控制信息结构.分布式控制规律用于分布式计算机控制系统时,既可提高系统的综合性能及整体最优性,又使各子系统间的通信次数很少,具有广阔的应用前景.     

企业经营管理的信息系统和决策咨询

在企业管理中,用计算机作为辅助手段是不是值得?能不能带来新的经济效益?为了探讨这一问题,下面从我国企业管理现行的信息系统和决策两个方面的现状出发,探讨计算机用于企业管理的必要性及可能性.     

大规模数据密集型系统中的去重查询优化

针对shared-nothing结构下大规模数据密集型系统去重查询的挑战,提出了一种有效的数据分布策略和并行处理方法分别对相关属性和无关属性去重进行优化:即自适应的散列和直方图相结合的数据分布策略,以及异步式并行查询中间件.前者在数据写入时保证数据均衡,并在数据量发生倾斜时自动调整数据的分布;后者充分发掘了去重查询处理中的粗粒度流水级并行,并消除了多节点同步等待的开销,尽早地返回结果.在生产系统DBroker上的测试表明,数据分布策略极大地改善相关属性的去重查询性能,而异步式并行查询引擎能够充分发掘并行性,对不相关属性的去重查询具有显著的性能提升.     

Composing Specific Domains for Large Scale Systems

DSM （domain-specific modeling） offers many advantages over general purpose modeling, but this type of modeling is effective just in narrow domains. The recent MDE （model driven engineering） approaches seek to provide a technology to compose different specific domains in order to cover large scale systems. In this context, this article proposes a new approach for composing specific domain models. First, we analyze some related works. On the basis of the key findings and conclusions drawn from the analysis, we propose a multidimensional approach based on the composition of crosscutting concerns contained in the source domain models. The approach is illustrated by a composition of service domains.     

Performance Properties of Large Scale Parallel Systems

There are several metrics that characterize the performance of a parallel system, such as parallel execution time, speedup, and efficiency. A number of properties of these metrics have been studied. For example, it is a well known fact that given a parallel architecture and a problem of a fixed size, the speedup of a parallel algorithm does not continue to increase with increasing number of processors. It usually tends to saturate or peak at a certain limit. Thus, it may not be useful to employ more than an optimal number of processors for solving a problem on a parallel computer. This optimal number of processors depends on the problem size, the parallel algorithm, and the parallel architecture. In this paper we study the impact of parallel processing overheads and the degree of concurrency of a parallel algorithm on the optimal number of processors to be used when the criterion for optimality is minimization of the parallel execution time. We then study a more general criterion of optimality and show how operating at the optimal point is equivalent to operating at a unique value of efficiency that is characteristic of the criterion of optimality and the properties of the parallel system under study. We put the technical results derived in this paper in perspective with similar results that have appeared in the literature before and show how this paper generalizes and/or extends these earlier results.