Kullback-Leibler distance-based enhanced detection of incipient anomalies

Accurate and effective anomaly detection and diagnosis of modern engineering systems by monitoring processes ensure reliability and safety of a product while maintaining desired quality. In this paper, an innovative method based on Kullback-Leibler divergence for detecting incipient anomalies in highly correlated multivariate data is presented. We use a partial least square (PLS) method as a modeling framework and a symmetrized Kullback-Leibler distance (KLD) as an anomaly indicator, where it is used to quantify the dissimilarity between current PLS-based residual and reference probability distributions obtained using fault-free data. Furthermore, this paper reports the development of two monitoring charts based on the KLD. The first approach is a KLD-Shewhart chart, where the Shewhart monitoring chart with a three sigma rule is used to monitor the KLD of the response variables residuals from the PLS model. The second approach integrates the KLD statistic into the exponentially weighted moving average monitoring chart. The performance of the PLS-based KLD anomaly-detection methods is illustrated and compared to that of conventional PLS-based anomaly detection methods. Using synthetic data and simulated distillation column data, we demonstrate the greater sensitivity and effectiveness of the developed method over the conventional PLS-based methods, especially when data are highly correlated and small anomalies are of interest. Results indicate that the proposed chart is a very promising KLD-based method because KLD-based charts are, in practice, designed to detect small shifts in process parameters.     

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。     

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.     

A hybrid approach to faults detection and diagnosis in batch and semi-batch reactors by using EKF and neural network classifier

This work deals with a new hybrid approach for the detection and diagnosis of faults in different parts of fed-batch and batch reactors. In this paper, the fault detection method is based on the using of Extended Kalman Filter (EKF) and statistical test. The EKF is used to estimate on-line in added to the state of reactor the overall heat transfer coefficient (U). The diagnosis method is based on a probabilistic neural network classifier. The Inputs of the probabilistic classifier are the input–output measurements of reactor and the parameter U estimated by EKF, while the outputs of the classifier are fault types in reactor. This new approach is illustrated for simulated as well as experimental data sets using two cases of reactions: the first is the oxidation of sodium thiosulfate by hydrogen peroxide and the second is alkaline hydrolyse of ethyl benzoate in homogeneous hydro-alcoholic. Finally, the combination of the estimated parameter U using EKF and probabilistic neural network classifier provided the best results. These results show the performance of the proposed approach to monitoring the semi-batch and batch reactors.     

Detection of small leakage from long transportation pipeline with complex noise

In the long distance pipeline remote monitoring system, small leak detection becomes an important issue. Weak singularities in small leak signals are usually difficult to detect precisely under complicated noise background, which may cause false alarm or miss alarm. The advantage of applying the harmonic wavelet method is explored in this paper. Pipeline small leak sensitive characteristics are recognized and the negative pressure wave inflexions are extracted by harmonic wavelet analysis, expressed in terms of harmonic wavelet time-frequency mesh map, time-frequency contour map, and time-frequency profile plot. This paper also presents a comparative study of both Daubechies wavelet and harmonic wavelet analysis when applied to pipeline small leak detection under complicated background noises. Results of simulating test and field experiment show that it is possible to distinguish weak non-stationarities from complicated noises by harmonic wavelet analysis in pipeline small leak detection system. The comparison clearly illustrates that harmonic wavelet based pipeline small leakage detection method is significantly more accurate than other wavelets analysis such as Daubechies wavelet. This work provides a reliable and safe guarantee for oil and gas long distance transportation, reducing petroleum product losses and protecting surrounding environment.     

一种新的基于图模型的间歇过程故障监测方法

目前多元统计方法被广泛用于间歇过程故障监测并已经取得了比较好的效果。但是,统计模型的可解释性能比较差,很难直接利用操作人员积累的安全经验。为应对这些不足,提出了一种基于图模型的间歇过程故障监测方法。利用提出的定性建模方法,过程机理及操作人员的安全经验能够方便地表达。利用在线的过程变量数据和正向推理算法推断生产应处于的状态,利用安全知识及时地发现生产异常。当推断的结论与在线测得的结果矛盾或过程变量超过设定的安全限时,给出解释性输出。通过一个间歇反应案例,验证了提出的方法在模型的可解释性和利用安全经验方面的优势。     

Fault detection and diagnostics of a three-phase separator

A high demand of oil products on daily basis requires oil processing plants to work with maximum efficiency. Oil, water and gas separation in a three-phase separator is one of the first operations that are performed after crude oil is extracted from an oil well. Failure of the components of the separator introduces the potential hazard of flammable materials being released into the environment. This can escalate to a fire or explosion. Such failures can also cause downtime for the oil processing plant since the separation process is essential to oil production. Fault detection and diagnostics techniques used in the oil and gas industry are typically threshold based alarm techniques. Observing the sensor readings solely allows only a late detection of faults on the separator which is a big deficiency of such a technique, since it causes the oil and gas processing plants to shut down.A fault detection and diagnostics methodology for three-phase separators based on Bayesian Belief Networks (BBN) is presented in this paper. The BBN models the propagation of oil, water and gas through the different sections of the separator and the interactions between component failure modes and process variables, such as level or flow monitored by sensors installed on the separator. The paper will report on the results of the study, when the BBNs are used to detect single and multiple failures, using sensor readings from a simulation model. The results indicated that the fault detection and diagnostics model was able to detect inconsistencies in sensor readings and link them to corresponding failure modes when single or multiple failures were present in the separator.     

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。     

基于深度学习的输煤皮带故障视觉检测方法研究

为了更好地检测皮带跑偏、撕裂和异物干扰等严重影响皮带安全运行的故障状态,围绕相关问题产生的原因及检测方法开展深入研究,通过对纵/横向裂缝、异物的检测分析、实验,提高基于视觉的检测精度。提出基于Canny边缘检测算法的皮带跑偏检测算法;基于深度学习的横向与纵向撕裂检测,尤其对于裂缝与纵向纹理区分不明显情况,提出一种红光透射的判别方式;基于最小距离分类算法将识别异物转换为分类问题,利用机器学习的方法对样本进行训练并建立无异物阈值,通过提取特征,最后利用最小距离分类算法得到有无异物的结果。研究结果表明:提出的视觉检测系统可以实时高效地检测出输煤皮带常见的3种故障,可进一步保障运输系统安全运行。     

Dispersion modeling approach for quantification of methane emission rates from natural gas fugitive leaks detected by infrared imaging technique

Recently, infrared optical imaging has been applied in the oil and gas industry as a method to detect potential leaks in pipelines, components and equipment. The EPA suggested that this impending technique is considered as a smart gas LDAR (leak detection, monitoring and repair) for its rapid recognition of leaks, accuracy and robustness. In addition, compared to the conventional method using Total Vapor Analyzer (TVA) or gas sniffer, it has several other advantages, such as the ability to perform real-time scanning and remote sensing, ability to provide area measurement instead of point measurement, and provide an image of the gas which is not visible to naked eye. However, there is still some limitation in the application of optical imaging techniques; it does not give any measurement of gas emissions rates or concentrations of the leaking gas. Infrared cameras can recognize a target gas and distinguish the gas from its surrounding up to a certain concentration, namely the minimum detectable concentration. The value of the minimum detectable concentration depends on the camera design, environmental conditions and surface characteristics when the measurement is taken. This paper proposed a methodology to predict gas emissions rates from the size of the dispersed gas plume or cloud to the minimum detectable concentration. The gas emissions rate is predicted from the downwind distance and the height of the cloud at the minimum detectable concentration for different meteorological conditions. Gas release and dispersion from leaks in natural gas pipeline systems is simulated, and the results are presented.     

环空带压气井保护液液位测试方法研究

为了有效检测环空保护液位,评估气井环空带压情况,在传统回声法检测的基础上,提出了基于频谱分析及自相关分析的2种液位检测方法;搭建了液位检测试验系统,开展了不同环空压力下的液位检测试验,计算了不同压力下的环空声速;设计了FIR低通滤波器对液面回波信号进行滤波处理,同时利用了上述2种方法计算液位值并与实际值及传统液位计算方法对比,得出了声衰减系数并评判了衰减过程对液面回波信号频谱分析的影响程度。研究结果表明:设计的FIR低通滤波器适用于所建液位检测试验的滤波处理过程;频谱分析及自相关分析都能够有效检测液位高度,最大误差分别为1.65%和0.61%,自相关分析方法具有更高的精确度;声衰减系数整体较小,对液面回波信号频谱分析结果影响较小。     

水体监测中氨氮测定的影响因素分析

目前,水质监测的范围非常广泛,通常包括易受污染水体和未受污染水体的监测。一般在水质监测过程中,首先需要确定的是科学合理的检测方法,然后对影响该检测方法精度的因素进行综合分析并进行有效监测,当今水质监测中氨氮浓度是评价水质好坏的重要指标之一。基于此,文章综合分析了水体中氨氮浓度检测的基本原理并重点探讨了水体监测中影响氨氮测定的主要因素。     

Simulation-based fault propagation analysis—Application on hydrogen production plant

Recently production of hydrogen from water through the Cu–Cl thermochemical cycle is developed as a new technology. The main advantages of this technology over existing ones are higher efficiency, lower costs, lower environmental impact and reduced greenhouse gas emissions. Considering these advantages, the usage of this technology in new industries such as nuclear and oil is increasingly developed. Due to hazards involved in hydrogen production, design and implementation of hydrogen plants require provisions for safety, reliability and risk assessment. However, very little research is done from safety point of view. This paper introduces fault semantic network (FSN) as a novel method for fault diagnosis and fault propagation analysis by using evolutionary techniques like genetic programming (GP) and neural networks (NN), to uncover process variables’ interactions. The effectiveness, feasibility and robustness of the proposed method are demonstrated on simulated data obtained from the simulation of hydrogen production process in Aspen HYSYS^®. The proposed method has successfully achieved reasonable detection and prediction of non-linear interaction patterns among process variables.     

随钻压力波在用于早期气侵检测时的扰动传播特性研究*

为了研究随钻测量装置（Measurement While Drilling,MWD）压力波信号在用于早期气侵检测时的扰动传播特性,基于油气井多相流流动理论,建立随钻压力波在环空气液两相流中的扰动传播模型,对多参数影响下的压力波传播与衰减特性进行模拟,并对压力波检测技术的现场应用效果进行分析。结果表明:含气率、角频率、系统压力、虚拟质量力、拖曳力和壁面剪切力的变化都会对压力波在环空气液两相流中的传播与衰减特性造成不同程度的影响;相比于常规的全烃量检测技术,压力波检测技术可以更早地检测到气侵的发生,可进一步提高油气井建井的安全性。     

地下氡气测量法-音频大地电磁法在隐伏断层探测中的应用*

为开展隐伏断层探测，明确断层空间位置和构造属性，在此基础上合理避让或采取有效的工程措施，可以有效减轻地震灾害损失。在昔格达断裂带某区域，采用地下氡气测量法与音频大地电磁法相结合的方法探测隐伏断层。结果表明:2种方法所推断的断层区域较为吻合。地下氡气测量法简单方便，但无法确定断裂结构的深部延伸及产状变化，音频大地电磁法勘探深度大、效率高但干扰因素较多，具有多解性，二者结合的断层探测方法在研究区效果良好，对类似地区开展断层探测具有一定的借鉴和指导意义。     

Detecting fatigue in car drivers and aircraft pilots by using non-invasive measures: The value of differentiation of sleepiness and mental fatigue

Introduction: Fatigue is one of the most crucial factors that contribute to a decrease of the operating performance of aircraft pilots and car drivers and, as such, plays a dangerous role in transport safety. To reduce fatigue-related tragedies and to increase the quality of a healthy life, many studies have focused on exploring effective methods and psychophysiological indicators for detecting and monitoring fatigue. However, those fatigue indicators rose many discrepancies among simulator and field studies, due to the vague conceptualism of fatigue, per se, which hinders the development of fatigue monitoring devices. Method: This paper aims to give psychological insight of the existing non-invasive measures for driver and pilot fatigue by differentiating sleepiness and mental fatigue. Such a study helps to improve research results for a wide range of researchers whose interests lie in the development of in-vehicle fatigue detection devices. First, the nature of fatigue for drivers/pilots is elucidated regarding fatigue types and fatigue responses, which reshapes our understanding of the fatigue issue in the transport industry. Secondly, the widely used objective neurophysiological methods, including electroencephalography (EEG), electrooculography (EOG), and electrocardiography (ECG), physical movement-based methods, vehicle-based methods, fitness-for-duty test as well as subjective methods (self-rating scales) are introduced. On the one hand, considering the difference between mental fatigue and sleepiness effects, the links between the objective and subjective indicators and fatigue are thoroughly investigated and reviewed. On the other hand, to better determine fatigue occurrence, a new combination of measures is recommended, as a single measure is not sufficient to yield a convincing benchmark of fatigue. Finally, since video-based techniques of measuring eye metrics offer a promising and practical method for monitoring operator fatigue, the relationship between fatigue and these eye metrics, that include blink-based, pupil-based, and saccade-based features, are also discussed. To realize a pragmatic fatigue detector for operators in the future, this paper concludes with a discussion on the future directions in terms of methodology of conducting operator fatigue research and fatigue analysis by using eye-related parameters.     

基于视频图像的交通枢纽综合体人体检测

针对交通枢纽综合体人体检测的问题,提出一种基于梯度方向直方图(HOG)人体模型特征的检测算法。该方法通过提取人体样本库的HOG特征,用支持向量机算法(SVM)对样本的HOG特征进行分类训练。为了提高算法的精确度和适用性,以南京南站的监控视频为依据建立交通枢纽综合体人体样本库。并以南京南站监控视频和校园拍摄的人员视频作为测试集。结果证明,本算法可以有效识别交通枢纽综合体各种特征人体。     

Optimum parameters for fault detection and diagnosis system of batch reaction using multiple neural networks

Batch process usually differs from the continuous process because of its time-varying variables and the process parameters. An early detection and isolation of faults in the process will help to reduce the process upsets and keep it safe and reliable. This paper discusses on the application of multi-layer perceptron neural network in detecting various faults in batch chemical reactor based on an esterification process that involves the reaction of ethanol and acetic acid catalyzed by sulfuric acid. A multi-layer feed forward neural network with double hidden layers has been used in the neural network architecture. The detection was based on the different patterns generated between normal and faulty conditions. An optimum network configuration was found when the network produced the minimal error with respect to the training, testing and data validation.     

煤矿井下故障电信号快速检测电路的动态研究

对利用动态整定实现故障电信号快速检测电路的动态特性进行了分析和模拟实验。用数字贮存示波器（ＤＳＯ）记录了各参量的动态变化过程,结果表明,快速检测电路在动态下能对故障电信号进行快速检测,采样时间短