基于递推规范变量残差和核主元分析的微小故障检测

微小故障由于故障征兆不明显从而很难在故障发生早期对其进行检测. 针对该问题, 本文提出了一种基于递推规范变量残差和核主元分析(RCVD–KPCA)的微小故障检测方法. 首先构造规范变量残差, 从中提取数据的线性特征. 利用指数加权滑动平均法对规范变量残差进行递推滤波处理, 提高规范变量残差对微小故障的敏感程度;然后使用KPCA提取规范变量残差中的非线性主成分作为非线性特征, 根据提取的特征提出了两个新的故障检测统计量; 此外, 利用核密度估计确定故障检测统计量的控制限. 由于同时提取了过程数据的线性和非线性特征, 有效地提高了非线性动态过程中微小故障的可检测性. 以闭环连续搅拌釜式反应器过程为例进行了仿真分析, 仿真结果表明本文所提方法具有较好的故障检测性能.     

Modified non-Gaussian multivariate statistical process monitoring based on the Gaussian distribution transformation

Independent component analysis (ICA) has been applied for non-Gaussian multivariate statistical process monitoring (MSPM) for several years. As the independent components do not satisfy the multivariate Gaussian distribution, a missed alarm occurs when monitoring with traditional statistics. In this paper, we propose a Gaussian distribution transformation (GDT)-based monitoring method. Independent components are first transformed into approximate Gaussian distributions through the proposed nonlinear mapping. Then, we propose new statistics and their control limits to reduce missed alarms. The proposed method is particularly suitable for slight magnitude fault and early-stage fault detection. The ratio part of the area above the curve (RPAAC) is developed to evaluate the performance in fault detection. The experimental results from a synthetic example show the effectiveness of our proposed method. We also apply our method to monitor an electrical fused magnesia furnace (EFMF), and eruption and furnace wall melt faults can be detected in time.     

基于Bagging-CVA的动态过程及质量相关故障检测

针对过程数据具有时序相关性以及过程故障是否影响产品质量的问题,提出一种基于Bagging思想和典型变量分析(CVA)的故障检测方法(Bagging-CVA).采用Bagging思想对建模数据随机抽样构成多组新的数据集,消除数据的时序相关性.分别在每组新的数据集基于CVA方法建立过程相关和质量相关的故障检测模型,同时监测...     

Dimension reduction method of independent component analysis for process monitoring based on minimum mean square error

This article proposes a novel dimension reduction method of independent component analysis for process monitoring based on minimum mean square error (MSE). Firstly, the order of the independent components (ICs) is ranked according to their importance estimated by MSE, and the mathematical proof is presented. Secondly, the top-n ICs are selected as dominant components and the dimension of ICs is reduced. The sum of the squared independent scores (I²) and the squared prediction error (SPE) are adopted as monitoring statistics. The control limits of I² and SPE are determined by the kernel density estimation (KDE). The proposed dimension reduction method is applied to fault detection in a simple multivariate process and the simulation benchmark of Tennessee Eastman process. Finally, two fault conditions of pulverizing system in power plant are analyzed by the proposed method. The experiments results verify the effectiveness of the proposed method.     

基于概率密度PCA的多模态过程故障检测

为了提高故障检测和分类能力,提出基于概率密度PCA的多模态过程故障检测算法。对各模态的训练数据建立PCA模型,计算各个模型的控制限和匹配系数。根据匹配系数计算各模态统一的控制限。对新来的数据,运用概率密度确定其模态。新来数据向对应模态的模型上投影并计算统一的统计量,比较统计量与控制限进行多模态过程故障检测。把该方法应用到数值例子和半导体过程中,仿真结果表明,该算法在分类及多模态过程故障检测方面具有很高的准确性。     

Fault Detection Based On Online Probability Density Function Estimation

This study presents a new fault detection scheme based on the probability density function (PDF) of system output. Unlike the classical fault detection and diagnosis methods, in the proposed method, distribution of the system output is estimated online. To achieve this goal, an algorithm is introduced to estimate PDF online using fuzzy logic. Furthermore, convergence of this algorithm is investigated. Then, a residual is constructed that can show the existence of a fault in the system. The main advantages of the proposed method are robustness against measurement noise, even though it does not need the exact model and measured data of inputs and states. Simulation results show that this scheme can detect abrupt faults very well.     

Fault diagnosis for high order systems based on model decomposition

Fault detection observer and fault estimation filter are the main tools for the model based fault diagnosis approach. The dimension of the observer gain normally depends on the system order and the system output dimension. The fault estimation filter traditionally has the same order as the monitored system. For high order systems, these methods have the potential problems such as parameter optimization and the real time implementation on-board for applications. In this paper, the system dynamical model is first decomposed into two subsystems. The first subsystem has a low order which is the same as the fault dimension. The other subsystem is not affected by the fault directly. With the new model structure, a fault detection approach is proposed where only the residual of the first subsystem is designed to be sensitive to the faults. The residual of the second subsystem is totally decoupled from the faults. Moreover, a lower order fault estimation filter (with the same dimension of the fault) design algorithm is investigated. In addition, the design of a static fault estimation matrix is presented for further improving the fault estimation precision. The effectiveness of the proposed method is demonstrated by a simulation example.     

基于故障可诊断性量化评价的传感器优化配置方法研究

提出了一种基于故障可诊断性量化评价的传感器优化配置方法.针对可能发生故障的非线性系统,首先,基于K-L散度思想,通过计算故障情形下残差概率密度函数的差异度,得到了系统不同故障下故障可检测性和可分离性的量化指标,由于稀疏内核密度估计和蒙特卡洛算法的引入,克服了K-L散度计算中残差概率密度函数难以估计和非线性结构的K-L散度计算复杂度高的困难;其次,以故障可诊断性的定量评价为基础,借助于动态规划方法给出了系统满足期望故障可诊断性的传感器最优集合;最后,通过数值仿真和实体实验仿真验证了文中方法在故障诊断系统传感器优化配置中的有效性.     

Robust fault diagnosis and fault‐tolerant control for non‐Gaussian uncertain stochastic distribution control systems

The purpose of fault diagnosis of stochastic distribution control systems is to use the measured input and the system output probability density function to obtain the fault estimation information. A fault diagnosis and sliding mode fault‐tolerant control algorithms are proposed for non‐Gaussian uncertain stochastic distribution control systems with probability density function approximation error. The unknown input caused by model uncertainty can be considered as an exogenous disturbance, and the augmented observation error dynamic system is constructed using the thought of unknown input observer. Stability analysis is performed for the observation error dynamic system, and the H_∞ performance is guaranteed. Based on the information of fault estimation and the desired output probability density function, the sliding mode fault‐tolerant controller is designed to make the post‐fault output probability density function still track the desired distribution. This method avoids the difficulties of design of fault diagnosis observer caused by the uncertain input, and fault diagnosis and fault‐tolerant control are integrated. Two different illustrated examples are given to demonstrate the effectiveness of the proposed algorithm. Copyright © 2016 John Wiley & Sons, Ltd.     

一种图像隐写分析方法

针对现有隐写分析算法对彩色图像的检测效果不理想的问题,提出一种基于小波统计特征的隐写分析算法来分析嵌入信息前后载体图像与隐写图像之间的统计特征的差异。该算法首先将图像变换到YCbCr空间,其次通过小波分解提取预测子带的小波系数概率密度函数矩(PDF)和预测误差图像的子带系数直方图特征函数矩(CF),最后运用支持向量机(SVM)进行分类。实验表明,该方法对于流行的F5、OutGuess和Jphide&Seek隐写方法具有较强的检测能力。     

基于统计差分LPP的多模态间歇过程故障检测

针对工业过程数据存在的非高斯和多模态特性,提出一种基于统计差分LPP的多模态间歇过程故障检测方法。首先将统计模量分析的方法应用到间歇过程训练数据集中,计算统计过程变量的均值和方差,将不等长的批次变成等长的统计量,保证统计模量近似服从高斯分布;然后运用差分算法使多模态变为单模态,最后运用LPP算法进行降维和特征提取,计算样本的T2统计量,并利用核密度估计确定控制限。对于新来的测试样本数据统计差分处理后,向LPP模型上进行投影,计算新数据的T2统计量并与控制限比较进行故障检测。最后通过半导体过程数据的仿真结果表明,该算法的故障检测效果最好,验证了所提方法的有效性。     

基于规范变量残差的化工过程微小故障检测与诊断

微小故障因其幅值低而易被噪声和过程扰动所掩盖,并且会随时间慢慢演变成过程中的严重故障.因此,微小故障的检测和诊断变得越来越重要.为了更有效地监测和诊断微小故障,提出了基于规范变量残差的化工过程微小故障检测和诊断方法.首先,对Hankel矩阵执行奇异值分解来获得主元和残差空间并根据过去和未来数据的差异,求得两个不同的规范变量残差d_1, d_2.其次,考虑数据的时间序列特性,提出了基于规范变量残差的两个加权平均统计量W_(D1), W_(D2)及其控制限,进行故障检测;然后,计算出各个统计量的归一化贡献并绘制二维贡献图,进行故障诊断.最后,在连续搅拌釜式反应器(CSTR)过程中进行两种微小故障的应用研究.结果表明,与传统的统计量T~2,Q以及规范变量差异分析(CVDA)中统计量D相比,基于规范变量残差的加权平均统计量W_(D1), W_(D2)不仅能够及时检测到微小故障,而且在故障检测率和诊断率方面,均有不同程度的提高.     

基于第二类统计量的K分布参数估计

为了高效估计出K分布的参数,提出了对数累积量参数估计方法。基于第二类统计量,先对K分布的概率密度函数进行Mellin变换,从而获得K分布的第二类第一特征函数;然后对第二类第一特征函数进行对数变换,由此获得K分布的第二类第二特征函数;最后对第二类第二特征函数求导数,进而获得K分布的前两阶对数累积量,由此可以估计K分布的参数。与传统的最大似然估计方法相比,K分布的对数累积量估计具有解析的表达式,易于计算。Monte Carlo仿真表明,基于第二类统计量的K分布对数累积量估计可获得较高的估计精度。     

基于贝叶斯网络的电力变压器局部放电故障检测

针对传统电力变压器故障检测方法对电力系统中潜藏的故障问题检测水平不足,准确率较低,无法及时准确的发现异常隐患等问题,提出基于贝叶斯网络的变压器局部放电故障检测方法;首先通过传感器获取电力变压器不同状态下运行过程中的参数数据,对局部放电故障发生的概率和范围进行合理性评估,提取评估概率数据综合为样本数据集,构建贝叶斯网络故障树;根据逻辑规则转化为贝叶斯网络,推演计算故障节点之间的算例关系,利用贝叶斯原理抽取故障特征指标与异常概率之间的关联关系,利用模糊描述方法构建故障特征关联函数,计算可得故障特征模糊函数动态变化关系,实现对变压器故障发生的概率与位置信息的判断与确定;实验结果表明,利用贝叶斯网络对电力变压器局部放电故障检测准确率达到85%以上,最高可达96%,说明该方法具有较高的检测准确率,能够有效提高电力变压器放电故障检测的有效性。     

A robust fault diagnosis scheme based on signal modal estimation

A novel fault detection/diagnosis technique for linear dynamic systems is proposed. In comparison with existing schemes, the proposed method achieves fault detection/diagnosis using neither observer residuals nor parameter estimation errors; instead, it relies on the estimation of the underlying modal parameters of the system. The estimated modal parameters are compared with pre-calculated characteristic patterns of the system, which are represented as a set of root loci in terms of the physical system parameters. The modal parameter estimation is carried out using a numerically robust least-squares algorithm based on singular value decomposition. A pattern recognition technique employing linear multiprototype distance functions is used to classify the faults according to the variations of physical parameters. The proposed method has been applied to a simulated DC servo system where faults are introduced as abrupt changes in physical system parameters. It is shown that the proposed scheme is capable of diagnosing most of changes in physical system parameters.     

基于瞬时幅值的光伏系统电流传感器微小故障检测及估计

电流传感器是光伏系统中用于系统控制和状态监测的重要元件,然而受运行环境影响,电流传感器易出现性能退化,影响系统运行安全.为了准确检测和估计出电流传感器微小故障,提出基于瞬时幅值的传感器微小故障检测和估计方法.首先,建立基于瞬时幅值的电流传感器微小故障模型,利用Hilbert变换(HT)估计相电流瞬时幅值将测量的三相正弦电流转换为相互独立的三维直流信号分量;其次,利用快速移动窗主成分分析(FWMPCA)对三维直流信号组成的数据矩阵进行特征提取,获得主元和残差子空间向量的概率密度分布函数;然后,利用Kullback-Leibler(KL)距离定量度量实际运行数据相对于无故障运行数据的微小变化,在此基础上,设置故障检测阈值,构建故障幅值估计模型,实现电流传感器微小故障检测和估计;最后,利用RT-LAB实验平台验证所提方法的有效性.     

A comparative study of probability estimation methods for reliability analysis

In this paper we investigate the performance of probability estimation methods for reliability analysis. The probability estimation methods typically construct the probability density function (PDF) of a system response using estimated statistical moments, and then perform reliability analysis based on the approximate PDF. In recent years, a number of probability estimation methods have been proposed, such as the Pearson system, saddlepoint approximation, Maximum Entropy Principle (MEP), and Johnson system. However, no general guideline to suggest a most appropriate probability estimation method has yet been proposed. In this study, we carry out a comparative study of the four probability estimation methods so as to derive the general guidelines. Several comparison metrics are proposed to quantify the accuracy in the PDF approximation, cumulative density function (CDF) approximation and tail probability estimations (or reliability analysis). This comparative study gives an insightful guidance for selecting the most appropriate probability estimation method for reliability analysis. The four probability estimation methods are extensively tested with one mathematical and two engineering examples, each of which considers eight different combinations of the system response characteristics in terms of response boundness, skewness, and kurtosis.     

Cascade filter structure for sensor/actuator fault detection and isolation of satellite attitude control system

This paper presents a new scheme for fault detection and isolation in a satellite system. The purpose of this paper is to develop detection, isolation and identification algorithms based on a cascade filter for both total and partial faults in a satellite attitude control system (ACS). The cascade filter consists of a decentralized Kalman filter (DKF) and a bank of interacting multiple model (IMM) filters. The cascade filter is utilized for detection and diagnosis of anticipated sensor and actuator faults in a satellite ACS. Other fault detection and isolation (FDI) schemes are compared with the proposed FDI scheme. The FDI procedure using a cascade filter was developed in three stages. In the first stage, two local filters and a master filter detect sensor faults. In the second stage, the FDI scheme checks sensor residuals to isolate sensor faults, and 11 Extended Kalman filters with actuator fault models detect wherever actuator faults occur. In the third stage of the FDI scheme, four filters identify the fault type, which is either a total or partial fault. An important feature of the proposed FDI scheme is that it can decrease fault isolation time and accomplish not only fault detection and isolation but also fault type identification using a scalar penalty in the conditional density function.     

基于局部保持投影–加权k近邻规则的 多模态间歇过程故障检测策略

针对多模态间歇过程故障检测问题,本文提出一种基于局部保持投影–加权k近邻规则(LPP--Wk NN)的故障检测策略.首先,应用局部保持投影(LPP)方法将原始数据投影到低维主元子空间;接下来,在主元子空间中,应用样本第k近邻的局部近邻集确定每个样本的权重并计算权重统计量Dw;最后,应用核密度估计方法确定Dw控制限并进行故障检测.本文方法应用LPP对过程数据进行维数约减,既能够降低训练过程中离群点对模型的影响,又能够降低在线故障检测的计算复杂度.同时,加权k近邻规则(Wk NN)方法通过引入权重规则能够使得过程故障检测统计量分布具有单模态结构.相比传统的k NN统计量,本文引入的权重统计量具有更高的故障检测性能.通过数值例子和半导体蚀刻过程的仿真实验,并与主元分析(PCA), k NN, Wk NN, LPP--k NN等方法进行比较,实验结果验证了本文方法的有效性.     

Fault detection,isolation and identification for hybrid systems with unknown mode changes and fault patterns

This article presents a solution to the problem of multiple fault detection, isolation and identification for hybrid systems without information on mode change and fault patterns. Multiple faults of different patterns are considered in a complex hybrid system and these faults can happen either in a detectable mode or in a non-detectable mode. A method for multiple fault isolation is introduced for situation of lacking information on fault pattern and mode change. The nature of faults in a monitored system can be classified as abrupt faults and incipient faults. Under abrupt fault assumption, i.e. constant values for fault parameters, fault identification is inappropriate to handle cases related to incipient fault. Without information on fault nature, it is difficult to achieve fault estimation. Situation is further complicated when mode change is unknown after fault occurrence. In this work, fault pattern is represented by a binary vector to reduce computational complexity of fault identification. Mode change is parameterized as a discontinuous function. Based on these new representations, a multiple hybrid differential evolution algorithm is developed to identify fault pattern vector, abrupt fault parameter/incipient fault dynamic coefficient, and mode change indexes. Simulation and experiment results are reported to validate the proposed method.