PLS-based EWMA fault detection strategy for process monitoring

Fault detection (FD) and diagnosis in industrial processes is essential to ensure process safety and maintain product quality. Partial least squares (PLS) has been used successfully in process monitoring because it can effectively deal with highly correlated process variables. However, the conventional PLS-based detection metrics, such as the Hotelling's T² and the Q statistics are ill suited to detect small faults because they only use information from the most recent observations. Other univariate statistical monitoring methods, such as the exponentially weighted moving average (EWMA) control scheme, has shown better abilities to detect small faults. However, EWMA can only be used to monitor single variables. Therefore, the main objective of this paper is to combine the advantages of the univariate EWMA and PLS methods to enhance their performances and widen their applicability in practice. The performance of the proposed PLS-based EWMA FD method was compared with that of the conventional PLS FD method through two simulated examples, one using synthetic data and the other using simulated distillation column data. The simulation results clearly show the effectiveness of the proposed method over the conventional PLS, especially in the presence of faults with small magnitudes.     

基于KPLS数据重构的非线性过程监测与故障辨识

为适应快速变化的化工产品需求,过程工业逐步向柔性生产发展,使得间歇过程的应用日益广泛。这一类工艺过程具有动态和非线性的特征,过程故障带来的工艺波动和安全风险是较为突出的挑战。采用基于核函数的偏最小二乘方法,在高维特征空间提取特征变量,这些变量包含了生产过程的非线性结构特征,也反应了过程工况的模式特征。针对传统线性方法存在的故障漏报等问题,利用核函数技巧,在特征空间进行数据重构,进而计算统计监控指标SPE,并通过对SPE的在线监测实现更加有效地故障辨识。本方法针对标准非线性测试对象进行了过程监测,实现结果充分说明了方法的有效性。     

Minimum-loss design of x-bar charts for correlated data

When the x-bar chart is used to monitor a production process, three parameters should be determined: the sample size, the sampling interval between successive samples, and the control limits for the chart. In 1956, Duncan presented a cost model to determine the three parameters, which is called the economic design of x-bar charts. In 1995, Alexander et al. combined Duncan’s cost model with the Taguchi’s quality loss function to present a loss model for determining the three parameters. When designing an x-bar chart, one usually assumes that the measurements within a sample are independent; however, this assumption may not be true for some processes. In this paper, we develop the minimum-loss design of x-bar charts for correlated measurements within a sample by incorporating the Taguchi’s quality loss function. An example of orange juice production process is presented to illustrate the solution procedure. From the results of sensitivity analysis, we find that as the measurements in the sample are positively correlated, highly correlated data result in a smaller sample size and a frequent sampling interval; however, as the measurements in the sample are negatively correlated, highly correlated data yield a smaller sample size and a narrower control limits.     

Economic-statistical design of multivariate control charts for monitoring the mean vector and covariance matrix

When a control chart is used to monitor a process, three test parameters should be determined: the sample size, the sampling interval between successive samples, and the control limits or critical region of the chart. In this paper, we present the procedure to conduct the economic-statistical design of multivariate control charts for monitoring the process mean vector and covariance matrix simultaneously; i.e. to economically determine the optimum values of the three test parameters such that the statistical constraints (including the type I error probability and power) of the control chart can be satisfied. The test statistic −2ℓnL is applied to develop this procedure and the cost model is established based on the function given by Montgomery and Klatt. A numerical example is provided to illustrate the solution procedure of the design and then the effects of cost parameters on the optimal design are examined.     

新一代极轨气象卫星FY3A-VIRR数据的地表火监测算法研究与评价

首先对新一代极轨气象卫星FY3-VIRR（可见光红外扫描辐射计）传感器热异常监测特性进行描述,比较了VIRR、AVHRR、MODIS三个传感器的4μm通道光谱响应特征,给出VIRR数据定标计算和几何校正的算法,在分析VIRR传感器热异常监测特性的基础上,给出考虑像元背景亮温的火点识别算法,经过试验得到云识别和火点检测参...     

基于免疫遗传算法的化工过程微小差别异常诊断

化工过程中同一类型的异常往往也存在微小差别,将不同程度的同一类型异常进行有效识别和归属,掌握同一类型异常的多样化情况,对于化工过程安全监控有重要意义。设计了一种基于人工免疫和遗传思想的微小差别异常诊断方法,并以一个典型的精馏过程Chem CAD仿真模型进行同类型不同程度的阀门异常模式设置,获取正常情况和各异常情况下的样本数据,然后利用所提出的诊断方法计算不同程度异常对其异常类型的隶属概率。结果表明,同一阀门不同开度异常尽管会导致过程监控参数的变化,但该微小差别异常诊断方法却能对其进行有效归属,且归属准确度经启发式算法多次运行后可达98%以上。最后进一步运用TE过程中具有微小差异输出结果的4种异常模式进行了验证试验,结果表明,该方法也可以在较少的异常数据基础上完成对异常模式的准确归属。     

一种改进的CSCA+PCA化工过程异常检测技术

为了解决化工过程异常检测时因参数众多且数据庞杂而导致一些异常无法被有效检出的问题,在Brownlee的克隆选择分类算法（CSCA）基础上,通过引入主成分分析（PCA）技术,进行数据降维和数据重整,探讨了人工免疫算法在化工过程异常检测中的适用效果和技术方案,以TE过程数据作为样本进行异常检测和分类实验。结果表明,过程异常数据的规模、属性的数目对CSCA异常检测效果具有明显影响,而通过主成分分析进行数据降维之后,CSCA检测效果有所提高;进一步的数据重整之后,CSCA对过程异常分类辨识的准确率可提升到85%以上;基于CSCA+PCA的数据降维及重构之后的过程异常检测技术方案,可以获得较高的异常检测准确率,从而一定程度上为化工过程安全运行提供技术保障。     

Statistical fault detection using PCA-based GLR hypothesis testing

Safe process operation requires effective fault detection (FD) methods that can identify faults in various process parameters. In the absence of a process model, principal component analysis (PCA) has been successfully used as a data-based FD technique for highly correlated process variables. Some of the PCA detection indices include the T² or Q statistics, which have their advantages and disadvantages. When a process model is available, however, the generalized likelihood ratio (GLR) test, which is a statistical hypothesis testing method, has shown good fault detection abilities. In this work, a PCA-based GLR fault detection algorithm is developed to exploit the advantages of the GLR test in the absence of a process model. In fact, PCA is used to provide a modeling framework for the develop fault detection algorithm. The PCA-based GLR fault detection algorithm provides optimal properties by maximizing the detection probability of faults for a given false alarm rate. The performance of the PCA-based GLR fault detection algorithm is illustrated and compared to conventional fault detection methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of these examples clearly show the effectiveness of the developed algorithm over conventional methods.     

基于HMM和DT的无人机异常检测方法

为了实时检测无人机异常状态,提出基于隐马尔可夫模型(Hidden Markov Model,HMM)和决策树(Decision Tree,DT)的无人机异常检测方法(HMMDT).首先根据异常致因将无人机异常分为干扰异常和硬件异常;然后结合HMM和DT建立无人机异常检测模型,定义无人机异常度衡量异常状态的严重程度,确定...     

基于电磁波传播的采空区AMT探测与异常快速圈定*——以云南东川烂泥坪铜矿区为例

为了研究电磁波的传播在采空区界面上发生反射和透射后对音频大地电磁测深（简称AMT）勘探的影响,模拟采空区推导出基于4层介质模型的阻抗计算公式。经过计算分析,采空区会引起AMT的电阻率和阻抗参数在高频段出现震荡,可利用高值异常波动来定性判断采空区。根据东川烂泥坪铜矿已知采空区的AMT探测资料,采空区引起视电阻率和阻抗在15~30 kHz频段出现高值异常,相对于视电阻率资料,阻抗对异常的反映更明显。提出采用阻抗平均法和最大阻抗法计算相对偏差,通过高值异常对采空区进行快速圈定。研究结果表明:2种方法得到的结果较为相似,基本符合实际采空区的分布情况,而最大阻抗法更适合采空区AMT探测使用。     

A methodology for evaluation and monitoring of recurring hazards in underground coal mining

This study provides a methodology for evaluation and monitoring of recurring hazards in underground coal mining. An important measure in this regard may be the ‘time between occurrences’ (TBO) of hazards that can be modeled in the similar fashion of ‘time between failures’ (TBF) data modeling which is practiced in reliability study. Typically, time between accidents is modeled in safety study. This study is therefore new in two counts: (i) statistically modeling hazard occurrences based on inspection reports and (ii) monitoring of safety status based on control charting of hazard occurrences. The methodology includes Weibull-distribution based hazard rate functions, Poisson-distribution based cumulative risk functions, and Weibull-distribution based control charts. The new methodology is applied to an underground coal mining worksystem and the results are discussed. The case study results show that hazards related to machinery, ground-fall, housekeeping, roadways, and materials are more frequently occurring. It is recommended that in addition to planned inspections for identification of hazards, a control chart based hazard mitigation scheme should be employed at the mine sections for better monitoring and control of hazards.     

A prediction model for the flash point of binary liquid mixtures

A prediction model based on the partial least squares of the multivariate statistical analysis methods was developed for the flash point (FP) of binary liquid mixtures. Estimation of the FP of flammable substances is important for safety measures in industrial processes. Since experimental FP data of liquid mixtures are scarce in the literature, there have been many researches to estimate the FP of liquid mixtures using physicochemical laws. In this study, the partial least squares (PLS) method using experimental data was used as a prediction model of the FP of binary liquid mixtures. The FPs predicted from the PLS method were also compared to results from the existing calculating methods using physicochemical laws such as Raoult's law and the Van Laar equation.     

基于计算机视觉的驾驶疲劳识别方法的研究

汽车驾驶员疲劳驾驶是引发交通事故的重要原因之一,许多国家在进行疲劳报警器的研究。在介绍和比较多种驾驶疲劳测评方法的基础上,阐述PERCLOS评价驾驶疲劳的机理,并与其他几种方法进行对比,以证明其优越性。设计一套驾驶员疲劳监控系统,应用图像差分的方法,灰度直方图、直方图均衡化等图像分析手段,定位和识别驾驶员眼睛睁开与闭合的变化过程,统计出眼睛闭合时间来确定疲劳程度,并通过利用MATLAB得以实现。给出了一种基于PERCLOS的驾驶员疲劳识别的新方法。     

煤矿生产系统干扰事件的离散小波变换检测

小波变换常用于数字信号传输的噪声检测与处理,为了检测煤矿生产系统中的突发事件,将其模拟为数字信号传输中的干扰,构建系统稳态方程和对应的扰动方程,根据离散小波变换原理,将煤矿生产系统中的监测数据分解为近似部分和细节部分,再进行离散变换。采用多尺度边缘检测法,由光滑后的一阶和二阶导数检测出信号的模极大值点即为瞬态突变点。将A矿记录到的瓦斯浓度监控数据进行可行性分析后,在MATLAB的wavelet工具中对该瓦斯浓度监控数据作离散小波变换,找到四个瓦斯浓度瞬态突变点,那么这四个时刻发生了煤与瓦斯突出事件。结果表明该方法简单、实用性强。据此从计算机实时监控和现场技术防突方面提出应对煤矿干扰事件的措施。     

不同去除浊度方法对水中总磷测定影响的比较

本文研究了去除浊度方法的不同是否会导致各监测单位测定结果可比性差异。首先,通过在线自动监测仪器法和实验室钼酸铵分光光度法测定水中总磷浓度的比对实验所得出的监测数据结果发现,两者的结果误差较小,说明自动监测仪器分析结果与常规实验室手工法测定结果基本一致,未发现可比性差异,同样稳定可靠。同时,通过实验系统研究了手工法中应用浊度-色度补偿法、过滤法及离心法等去除浊度干扰的前处理方法,实验结果表明:三种分析方法测定结果基本一致,精密度和加标回收率均符合质量控制要求,测定结果无显著差异。     

A new fault detection method for non-Gaussian process based on robust independent component analysis

Conventional fault detection method based on fast independent component analysis (FastICA) is sensitive to outliers in the modeling data and thus may perform poorly under the adverse effects of outliers. To solve such problem, a new fault detection method for non-Gaussian process based on robust independent component analysis (RobustICA) is proposed in this paper. A RobustICA algorithm which can effectively reduce the effects of outliers is firstly developed to estimate the mixing matrix and extract non-Gaussian feature called independent components (ICs) by robust whitening and robust determination of the maximum non-Gaussian directions. Furthermore, a monitoring statistic for each extracted IC is constructed to detect process faults. Simulations on a simple example of the mixing matrix estimation and a fault detection example in the continuous stirred tank reactor system demonstrate that the RobustICA achieves much higher estimation accuracy for the mixing matrix and the ICs than the commonly used FastICA algorithm, and the RobustICA-based fault detection method outperforms the conventional FastICA-based fault detection method in terms of the fault detection time and fault detection rate.     

基于LabVIEW的水质微型站实时故障诊断专家系统

设计一种基于微试剂原位水质检测仪的多参数水质微型站监控系统。通过对水质检测流程合理分析以及各监控点反馈信号诊断,结合LabVIEW可视化编程技术、信号分析技术和测控技术,分别从硬件和软件方面,设计了一种基于多传感器数据融合的水质微型站监控感知的检测流程测控模型。现场应用结果表明,采用诊断专家系统的水质微型站监控系统满足水质测控的需求,极大提升系统的稳定性、安全性。     

Risk assessment of an oil depot using the improved multi-sensor fusion approach based on the cloud model and the belief Jensen-Shannon divergence

At present, enterprises have introduced the Internet of Things (IoT) technology to monitor and evaluate the safety status of oil depots, allowing for the collection of a substantial amount of multi-source monitoring data from factories. However, sensor monitoring data is often inaccurate and fuzzy. To improve the reliability of risk prevention and control based on multi-source sensor data, this study proposed a CM-BJS-DS model based on the cloud model (CM), the Belief Jensen-Shannon (BJS) divergence and Dempster-Shafer(D-S) evidence theory. First, the relevant evaluation factors of the accident and their threshold intervals of different risk levels were determined, and the fuzzy cloud membership functions (FCMFs) corresponding to different risk levels were constructed. Then, the sensor monitoring data were processed using the correlation measurement of the FCMF, and basic probability assignments (BPAs) were generated under the risk assessment frame of discernment. Finally, the BPAs were pre-processed by the improved evidence fusion model and the accident risk level was evaluated. Based on the monitoring data, a case study was performed to assess the risk level of vapor cloud explosion (VCE) accidents due to liquid petroleum gas (LPG) tank leaks. The results show that the proposed method presents the following characteristics: (i) The BPAs were constructed based on the monitoring data, which reduced the subjectivity of the construction process; (ii) Compared with single sensors, the multiple sensor fusion evaluation yielded more specific results; (iii) When dealing with highly conflicting evidence, the evaluation results of the proposed method exhibited a higher belief degree. This method can be used as a decision-making tool to detect potential risks and identify critical risk spots to improve the specificity and efficiency of emergency response.     

有机过氧化物反应热的危险性预测研究

为了采用非实验的方法对安全物质学的研究内容及研究方法进行初探,基于定量结构-性质关系法,选择13种与有机过氧化物热危险性的影响因子密切相关的描述符,分别对起始分解温度T0和分解热△H的实验数据进行多元线性回归、偏最小二乘和支持向量机回归分析,从而获得3种相应的预测模型。对比T0与△H的实验值和预测值,结果发现:SVM预测模型的精度高于PLS预测模型,MLR预测模型的精度最低;同种预测模型对分解热的预测结果均优于起始分解温度。此外,分析各预测模型的稳定性数据发现:MLR模型的预测过程发生了过拟合现象,不具备预测能力;PLS模型的交互验证系数均大于0.5,具备较稳定的预测能力;SVM模型的交互验证系数均大于0.9,具备非常稳定的预测能力。