两状态非单调关联系统的形式理论

两状态单调关联系统理论已经成为可靠性数学理论的基础之一,正在向着多状态单调关 联系统理论的方向发展. 1977年Lapp&Powers提出非单调关联系统模型[1],引起可靠性学术界热烈讨论.一般 自动控制系统中存在着反馈回路,这导致系统的非单调关联性.为处理此类系统,本文推广 两状态单调关联系统理论的某些结果,提出两状态非单调关联系统的形式理论.     

非单调关联系统的可靠性计算

一般反馈控制系统往往都是非单调关联系统(NMS),NMS与单调关联系统(MS)具有 一系列不同的重要特性,如故障有序性,维修有序性等.本文给出了基于马尔可夫模型求一 般NMS首次故障时间分布的算法,对高维复杂NMS的分析具有优越性.     

多态关联系统可用度的计算模型

可用度是用来评估系统可用性的衡量指标。本文主要对多状态下的关联系统可用度计算模型进行了研究。首先对多状态单系统可用度计算方法进行了分析,然后在此基础上给出了多状态下串联系统、并联系统和k/m系统可用度的计算模型。     

基于扩张状态观测器的分散模型预测控制

针对复杂关联系统中分散控制方法无法有效解决子系统间的耦合和干扰问题, 提出一种基于扩张状态观测器的分散模型预测控制算法. 首先将复杂关联系统分解为多个状态维数较低、控制变量较少的子系统, 并为每个子系统设计本地预测控制器; 然后, 采用扩张状态观测器对子系统的耦合项以及干扰项进行估计, 进而利用估计值对子系统进行前馈补偿, 从而降低复杂关联系统的计算复杂度, 提高系统的稳定性和抗干扰能力; 最后, 利用液位控制系统验证了所提出算法的有效性.

     

基于有损网络的任意关联系统的分散控制

研究当子系统之间的通讯网络的信息传输存在随机丢包时, 在任意拓扑结构关联系统中的控制问题. 将网络通信状态的随机变量视为一种模型不确定性, 系统建模为关于确定性标称系统和一个随机模型不确定性的线性分式变换. 然后基于鲁棒控制理论对系统进行分析, 通过求解μ综合问题得到整个关联系统保持均方稳定所能容忍的最大丢包概率. 基于线性矩阵不等式设计了分散状态反馈控制器, 使得整个关联系统在某一给定的丢包概率下均方稳定. 最后通过一个例子验证该模型及算法的有效性.     

含任意拓扑和通信故障的关联系统分散控制

针对一类含任意拓扑结构连接以及不可靠通信连接而引起的关联系统子系统间信息交换故障情况问题,提出了一种关联系统分析以及分散控制器设计方法。该方法首先把关联系统子系统间通信故障的随机变量看作是关联系统模型不确定性来源,并且用分式线性变换形式建立关联系统。然后,借助鲁棒控制理论推导出在最大通信连接故障情况下关联系统均方稳定的充分条件。在给定某一通信故障率下,基于线性矩阵不等式理论,通过设计分散状态反馈控制器来保证整个大系统是均方稳定的。最后,通过中国南方电网局部缩减模型作为仿真实例说明所提出模型和方法的有效性。仿真结果表明,当南方电网局部电力元件之间的通信发生故障时使用本文的方法能够很好地进行电网设计和广域控制,并使整个局部电网是均方稳定的。     

一种多系统融合的电力移动巡检系统研究与设计

本文利用移动互联网、云计算和大数据技术,提出一种多系统融合的移动电力巡检系统。该系统通过云服务平台整合多个电力巡检关联系统,实现关联系统之间的实时数据交互,并依据变异系数对多系统融合的稳定性进行控制,以便实现系统资源的分配和调度,从而提供决策分析和预警。基于本框架的系统实现和实际应用测试结果表明,该系统能有效提高电力企业巡检的准确性和效率。     

时滞随机关联系统的群稳定性

在假定激励是参数白噪声的前提下, 基于箱体理论, 研究了无限维时滞随机关联系统中各子系统的内部联系. 利用向量Lyapunov 函数法, 研究了无限维时滞随机关联系统的群稳定性, 分别得到了无限维时滞非线性复合随机系统、无限维时滞弱耦合随机系统, 以及无限维时滞车辆跟随随机系统指数群稳定性的充分条件. 最后给出一个算例, 用以说明定理在实际中便于应用.     

光学元件管理系统的实现

为了提高管理工作的自动化和效率,针对高功率激光系统涉及的光学元件种类多、数量大、元件管理工作繁琐的实际情况,开发了光学元件管理系统.本系统可动态添加或删除元件的种类,允许自定义每种元件的参数类型.在提供对参数进行方便的录入和修改的同时,可对所有参数进行检索,并能够对元件的外借和归还进行简便的管理,还可以察看所有光学元件的公共状态和元件的使用记录,也具有打印和备份功能.     

一种基于Petri网的可靠性分析方法

冲突和并发是Petri网的两种典型的行为,本文基于系统可靠性Petri网模型的逆模型和基于ECS的解冲突算法,得到一种新的求解单调关联系统最小割集的算法,另外,根据这一算法,能够获得与系统当前状态有关的所有可能的演化。     

Variable-structure coherent systems†

The notion of coherent systems plays an essential role in conventional reliability theory. A system is said to be coherent if all of its components are relevant and the system reliability is improved as the component reliabilities are improved. However, in many complex systems or networks, not all the components are unconditionally relevant. As a result, in this paper we introduce the notion of variable-structure coherent systems to describe those systems that extensively exist and demonstrate essentially distinct features not observed in conventional coherent systems. A variable-structure coherent system consists of a number of substructures that are each a coherent system in conventional sense themselves. We then analyze the structural properties of variable-structure coherent systems; define the system operational profile, the system reliability, and the system structural profile. We study the system life distribution, the substructure importance, and the component importance. Finally, we deal with phase-cyclic systems in the context of variable-structure coherent systems.     

Positive invariance, monotonicity and comparison of nonlinear systems

In this paper, the interrelation between positive invariance, monotonicity and comparison of iterated nonlinear systems defined in partially ordered sets is studied. First, necessary and sufficient conditions guaranteeing the positive invariance of sets defined by relations of the form v(x)≤w with respect to nonlinear systems are established. Then, various characterizations of monotone nonlinear systems are developed and a connection between positive invariance and monotonicity is established. Finally, necessary and sufficient conditions for a, not necessarily monotone, nonlinear system to be a comparison system are established. No specific algebraic structure is necessary for developing the main results of this work. Numerical examples illustrating the applicability of all these results to the case of discrete-time dynamical systems are also given.     

离散非线性时变凸多面体系统族的鲁棒正不变集

动态系统的状态约束和控制约束等问题可归结为状态空间中某些集合的正不变性.利用 混合单调分解方法研究离散非线性、时变凸多面体系统族的线性状态约束集合的鲁棒正不变性． 对由矩阵凸多面体和加性区间扰动描述的线性时变离散系统族,得到了鲁棒正不变集的充分必要 条件;对非线性系统族则得到有关充分条件.这些条件均由系统族的顶点表述,易于检验,同时给 出示例.     

二阶多智能体系统在自适应间歇牵制控制下的部分分量一致性

部分分量一致指的是多智能体系统中所有状态变量的一些分量渐近趋于恒同这样一种现象,它是一种比恒同一致弱的群体动力学行为.基于二阶多智能体系统的动力学模型,设计一种自适应间歇牵制控制协议,通过置换矩阵方法将原偏差系统中待研究的状态分量转换为新偏差系统中前面部分的状态分量,并运用矩阵理论和部分变元稳定性理论,导出部分分量一致性准则,从而确保在指数稳定意义下该多智能体系统的部分分量一致性得以实现.最后,通过数值模拟验证理论分析结果.     

基于独立成分分析的状态监控方法

针对目前部分的工业系统存在系统模型或系统独立变量不确定的情况,提出了一种基于独立成分分析(Independent Component Analysis,ICA)的状态监控方法。该方法不需要有关系统模型和系统变量的先验知识,解决了系统模型或系统变量不确定的问题。运用独立成分分析法对系统的观测信号进行特征提取,通过监控分离得到的特征信号来实现对整个系统的状态进行监控。分别对一个离散系统和一个双容水箱液位系统进行仿真验证,仿真结果表明该方法能够有效地实现系统监控。     

Attribute reduction in ordered information systems based on evidence theory

Attribute reduction is one of the most important problems in rough set theory. However, in real-world lots of information systems are based on dominance relation in stead of the classical equivalence relation because of various factors. The ordering properties of attributes play a crucial role in those systems. To acquire brief decision rules from the systems, attribute reductions are needed. This paper deals with attribute reduction in ordered information systems based on evidence theory. The concepts of plausibility and belief consistent sets as well as plausibility and belief reducts in ordered information systems are introduced. It is proved that a plausibility consistent set must be a consistent set and an attribute set is a belief reduct if and only if it is a classical reduction in ordered information system.     

A theoretical framework for multiple neural network systems

Multiple neural network systems have become popular techniques for tackling complex tasks, often giving improved performance compared to single network systems. For example, modular systems can provide improvements in generalisation through task decomposition, whereas multiple classifier and regressor systems typically improve generalisation through the ensemble combination of redundant networks. Whilst there has been significant focus on understanding the theoretical properties of some of these multi-net systems, particularly ensemble systems, there has been little theoretical work on understanding the properties of the generic combination of networks, important in developing more complex systems, perhaps even those a step closer to their biological counterparts. In this article, we provide a formal framework in which the generic combination of neural networks can be described, and in which the properties of the system can be rigorously analysed. We achieve this by describing multi-net systems in terms of partially ordered sets and state transition systems. By way of example, we explore an abstract version of learning applied to a generic multi-net system that can combine an arbitrary number of networks in sequence and in parallel. By using the framework we show with a constructive proof that, under specific conditions, if it is possible to train the generic system, then training can be achieved by the abstract technique described.     

基于Vague集的电子商务推荐系统研究

基于内容推荐系统发展的瓶颈在于商品特征的表示。借助Vague集方法,并利用高斯函数和Vague集中未知度的一些性质,将商品的特征用Vague值形象表示。在此基础上,给出了运用Vague集方法进行基于内容推荐的步骤,为推荐系统的设计提供了一种新的指导思想和方法。最后给出了多个推荐公式,并进行简单的比较,有利于在实际推荐工作中酌情选择不同的推荐公式,提高商品推荐质量和推荐精度。