Fault Isolation by Partial Dynamic Principal Component Analysis in Dynamic Process

Principal component analysis （PCA） is a useful tool in process fault detection, but offers little support on fault isolation. In this article, structured residual with strong isolation property is introduced. Although it is easy to get the residual by transformation matrix in static process, unfortunately, it becomes hard in dynamic process under control loop. Therefore, partial dynamic PCA（PDPCA） is proposed to obtain structured residual and enhance the isolation ability of dynamic process monitoring, and a compound statistic is introduced to resolve the problem resulting from independent variables in every variable subset. Simulations on continuous stirred tank reactor （CSTR） show the effectiveness of the proposed method.

Loss-of-Main Monitoring and Detection for Distributed Generations Using Dynamic Principal Component Analysis

In this paper, our previous work on Principal Component Analysis (PCA) based fault detection method is extended to the dynamic monitoring and detection of loss-of-main in power systems using wide-area synchrophasor measurements. In the previous work, a static PCA model was built and verified to be capable of detecting and extracting system faulty events;however the false alarm rate is high. To address this problem, this paper uses a well-known ‘time lag shift’ method to include dynamic behavior of the PCA model based on the synchronized measurements from Phasor Measurement Units (PMU), which is named as the Dynamic Principal Component Analysis (DPCA). Compared with the static PCA approach as well as the traditional passive mechanisms of loss-of-main detection, the proposed DPCA procedure describes how the synchrophasors are linearly auto- and cross-correlated, based on conducting the singular value decomposition on the augmented time lagged synchrophasor matrix. Similar to the static PCA method, two statistics, namely T2 and Q with confidence limits are calculated to form intuitive charts for engineers or operators to monitor the loss-of-main situation in real time. The effectiveness of the proposed methodology is evaluated on the loss-of-main monitoring of a real system, where the historic data are recorded from PMUs installed in several locations in the UK/Ireland power system.

DYNAMIC VARIATIONS OF WATER QUALITY IN TAIHU LAKE AND MULTIVARIATE ANALYSIS OF ITS INFLUENTIAL FACTORS

Dynamic variation of water quality in Meiliang Bay and part of West Taihu Lake has been analysed based on data from 1991 to 1992. Principal Component Analysis is used to reveal the mutual relationships of various factors. It is shown that there existis an obvious spatial and temporal variation in the main factors of water quality. Annual values of TP, CON, TN, Chl-a and conductivity decrease evidently from inner Meiliang Bay to the outer from north to south. TP and TN fluctuate seasonally with much higher value in winter. This is particularly true for the mouth of Liangxi River. In addition, the Chl-1 has a synchronous variation with water temperature, although being lagged a little, and closely relates to TP and TN. Finally, the results from Principal Component Analysis show that TP, TN, SS (or SD), water temperature and Chl-a are the most influential factors to water qualuty in this area, and both suspensions and algae can contribute to transparency to Taihu Lake.

DYNAMIC VARIATIONS OF WATER QUALITY IN TAIHU LAKE AND MULTIVARIATE ANALYSIS OF ITS INFLUENTIAL FACTORS

ＤＹＮＡＭＩＣ ＶＡＲＩＡＴＩＯＮＳ ＯＦ ＷＡＴＥＲ ＱＵＡＬＩＴＹ ＩＮ ＴＡＩＨＵ ＬＡＫＥ ＡＮＤ ＭＵＬＴＩＶＡＲＩＡＴＥ ＡＮＡＬＹＳＩＳ ＯＦ ＩＴＳ ＩＮＦＬＵＥＮＴＩＡＬ ＦＡＣＴＯＲＳＤＹＮＡＭＩＣＶＡＲＩＡＴＩＯＮＳＯＦＷＡＴＥ...

Nonlinear Dynamic Analysis of MPEG-4 Video Traffic

The main research motive is to analysis and to veiny the inherent nonlinear character of MPEG-4 video. The power spectral density estimation of the video trafiic describes its 1/f^β and periodic characteristics.The priraeipal compohems analysis of the reconstructed space dimension shows only several principal components can be the representation of all dimensions. The correlation dimension analysis proves its fractal characteristic. To accurately compute the largest Lyapunov exponent, the video traffic is divided into many parts.So the largest Lyapunov exponent spectrum is separately calculated using the small data sets method. The largest Lyapunov exponent spectrum shows there exists abundant nonlinear chaos in MPEG-4 video traffic. The conclusion can be made that MPEG-4 video traffic have complex nonlinear be havior and can be characterized by its power spectral density,principal components, correlation dimension and the largest Lyapunov exponent besides its common statistics.

Insight into the Urea Binding and K166R Mutation Stabilizing Mechanism of TIpB： Molecular Dynamics and Principal Component Analysis Study

Chemoreceptor TlpB（Tlp=transducer-like protein）, which has been demonstrated to respond to pH sensing function, is crucial for the survival ofHelicobacterpylori（H, pylori） in host stomach. Urea was proposed to be essen- tial for TlpB＇s pH sensing function via binding with the Per-ARNT-Sim（PAS） domain of TlpB. Additionally, KI66R mutation of the TlpB protein has also been proven to have a similar effect on TlpB pH sensing as urea binding. Al- though X-ray crystallographic studies have been carried out for urea-bound Tlpl3, the molecular mechanism for the stabilization of TIpB induced by urea binding and K166R mutation remains to be elucidated. In this study, molecular dynamics simulations combined with principal component analysis（PCA） for the simulation results were used to gain an insight into the molecular mechanism of the stabilization of urea on TlpB protein. The formed H-bonds and salt-bridges surrounding Aspll4, which were induced by both urea binding and K166R mutation of TIpB, were im- portant to the stabilization of TlpB by urea. The similarity between the urea binding and K166R mutation as well as their differences in effect has been explicitly demonstrated with computer simulations at atomic-level. The findings may Dave the wav for the further researches of TlpB.

Adaptive multiblock kernel principal component analysis for monitoring complex industrial processes

Multiblock kernel principal component analysis (MBKPCA) has been proposed to isolate the faults and avoid the high computation cost. However, MBKPCA is not available for dynamic processes. To solve this problem, recursive MBKPCA is proposed for monitoring large scale processes. In this paper, we present a new recursive MBKPCA (RMBKPCA) algorithm, where the adaptive technique is adopted for dynamic characteristics. The proposed algorithm reduces the high computation cost, and is suitable for online model updating in the feature space. The proposed algorithm was applied to an industrial process for adaptive monitoring and found to efficiently capture the time-varying and nonlinear relationship in the process variables.

Dynamic model for predicting nitrogen oxide concentration at outlet of selective catalytic reduction denitrification system based on kernel extreme learning machine

To solve the increasing model complexity due to several input variables and large correlations under variable load conditions,a dynamic modeling method combining a kernel extreme learning machine(KELM)and principal component analysis(PCA)was proposed and applied to the prediction of nitrogen oxide(NO_(x))concentration at the outlet of a selective catalytic reduction(SCR)denitrification system.First,PCA is applied to the feature information extraction of input data,and the current and previous sequence values of the extracted information are used as the inputs of the KELM model to reflect the dynamic characteristics of the NO_(x)concentration at the SCR outlet.Then,the model takes the historical data of the NO_(x)concentration at the SCR outlet as the model input to improve its accuracy.Finally,an optimization algorithm is used to determine the optimal parameters of the model.Compared with the Gaussian process regression,long short-term memory,and convolutional neural network models,the prediction errors are reduced by approximately 78.4%,67.6%,and 59.3%,respectively.The results indicate that the proposed dynamic model structure is reliable and can accurately predict NO_(x)concentrations at the outlet of the SCR system.

DYNAMICS OF REGIONAL URBANIZATION IN FUJIAN COASTAL AREA

Since the economic reform in 1978, urban development in China has become much more rapid and the dynamic mechanisms of urbanization more diversified. The 'Bottom Up' strategy becomes as important as, or even more important than, the 'Top Down' strategy as the dynamic mechanisms of regional urbanization. On the basis of major theories of development economics and regional economics, this paper analyzes the major dynamic mechanisms of regional urbanization in coastal area of Fujian Province from 1978 to 1989, and describes quantitatively the territorial differentiation of regional urbanization process under two major dynamic mechanisms using principal componet analysis.

Factor-Cluster Analysis and Effect of Particle Size on Total Recoverable Metal Concentration in Sediments of the Lower Tennessee River Basin

Total recoverable concentration of five elements of concern: Aluminum, Iron, Manganese, Arsenic and Lead (Al, Fe, Mn, As, Pb) were measured by inductively coupled plasma atomic emission spectrometry, and mass spectrometry. The results show that sediment texture plays a controlling role in the concentrations and their spatial distribution. Principal Component Analysis and Cluster Analysis were used to analyze the grain sizes of the sediments. Result of texture analysis classified the samples into three main components in percentages: sand, silt, and clay. Significant differences among the element concentrations in the three groups were observed, and the concentrations of the elements in each group are reported in this study. Most of the elements have their highest concentrations in the fine-grained samples with clay playing an important role, in comparison with the sand component of the soil/sediment samples. There appears to be a strong correlation between samples with high silt, and clay content with the areas of elevated concentrations for Al, Fe, and Mn. There was a strong correlation between aluminum and lead with clay;lead with silt;and sand with manganese, aluminum, and lead. However, there was no strong relationship between the soil textures and iron or arsenic. All elements measured were statistically significant (at P ≤ 0.05) by watershed. The upland areas, and depositional areas’ spatial variation of element concentrations in the sediments were also observed, which was in line with the spatial distribution of the grain size and was thought to be related to the watersheds hydrological dynamics.

基于DPCA-BP神经网络的中长期电力负荷预测方法

针对PCA-神经网络预测方法解决预测问题时,忽视数据自相关性而造成的预测结果难以满足实际工程要求精度的研究现状,建立了预测数据的增广矩阵.通过计算前l时刻数据确定增广矩阵的维数,并把得到增广后的预测数据作为BP神经网络的输入变量,建立了基于DPCA-BP神经网络的预测模型,给出了模型结构.该模型能有效地去除自变量系统中与因变量无关的数据信息,增加自变量系统中数据的自相关性.算例比较分析表明,所建立模型的模型成分解释性增强,预测精度提高,预测效果优于PCA-BP神经网络方法.

基于DPCA与LSSVM的飞机发动机异常状态识别

针对飞机发动机异常状态识别精度差、效率低和易误诊漏诊等问题,提出了一种基于动态主元分析(dynamic principal component analysis,DPCA)和最小二乘支持向量机(least square support vector machine,LSSVM)的飞机发动机润滑系统异常状态识别方法;首先对发动机润滑系统参数进行DPCA处理以及在线检测是否有故障发生,如果有故障发生,再采用LSSVM方法进行异常状态识别;以某型飞机发动机润滑系统为例,对文中所提方法的准确性进行试验验证,由试验结果得出文中方法能有效提高飞机发动机异常状态识别准确率。

人脸图像识别中非贪婪L1范数的2DPCA最大化鲁棒算法

基于L1范数的二维主成分分析是近年来提出的一种在图像域降维和特征提取的方法。通常,直接求解L1范数最大化问题很困难,因此,一种贪婪的策略被提出来了。然而,这种策略的初始化投影是随意选取的,为了获得更好的投影向量,得到一个最优的局部解,提出了一个非贪婪的L1范数最大化算法,该非贪婪算法具有三大优势:使用L1范数和非贪婪策略对于异常值很稳健;与PCA-L1相比较,更多的空间结构信息得到了保留;相比2DPCA-L1,所有的投影方向可以被优化并且可以获得更好的解决方案。人脸和其他数据集上的实验结果验证了所提出的方法更加有效。

基于DPCA-KD的污水处理过程故障诊断

污水生化处理过程是一类强非线性、变量耦合、工况复杂的过程。由于环境恶劣,污水生化处理过程传感器故障频发,导致传统基于动态主元分析的故障检测方法漏报率较高、检测率较低。提出了一种基于Kantorovich距离的动态主元分析故障检测方法。首先,通过动态主元分析构建增广矩阵,对多维数据进行降维,降低数据的自相关性。其次,通过Kantorovich距离对过程的数据进行故障检测。最后,基于国际水协会的基准仿真模型BSM1对所提方法进行验证。结果表明,所提出的方法相较于传统的动态主元分析方法,降低了故障误报率、提高了检测率。

BDPCA在线过程监测方法

针对基于多向主元分析(Multiway Principal Component Analysis,MPCA)的方法在批过程故障监测中以样本观测相互独立作为假设前提条件,没有考虑到时间序列相关性的影响及需要对新批次未反应完的数据进行预估的缺陷,提出一种批过程动态主元分析(Batch Dynamic PCA,BDPCA)在线监测方法。该方法采用时滞变量将过程的静态和动态特征相结合,有效地去除了测量变量时间序列的自相关关系,并通过时滞窗口提供了在线监测方案,避免了对新批次未反应完的数据进行预估的需要,提出确定时滞变量的算法。将BDPCA应用于β-甘露聚糖酶发酵批过程的仿真监测,与移动窗多向主元分析(Moving Window MPCA,MWMPCA)法相比,仿真结果表明该方法能够更精确地对过程故障行为进行描述,具有良好的准确性和实时性。

基于DPCA方法的传感器故障检测与诊断

针对多传感器的相关时序测量数据,在假设只存在传感器故障的前提下,提出了一种基于动态主成分分析(DPCA)的传感器故障检测方法。根据测量数据建立传感器的DPCA模型,在该模型基础上利用T2和SPE统计量进行传感器的故障检测。同时,将基于主成分分析(PCA)模型的传感器有效度指标SVI推广应用于DPCA模型中。通过对污水处理系统中重要传感器的故障诊断仿真实验表明:该方法能有效地检测和识别出故障传感器。

基于DPCA方法的故障检测与诊断分析

主元分析(PrincipalComponentAnalysis,PCA)已广泛应用于复杂工业过程的运行状态监控。然而,传统的PCA方法仅构造了生产过程的静态线性关系,无法从根本上有效处理具有较强动态特性的实际工业生产过程。动态主元分析(DynamicPCA,DPCA)是一种将传统PCA分析推广到动态多变量过程的方法,但其较大的计算负荷阻碍了其实际应用。本文对文献中的DPCA作了算法上的简化,减少了实施中的计算量,并将其应用于重油分馏塔的动态运行故障监测与诊断。研究结果表明了方法的有效性。

基于分块的2DPCA人脸识别方法

文章将分块理论与2DPCA方法相结合，研究分块二维主成分分析法（M-2DPCA）在人脸识别中的应用。对人脸图像矩阵进行分块，用形成的子图像矩阵直接构造总体散布矩阵并求解对应的特征向量，利用提取的特征向量对图像进行特征的提取与分析，进行人脸识别。基于Yale人脸数据库的实验显示，在相同训练样本和特征向量条件下，M-2DPCA比2DPCA算法具有更高的识别率。结论 M-2DPCA充分利用了图像的协方差信息，在人脸识别方面具有较高的识别率和鲁棒性方面，对进一步研究人脸识别具有重要的意义。

DPCA与GA-SVM融合的智能台车液压系统故障诊断

针对智能台车液压系统故障原因复杂多样及故障诊断效率低等问题,提出动态主成分分析(DPCA)与遗传算法改进支持向量机(GA-SVM)相结合的液压系统故障诊断方法。首先,采用AMEsim软件建立液压系统仿真模型采集故障数据并进行预处理;然后采用DPCA对故障特征向量进行降维,解除特征间的相关性和缩短训练时间;再运用遗传算法对SVM进行参数优化,将抽取出来的故障特征参数样本输入优化后的SVM中进行训练,获得分类模型,从而实现故障诊断。测试结果表明该方法的效率高于常规PCA-SVM及BP神经网络,为台车设备的维修和保养提供了指导,具有良好的应用价值和前景。