ANALYSIS AND DETECTION OF OUTLIERS AND SYSTEMATIC ERRORS IN INDUSTRIAL PLANT DATA

This article describes the analysis of industrial process data to detect outliers and systematic errors. Data reconciliation is an important step in adjusting mathematical models to plant data. The quality of the data directly affects the quality of adjustment of the model for modeling, simulation, and optimization purposes. To detect these errors in a multivariable system is not an easy task. If the origin of the abnormal values is known, these values can be immediately discarded. On the other hand, if an error or an extreme observation is not clearly justified, the decision whether or not to discard these values must be based on statistical analysis. In this work, in addition to process knowledge, the methodology employed involves an approach based on statistical analysis, first-principle equations, neural network models, and a composite of these. The neural network based approach was used to represent the process in order to classify similar inputs and outputs, i.e., to identify clusters. The elimination of gross errors was performed by the similarity principle or by hypothesis testing for means. The system studied is the Isoprene Production Unit of BRASKEM, the largest Brazilian petrochemical plant. The analysis of the process was undertaken by using a one-year database. The frequency of the data collection of the monitoring variables was 15 minutes.     

DETECTION OF GROSS ERRORS IN NONLINEARLY CONSTRAINED DATA: A CASE STUDY

The MIMT algorithm previously developed for gross error detection in linearly constrained systems was extended to nonlinear systems. The algorithm was tested by means of computer simulation using data from an industrial grinding circuit. The overall performance of the algorithm on the nonlinear system was found to be comparable to that exhibited on a linear system of approximately the same size. The algorithm correctly detected approximately 80% of all systematic errors in the data and achieved an average reduction in total error of more than 60%. The detection rate for the more significant (gross) systematic errors was approximately 90%. These results represent the first detailed performance evaluation of a gross error detection algorithm applied to a nonlinear system of industrial significance.     

ON THE CHOICE OF REGRESSION VARIABLES IN NONLINEAR DATA RECONCILIATION

Reconciliation of process data constrained by multicomponent material balance equations can be accomplished by a regression calculation in which either total flow rates and compositions (primary variables) or total and component flow rates (secondary variables) are used as the regression variables. It is shown by means of computer simulation experiments that the choice of secondary variables for regression can adversely affect the performance of an associated gross error detection algorithm. The principal effect is an increased tendency of the algorithm to make Type I errors     

ANALYSIS OF RANDOM AND SYSTEMATIC ERROR EFFECTS ON UNCERTAINTY PROPAGATION IN PROCESS DESIGN AND SIMULATION USING DISTRIBUTION TAIL CHARACTERIZATION

An exponential and heavy tail analysis method is presented to study the effect of systematic and random errors present in thermodynamic data on chemical process design and simulation. The true distribution tail characteristics (important for high levels of quality assurance) can be far from the estimates obtained with typical Gaussian distribution analysis. Pareto (heavy) or exponential distributions may represent the tail behavior better under many circumstances. Heavy tails diminish at an algebraic rate rather than at an exponential rate. Different error types such as random and systematic error can potentially cause different effects in the behavior of probability distributions, particularly in the tails. In this work, we use the tail behavior of cumulative frequency distributions produced from uncertainty analyses to characterize the error propagation in process design and simulation. The diminishing rate of the tail of a given distribution can be related to the error types involved in the process and also can be used to determine which error exhibits stochastic dominance. Case studies of process performance evaluations for liquid-liquid extraction operations are presented to illustrate the approach. It is observed that random and systematic errors (coupled with typical nonlinear chemical engineering models) can cause the tails of the uncertainty probability distributions to be exponential.     

ANALYSIS OF RANDOM AND SYSTEMATIC ERROR EFFECTS ON UNCERTAINTY PROPAGATION IN PROCESS DESIGN AND SIMULATION USING DISTRIBUTION TAIL CHARACTERIZATION

An exponential and heavy tail analysis method is presented to study the effect of systematic and random errors present in thermodynamic data on chemical process design and simulation. The true distribution tail characteristics (important for high levels of quality assurance) can be far from the estimates obtained with typical Gaussian distribution analysis. Pareto (heavy) or exponential distributions may represent the tail behavior better under many circumstances. Heavy tails diminish at an algebraic rate rather than at an exponential rate. Different error types such as random and systematic error can potentially cause different effects in the behavior of probability distributions, particularly in the tails. In this work, we use the tail behavior of cumulative frequency distributions produced from uncertainty analyses to characterize the error propagation in process design and simulation. The diminishing rate of the tail of a given distribution can be related to the error types involved in the process and also can be used to determine which error exhibits stochastic dominance. Case studies of process performance evaluations for liquid-liquid extraction operations are presented to illustrate the approach. It is observed that random and systematic errors (coupled with typical nonlinear chemical engineering models) can cause the tails of the uncertainty probability distributions to be exponential.     

炭素生产中监视和测量结果数据的分析方法与应用

在炭素产品生产的各个阶段，会产生大量监视和测量（结果的）数据。本文介绍了分析这些数据的几种方法，只要对所获得的数据进行有效的分析，通过持续改进，产品的质量就会得到逐步的提高。     

CLUSTER ANALYSIS OF MINERAL PROCESS DATA WITH AUTOASSOCIATIVE NEURAL NETWORKS

Experimental results indicate that autoassociative neural networks provide a robust method for the identification of clusters in process data. Cluster identification is accomplished by extracting a single feature from each multivariate data vector. The ranked features can be used to construct a feature curve, which is subsequently used as a basis for partitioning of the data space. In three case studies, involving two sets of ore samples, and a set of flotation froth features, with 11, 13 and 5 variables respectively, the clusters identified with the neural network appeared to be better than those obtained by conventional means.     

CLUSTER ANALYSIS OF MINERAL PROCESS DATA WITH AUTOASSOCIATIVE NEURAL NETWORKS

Experimental results indicate that autoassociative neural networks provide a robust method for the identification of clusters in process data. Cluster identification is accomplished by extracting a single feature from each multivariate data vector. The ranked features can be used to construct a feature curve, which is subsequently used as a basis for partitioning of the data space. In three case studies, involving two sets of ore samples, and a set of flotation froth features, with 11, 13 and 5 variables respectively, the clusters identified with the neural network appeared to be better than those obtained by conventional means.     

尿素造粒喷头性能测试中的误差分析

分析在尿素造粒喷头性能测试中容易产生误差的环节,介绍减少误差的办法。     

合成氨装置变换系统技术改造及其效果分析

介绍年产50万吨合成氨装置一氧化碳变换系统流程及变换技术特点,分析系统中存在的问题,针对性地提出改造措施。措施实施后从根本上消除了系统运行隐患,保证了变换系统的安全稳定运行,为全装置长周期、高负荷、低能耗运行提供了强有力的保障。     

乙烯装置绩效改进实施综述

介绍“SINO BEST”PPP项目在扬子乙烯实施的概况，以及实施的效果。     

数据协调中冗余性分析新方法

提出一种数据协调中的冗余性分析新方法 .根据尽量删除带有显著误差的变量 ,同时确保协调精度的原则 ,对所有检测出来的带有显著误差的变量 ,通过计算不同阶次的冗余度 ,按照冗余度的大小逐个选择可以删除的变量 .这样既充分利用了数据信息 ,又确保了协调精度 .仿真结果证实了该方法的有效性 .     

数据模型化的分析准则

实验数据的准确、可靠的模型关联是对过程进行有效模拟和设计的基础。通过气体吸附平衡数据的模型化的研究，结合统计技术的分析，对各种吸附平衡模型的适用性进行了探讨，包括提出的吸附平衡模型。讨论了依据统计分析进行模型选择应考虑的几个问题。     

中国西南山区氢能资源开发的工业经济分析

制氢有3种方法,其中电解水制氢的方法,在严格有限的范围内、在水力资源丰富、水电价格便宜的许多国家都在工业上普遍应用。中国西南山区是一个水力资源丰富、水电价格十分便宜的地区,水电资源可达200 kW·km~(-2)以上,占全国水力资源总量的67％。在这里建立梯级水电站群,并站联网后建立一个庞大的氢能生产基地,实现水能→电能→氢能的转变。在此地区利用水电解法获取氢能的成本仅为美国等先进国家氢能成本的1/10左右,在国内、国际市场具有很强的竞争力,是中国西部大开发中一项举足轻重的科学试验项目。     

化肥装置取用水现状与节水减排潜力分析

分析中石化巴陵分公司化肥事业部取用水现状,提出节水减排的工作思路和具体节水减排措施,分析化肥装置节水减排的潜力。     

数据处理与优化方法在化工试验中的应用

在化工试验过程中，选用适当的试验设计方法进行数据处理与优化，可以快捷，准确地得出结果数据。使试验次数大幅度降低。节约时间和经费，并达到最优化的方案和最好的试验结果。     

芳烃联合装置分离系统模拟与分析

建立了芳烃装置分离系统的数学模型,开发芳烃装置分离系统模拟软件APSS,并应用于现场装置的模拟计算分析,模拟值与设计值基本吻合。该软件可用于指导实际的生产操作和设备改造,优化生产操作参数,提高企业的计算机管理水平。     

PERFORMANCE EVALUATION OF PCA TESTS IN SERIAL ELIMINATION STRATEGIES FOR GROSS ERROR IDENTIFICATION

In this paper, the performance of the Principal Component Measurement Test (PCMT) is evaluated when used for the identification of multiple biases. A serial elimination strategy is implemented where a statistical test based on principal component analysis is used to identify the measurement to eliminate. A simulation procedure involving random measurement errors and fixed gross error sizes is applied to evaluate its performance. This performance is compared with the one obtained using serial elimination using the conventional Measurement Test (MT), as it is performed in some commercial simulators. The analysis indicates that principal component tests alone, without the aid of other collective tests, do not significantly enhance the ability in identification features of this strategy, performing worse in some cases. A few cases of severe failure of this strategy are shown and a suggestion to test other strategies is offered.     

IDENTIFICATION AND ROBUST CONTROL OF AN INDUSTRIAL STEAM REFORMER

The dynamic behaviour of a methane fired steam reformer is studied in an industrial environment. A system identification approach is used to determine transfer functions linking coil outlet temperature and fuel rate. All models have in common an unstable zero and large gain fluctuation over different operating conditions. A statistical clustering technique was used to determine the reactor dynamic model, for different operating conditions. To counteract the impact of gain fluctuation, a minimax error process model selection was used to determine a nominal model for the design of an IMC controller. The tuning of the controller was shown to be equivalent to the tuning of the time constant of the robustness filter. Improved temperature regulation can be used to increase the operating temperature of the reactor, thus leading to reducing the steam to carbon ratio and in turn improving process ecconomics.