分布式随机离散事件系统的模式故障预测EI北大核心CSCD

本文以随机自动机为模型,研究分布式随机离散事件系统的模式故障预测问题.首先根据分布式随机离散事件系统的分布特性对模式故障协同可预测性概念进行形式化,并通过构造模式故障识别器来识别系统中发生的模式故障.然后,构造一个模式故障协同预测器,提出一种基于协同预测器的具有多项式复杂度的算法,得到关于分布式随机离散事件系统模式故障协同可预测性的充分必要条件,实现对分布式随机离散事件系统的模式故障预测.     

基于不确定观测下离散事件系统可诊断性的研究

在实际应用系统中,由于传感器故障、传感器限制和网络中的数据包丢失等原因,事件的可观测值变得不确定,使得观测系统行为变得尤为复杂。针对离散事件系统中,同个事件串可能有多个观测值以及不同状态下同个事件观测值也可能不同的问题,提出一种不确定观测下故障诊断验证的方法。首先对不确定观测的离散事件系统的可诊断性进行形式化,然后构建出用于上述故障诊断验证的验证器;基于验证器提出了系统基于不确定观测下可诊断的充要条件及验证算法;最后,实例说明不确定观测下故障诊断验证算法的应用。与现有研究相比,提出的方法对故障事件的观测值没有约束,可以为0个或多个观测值,使此方法应用的场景更为广泛。     

分布式模糊离散事件系统的故障预测

本文研究分布式模糊离散事件系统的故障预测问题.先根据系统的模糊特性,提出一种分布式模糊离散事件系统的协同可预测性的形式化方法,使分布式模糊离散事件系统的协同可预测度不小于各分站点的局部可预测度.通过构造协同预测验证器,提出一种基于协同预测验证器的协同预测算法,并得到一个关于分布式模糊离散事件系统协同可预测性的充分必要条件.     

基于离散HSMM的故障预测模型

提出了一种基于离散HSMM的故障预测模型,根据部分观测矢量预测系统下一时刻处于各个状态的概率。结合HSMM的前向—后向(FB)算法,给出了部分观测下HSMM的状态预测算法。将提出的模型应用于减速箱故障预测中,结果表明该方法可以有效地进行故障预测。     

带乘性噪声和延时观测广义系统的多步最优预测器

本文研究了状态空间描述的离散广义系统最优预测器的设计问题,该系统带有即时和延时观测,所有观测中带有乘性噪声.论文在两个基本假设条件下采用标准的奇异值分解方法给出了受限等价时滞系统,对于此类系统没有采用状态增广方法,而是采用新息重组分析理论给出了多步预测器.因为延时观测的存在,所给出的多步预测器包含了两套递推的广义系统黎卡提方程.本文给出了一个数学算例验证了所提方法的正确性和有效性,并给出了四幅图片,根据算例可以看出一般情况下预测的步数越少,预测的结果越好.本文方法可以进一步来研究更复杂的一些问题,如延时广义系统的H_∞滤波和控制问题.     

基于动态观测的随机离散事件系统故障诊断

近年来,离散事件系统故障诊断研究引起国内外学者广泛关注.鉴于此,研究动态观测下随机离散事件系统的故障诊断.首先引入一种动态观测,使事件的可观测性随着系统的运行而动态变化;然后分别对基于动态观测的随机离散事件系统的单故障可诊断性和模式故障可诊断性进行形式化;最后通过构造相应的诊断器,分别得到关于单故障可诊断性和模式故障可...     

基于非策略Q-学习的网络控制系统最优跟踪控制

针对具有数据包丢失的网络化控制系统跟踪控制问题,提出一种非策略Q-学习方法,完全利用可测数据,在系统模型参数未知并且网络通信存在数据丢失的情况下,实现系统以近似最优的方式跟踪目标.首先,刻画具有数据包丢失的网络控制系统,提出线性离散网络控制系统跟踪控制问题;然后,设计一个Smith预测器补偿数据包丢失对网络控制系统性能的影响,构建具有数据包丢失补偿的网络控制系统最优跟踪控制问题;最后,融合动态规划和强化学习方法,提出一种非策略Q-学习算法.算法的优点是:不要求系统模型参数已知,利用网络控制系统可测数据,学习基于预测器状态反馈的最优跟踪控制策略;并且该算法能够保证基于Q-函数的迭代Bellman方程解的无偏性.通过仿真验证所提方法的有效性.     

基于AR 模型的预测算法在力觉临场感遥控机器人系统中的应用

为解决力觉临场感遥控机器人系统的时延问题 ,本论文提出了将预测器引入该系统的新思路 .该预测器是基于AR模型对从手与环境间的作用力进行预测的 .本论文模拟了具有预测器的单自由度力觉临场感遥控机器人系统 ,并对该预测器的算法进行了仿真研究 .在时延条件下 ,该思路可实现对从机械手较为准确的控制     

测量数据丢失的随机不确定系统鲁棒滤波递推算法

针对一类具有测量数据丢失的不确定离散随机系统,研究了鲁棒状态估计问题,基于间断观测滤波算法和规则最小二乘优化理论,给出一种Kalman形式的递推滤波算法.对于测量数据丢失的问题,采用已知概率的Bernoulli随机序列,使得对于所有可能的测量数据丢失和所能容许的不确定性,间断观测鲁棒状态估计递推算法是稳定的.最后,通过数值仿真和对比结果验证了所提出算法的可行性.     

基于SVR的瓦斯传感器故障诊断方法

介绍了回归型支持向量机（SVR）的基本原理,建立了基于SVR的传感器时间预测模型,对时间预测器实行离线训练、在线应用的方法,用训练好的SVR模型模拟煤矿井下瓦斯传感器系统的动态特性,阐述了瓦斯传感器故障诊断和信号恢复的实现过程。仿真结果表明：SVR时间预测器能准确预测和跟踪瓦斯传感器的输出,及时诊断出传感器故障信息,并对传感器信号进行恢复,实验验证了该方法的正确性和有效性。     

Decentralized Diagnosis of Stochastic Discrete Event Systems

We investigate the decentralized diagnosis of stochastic discrete event systems (SDESs) by using multiple local stochastic diagnosers, each possessing its own sensors to deal with different information. We formalize the notions of decentralized diagnosis for SDESs by defining the concept of codiagnosability for stochastic automata, in which any communication among the local stochastic diagnosers or to any coordinators is not involved. These notions are weaker than the corresponding notions of decentralized diagnosis of classical DESs. A stochastic system being codiagnosable means that a fault can be detected by at least one local stochastic diagnoser within a finite delay. We construct a codiagnoser from a given stochastic system with a finite number of projections whose each diagnosis component uses the complete model of the system. We also deal with a number of basic properties of the codiagnoser. In particular, a necessary and sufficient condition of the codiagnosability for SDESs is presented, which generalizes the corresponding results of centralized diagnosis for SDESs. Also, we give a computing method in detail to check the codiagnosability of SDESs. As an application of our results, some examples are described.     

Safe Diagnosability of Stochastic Discrete Event Systems

Recently, safe diagnosability of discrete event systems (DESs) was investigated by Paoli and Lafortune, which was viewed as the first necessary step of fault-tolerant supervision. In this paper, we consider the problem of safe diagnosability in the framework of stochastic discrete event systems (SDESs). We define the notion of safe diagnosability for stochastic automata, in which fault detection occurs before any given forbidden string in the failed mode of system is executed. The relationship between diagnosability and safe diagnosability for SDESs is analyzed. In particular, a necessary and sufficient condition for safe diagnosability of SDESs is presented by constructing the recognizer of illegal language and the safe diagnoser. Some examples are described to illustrate the results.      

Fault estimation and prediction for nonlinear stochastic system with intermittent observations

This paper is concerned with the fault estimation and prediction problems for a class of nonlinear stochastic systems with intermittent observations. Based on the extended Kalman filter and Kalman filter, the fault and state are simultaneously estimated, and then, it is extended to the case of intermittent observations. Meanwhile, the boundedness of the estimation error is also discussed. Once the fault is detected, the parameters of each fault are identified by the linear regression method. Then, the future fault signal can be predicted by the parameters of the fault. The effectiveness of the proposed algorithm is verified by the simulation of the 3‐tank system.     

Sufficiency for diagnosability of stochastic discrete-event systems and a polynomial-time verification

The notion of diagnosability of stochastic discrete-event systems (SDESs) was introduced in the literature, but the method for verifying the diagnosability of SDESs was exponential. However, is there a sufficient condition of diagnosability of SDESs that can be tested with polynomial method? In this paper, we present a sufficient condition of diagnosability of SDESs based on a constructive testing automaton, and a polynomial method is proposed to test the condition.     

改进SOM神经网络在电力调度故障诊断中的应用

为了解决电力调度自动化系统中故障、安全监测不到位,尤其是缺少精确定位和关联分析等问题,利用改进的SOM神经网络提出了一种故障诊断模型.首先,在分析调度系统历史数据基础上,提取故障的特征向量,建立学习样本.接着通过算法训练输入和输出间的内在联系,供后续测试验证使用.最后,在已具备数据内在映射关系的网络中,测试待检测数据,验证其故障诊断的效果.最后的结果表明,此模型对不同类型故障识别和诊断能力较强,是一种行之有效的人工智能诊断方法.     

Time domain single-phase reclosure scheme for transmission lines based on dual-Gaussian mixture models

The basic principle of new adaptive reclosures are to first identify whether a fault is transient or permanent and consequently to determine the reclosing moment. In this paper a novel method to enhance self-adaptive single phase autoreclosure of transmission lines is presented. Using Gaussian Mixture Models (GMM) the redundancy of setting the threshold is omitted. The proposed algorithm could prevent closing command in permanent faults and adapt dead time in temporary events. The method is derived by processing line terminal voltage around the period of dead time. The proposed scheme uses two sampled windows from the inception of the fault and two groups of GMM. Simulations performed in EMTP/ATP environment advocate the validity of the proposed algorithm convergence speed as well as fast and accurate protection scheme for reclosing relaying. The design of GMM is easy and the relative factors of the structure elements can be regulated due to the desirable effects. Since the discrimination method is done with stochastic characteristics of signals in time domain without application of any deterministic index, more reliable and accurate classification is achieved.     

Fault detection for a class of nonlinear stochastic systems with Markovian switching and mixed time-delays

This article is concerned with the fault detection (FD) problem for a class of nonlinear stochastic systems with Markovian switching and mixed time-delays. The stochastic system under consideration contains discrete and distributed mode-dependent delays, nonlinearities as well as Markovian switching. Considering a new sensor fault model, which can express the failures of the loss of effectiveness and the outage, a novel FD scheme is proposed such that it is valid for this class of sensor faults. In addition, delay-dependent conditions, which include some existing results, that capture the sensitivity performance and attenuation performance are derived. Then the FD filter gains are characterised in terms of the solutions to a convex optimisation problem. Finally, an aircraft application is given to demonstrate the effectiveness of the proposed method.     

基于模糊贴近度的粒子滤波故障预测

复杂设备的故障特征具有不确定性,非线性等特点,为预防故障可能造成的严重后果,提高故障预测准确性是非常必要的.针对故障预测具有不确定性的特点,本文将模糊数学中的模糊贴近度和粒子滤波算法相结合设计故障预测的方法.新方法利用隶属度函数设计了描述系统运行正常的正常模糊子集和运行异常的异常模糊子集,利用粒子滤波算法计算系统运行的预测值,并计算预测值的正常隶属度;再分别计算预测值的正常隶属度与正常模糊子集和异常模糊子集的贴近程度来实现故障预报.该方法通过三容水箱系统T2水箱水位变化预测三容水箱系统是否出现故障和通过UH-60行星齿轮盘裂纹何时开始增大的故障进行实验,并同基于改进余弦相似度的粒子滤波故障预报、基于随机摄动粒子滤波器的故障预报算法和基于粒子滤波的FDI方法进行了对比.实验验证了该方法的可行性,可及时准确地预测出系统故障.     

基于交互容积卡尔曼的随机切换系统故障诊断

针对一类随机切换非线性系统的故障检测和故障估计问题,提出了一种基于交互式多模型和容积卡尔曼滤波（IMM CKF）的系统状态估计算法。该算法利用容积卡尔曼滤波（CKF）在不同时刻对每个子系统进行状态估计,把不同子系统状态估计结果融合得到最终的状态估计,实现对系统真实状态的估计。针对一类随机切换非线性系统发生执行器故障,采用IMM CKF估计系统状态;然后分析了IMM CKF算法的稳定性;根据状态估计结果,构造残差信号,设计残差评价函数,检测故障发生。当检测到故障发生时,设计增广系统,对故障幅值进行估计。通过仿真实验验证提出算法的有效性,结果表明该算法可以较为准确地诊断系统故障。