基于图模型的间歇过程误操作风险辨识

提出了一套基于图模型的建模方法,该模型以Petri网和符号有向图（SDG）为基础将模型分为上部和下部,加入操作点、控制库所、判断模块等模型元素,并结合间歇过程的特点,提出了关联变量、目标变量检查表、操作点检查表和关联变量检查表等概念。根据误开误关、过早过晚、步骤添加删除等误操作,对模型进行验证。验证结果表明,该模型解决了先前方法难以描述具体操作的缺陷,并且模型结构简单,具有整体性。不仅能够方便地描述3种不同类型的误操作,还能有效地对误操作进行风险辨识,具有一定的适用性。     

间歇过程自动化的进展

本文从控制角度分析间歇过程的特征，并介绍了间歇控制装置和控制软件的现状和发展趋势，特别是配方管理和生产调度在线应用。     

间歇结晶过程的分析

本文概要地介绍了用于研究在间歇悬浮结晶器中的结晶动力学和晶体粒度分布（CSD）的实验技术和数据分析的新进展,讨论了包括最大CSD的特征化、累积CSD方法、热响应技术、最大允许成长速率在内的几个简单而有用的方法。     

间歇生产过程的自动控制

本文简要介绍了间歇生产过程及其特点,结合间歇生产过程对控制的要求,提出了相应的自动控制方案。阐述了间歇生产过程自动化的实现方法。     

间歇化工过程与计算机应用

首先指出间歇过程是一门独立的、有自身特点的、在化学工业中具有独特地位的学科。然后扼要回顾了计算机在间歇过程设计和生产调度中的应用，并介绍了几个典型的用于间歇过程的软件。最后简单介绍了由ＩＳＡ颁布的间歇控制标准ＳＰ８８，以便读者了解该领域的最新国际动态     

基于多向核熵成分分析的微生物发酵间歇过程监测研究

针对微生物发酵间歇过程监测算法只考虑数据信息最大化未考虑数据簇结构信息的不足,提出了基于多向核熵成分分析(Multi-way Kernel Entropy Component Analysis,MKECA)间歇过程监测的新方法。该方法首先引入AT展开策略对三维历史数据进行预处理,然后通过核映射将数据从低维空间映射到高维特征空间,解决数据的非线性特性,并在高维特征空间依据核熵的大小对数据进行降维,使降维后的数据能够最大化地保留原始数据的分布;同时理论证明了所提方法在特定条件下等同于传统方法,也就是说MKECA既能兼顾传统方法的优势,又能弥补传统方法的不足;最后通过青霉素仿真数据进行验证,表明MKECA方法具有更可靠的监控性能,能及时、准确地监测出故障。     

间歇生产过程的控制系统

论述了间歇生产过程控制的特点、硬件系统的选择原则和软件的开发方法及功能 ,并具体介绍了一个间歇精细化工的特殊生产过程控制的实例 ,对从事间歇生产过程控制开发、设计工作的人员 ,具有实用参考价值     

基于MKPLS和SQP方法的间歇过程迭代优化控制

采用多向核偏最小二乘（MKPLS）方法建立间歇过程的模型并进行操作条件的优化。由于存在模型失配和未知扰动，基于MKPLS模型的最优控制轨迹在实际对象上往往难以实现最优的产品质量指标。本文利用间歇过程批次间的重复特性与序贯二次规划（SQP）优化算法中迭代计算的相似特点，提出了一种基于MKPLS模型的批次间优化调整策略，使得经过逐步优化调整得到的控制轨迹作用于实际对象时，可以得到更优的质量指标。该方法的有效性在苯乙烯聚合反应器和乙醇流加发酵过程的仿真对象上得到了验证。     

基于MPCA-MDPLS的间歇过程的故障诊断

针对间歇过程的故障诊断问题,提出了一种新的混合模型方法——MPCA-MDPLS.这种方法包括两个模型：多向主元分析(MPCA)模型和多向判别部分最小二乘(MDPLS)模型.这两个模型的建模数据不仅包括正常工况的数据,而且还包含了各种已知故障数据.因此,MPCA模型具有检测未知故障的能力.给出了MDPLS模型故障诊断限,对经MPCA模型检测不是未知故障的故障做进一步诊断.如果故障是未知的,可以采取其他的方法来分析新的故障,并按不同类别存入到数据库中.当多次出现这种故障之后（一般≥5次）,把新的故障数据加入到建模数据中,并重新建立MPCA-MDPLS模型.通过对实际工业链霉素发酵过程数据的分析,表明了提出的算法是可行的、有效的,并具有识别未知新故障的能力.     

一种新的间歇过程故障诊断策略

间歇过程的在线故障诊断近年来受到了越来越多的关注,目前比较通用的方法主要是多变量统计的方法。然而在实际过程尤其是多阶段的间歇过程中故障诊断效果往往不够理想,误诊率比较高。为解决上述问题,本文基于动态轨迹分析（DLA）和在线的动态时间规整方法（DTW）,将二者的优点有效地结合在一起提出了一种在线故障诊断策略,提高了故障诊断效率和准确性。青霉素发酵过程的在线故障诊断应用实例表明该方法具有比较好的诊断效果。     

基于核Fisher包络分析的间歇过程故障诊断

随着间歇过程越来越受重视,其过程监控和故障诊断技术也成为研究热点。以核Fisher判别分析为基础,提出了基于核Fisher的正常工况与故障包络面模型,给出了基于该模型的在线故障诊断流程。此方法利用了Fisher判别分析对类别的划分特点,分别针对正常工况数据和各故障类型数据建立包络曲面模型。与多向Fisher判别分析相比,该方法按批次方向将数据展开,能够解决生产周期不一致问题,在线故障诊断时也不需要预报完整的生产轨迹,并且加入核函数来处理复杂的非线性。最后在青霉素发酵过程的仿真平台上对该方法进行测试,与改进多向Fisher判别分析方法进行对比,该方法获得了满意的诊断效果：能够及早诊断出故障的发生,并在有效识别已有故障的同时还具有对新故障的自学习能力。     

Enhanced process comprehension and quality analysis based on subspace separation for multiphase batch processes

Phase‐based subpartial least squares (subPLS) modeling algorithm has been used for online quality prediction in multiphase batches. It strictly assumes that the X – Y correlations are identical within the same phase so that they can be defined by a uniform regression model. However, the accuracy of this precondition has not been theoretically checked when put into practical application. Actually it does not always agree well with the real case and may have to be rejected for some practical processes. In the present work, it corrects the “absolute similarity” of subPLS modeling by a more general recognition that only one part of the underlying correlations are time‐wise common within the same phase while the other part are time‐specific, which is referred to as “partial similarity” here. Correspondingly, a two‐step phase division strategy is developed, which separates the original phase measurement space into two different parts, the common subspace and uncommon subspace. It is only in the common subspace where the underlying X – Y correlations are similar, a phase‐unified regression model can be extracted for online quality prediction. Moreover, based on the subspace separation, offline quality analyses are conducted in both subspaces to explore their respective cumulative manner and contribution in quality prediction. The strength and efficiency of the proposed algorithm are verified on a typical multiphase batch process, injection molding. © 2010 American Institute of Chemical Engineers AIChE J, 2011     

基于DHSC的多模态间歇过程测量数据异常检测方法

多模态间歇过程测量数据异常直接影响数据驱动的多元统计分析过程建模的准确性,导致间歇过程的监控性能降低。针对多模态间歇过程测量数据异常问题,提出了一种基于动态超球结构变化（DHSC）的多模态间歇过程测量数据异常检测方法。该方法通过引入时序约束的模糊C均值聚类（SCFCM）,利用隶属度变化划分多模态间歇过程的模态;针对不同模态,采用支持向量数据描述（SVDD）建立基于训练数据的静态超球体和基于待检数据的动态超球体,选择重要的支持向量作为球体结构,进而通过识别超球体发生结构变化实现过程测量数据异常检测。青霉素发酵过程仿真实验表明,所提出的方法能够实现多模态间歇过程的模态划分,减少了模态切换对过程测量数据异常检测精度的影响,并能够根据超球体结构变化检测过程测量数据异常,具有较高的检测精度,降低了误检率。     

Nonlinear quality prediction for multiphase batch processes

Typically, a multiphase batch process comprises several steady phases and transition periods. In steady phases, the data characteristics remain similar during the phase and have a significant repeatability from batch to batch; thus most data nonlinearities can be removed through the batch normalization step. In contrast, in each transition period, process observations vary with time and from batch to batch, so nonlinearities in the data may not be eliminated through batch normalization. To improve quality prediction performance, an efficient nonlinear modeling method—relevance vector machine (RVM) was introduced. RVMs were formulated for each transition period of the batch process, and for combining the results of different process phases. For process analysis, a phase contribution index and a variable contribution index are defined. Furthermore, detailed performance analyses on the prediction uncertainty and variation were also provided. The effectiveness of the proposed method is confirmed by an industrial example. © 2011 American Institute of Chemical Engineers AIChE J, 58: 1778–1787, 2012     

Consequence Identification for Maloperation in Batch Process

Batch processes are important in chemical industry, in which operators usually play a major role and hazards may arise by their inadvertent acts. In this paper, based on hazard and operability study and concept of qualitative simulation, an automatic method for adverse consequence identification for potential maloperation is proposed. The qualitative model for production process is expressed by a novel directed graph. Possible operation deviations from normal operating procedure are identified systematically by using a group of guidewords. The proposed algorithm is used for qualitative simulation of batch processes to identify the effects of maloperations. The method is illustrated with a simple batch process and a batch reaction process. The results show that batch processes can be simulated qualitatively and hazards can be identified for operating procedures including maloperations. After analysis for possible plant maloperations, some measures can be taken to avoid maloperations or reduce losses resulted from maloperations.     

Covariance‐oriented qualitative and quantitative calibration analysis for multistage batch processes

An improved stage‐specific multivariate calibration scheme is developed for multistage batch processes based on the covariance analysis unit. First, the process duration is automatically and properly divided into different stages, which reveals the changes of quality‐related process correlation characteristics. The concept of stage‐representative average process behaviour is then introduced, which is comprehensibly realized by averaging all covariance patterns within the same stage based on different weights. In this way, it stacks the cumulative effects of process variations on quality within each stage and meanwhile considers their time‐varying characteristics. Subsequently, covariance‐oriented OSC and variable selection are effectively combined, which can simplify the calibration model structure and enhance the causal relationship between predictors and quality by excluding the redundant latent factors and input variables. Finally, stage‐representative PLS regression models are developed focusing on the critical‐to‐quality stages for online quality prediction. It shows that a complete multistage calibration solution is readily achieved from an “overall” stage perspective by smartly making use of covariance. The illustration study to injection molding shows the effectiveness of the proposed method for improving process comprehension and quality prediction.