基于Ms-LWPLS的化工过程网络化性能分级评估方法

针对化工过程输入输出数据间非线性关系问题,提出一种基于多数据空间局部加权潜结构映射(multispace locally weighted projection to latent structures,Ms-LWPLS)的网络化性能分级评估方法。该方法将历史数据分成不同性能等级的集合,利用Ms-LWPLS方法提取不同性能等级训练数据的过程变化,获得训练数据与性能等级标签之间的非线性映射结构,实现输入数据与性能等级之间的网络化"离线建模"。得到模型后,以数据滑动时间窗为评估单元,将滑动窗口数据输入到训练好的神经网络模型中,根据网络输出划分过程当前性能等级,并构造过渡性能系数,将稳态性能等级和过渡性能等级进行识别和区分。最后,将该方法应用到乙烯裂解过程在线性能评估中,说明此性能评估方法的有效性和准确性。     

基于Ms-LWPLS的化工过程网络化性能分级评估方法

针对化工过程输入输出数据间非线性关系问题，提出一种基于多数据空间局部加权潜结构映射(multi-space locally weighted projection to latent structures，Ms-LWPLS)的网络化性能分级评估方法。该方法将历史数据分成不同性能等级的集合，利用Ms-LWPLS方法提取不同性能等级训练数据的过程变化，获得训练数据与性能等级标签之间的非线性映射结构，实现输入数据与性能等级之间的网络化“离线建模”。得到模型后，以数据滑动时间窗为评估单元，将滑动窗口数据输入到训练好的神经网络模型中，根据网络输出划分过程当前性能等级，并构造过渡性能系数，将稳态性能等级和过渡性能等级进行识别和区分。最后，将该方法应用到乙烯裂解过程在线性能评估中，说明此性能评估方法的有效性和准确性。     

基于多数据空间全潜结构映射的化工过程性能评估方法

针对化工过程运行状态在线评估的问题,提出多数据空间全潜结构映射（multi-space total projection to latent structures,MsT-PLS）性能评估方法。该方法采用“离线建模,在线评估”的评估策略。首先对历史多数据输入空间进行全面分解,结合多数据空间基向量提取方法,剔除多数据输入空间中与质量变量无关信息的干扰。在与质量变量相关的多数据输入空间上,建立不同运行性能等级的离线数据网络分类模型,实现“离线建模”。“在线评估”阶段,以数据滑动时间窗为评估单元,将过程性能分为稳定和过渡性能等级,把在线数据与历史性能等级进行相似度匹配。利用过程变量相对贡献度,对性能变化起决定性影响的过程变量进行识别和贡献度分析,为系统性能劣化原因的识别提供了参考。最后,应用到乙烯裂解过程在线性能评估中,说明了本评估方法可以对系统进行准确的在线性能评估。     

基于多数据空间全潜结构映射的化工过程性能评估方法

针对化工过程运行状态在线评估的问题,提出多数据空间全潜结构映射(multi-space total projection to latent structures,Ms T-PLS)性能评估方法。该方法采用"离线建模,在线评估"的评估策略。首先对历史多数据输入空间进行全面分解,结合多数据空间基向量提取方法,剔除多数据输入空间中与质量变量无关信息的干扰。在与质量变量相关的多数据输入空间上,建立不同运行性能等级的离线数据网络分类模型,实现"离线建模"。"在线评估"阶段,以数据滑动时间窗为评估单元,将过程性能分为稳定和过渡性能等级,把在线数据与历史性能等级进行相似度匹配。利用过程变量相对贡献度,对性能变化起决定性影响的过程变量进行识别和贡献度分析,为系统性能劣化原因的识别提供了参考。最后,应用到乙烯裂解过程在线性能评估中,说明了本评估方法可以对系统进行准确的在线性能评估。     

基于相似度网格模型的在线经济性能分级评估方法

针对经济性能评估方法中目标函数难以在线计算问题提出一种基于过程数据的在线经济性能分级评估方法。采用自回归潜结构映射(AR-PLS)算法对输入数据矩阵进行分解,在与输出潜变量相关的子空间上建立不同性能等级的离线模型,从而排除无关变化的干扰。然后采用"先标定分区,再对比邻级相似度"的策略设计一个相似度网格模型,将过程性能分为稳定性能级状态和过渡状态,并对离线模型中未出现过的因素造成的性能变化进行识别,以进一步丰富离线数据库。对于不属于最优性能级的过程数据,能够根据变量贡献度诊断造成性能变差的原因。乙烯裂解过程的现场数据测试实验表明本方法可以及时、准确地检测到经济性能的偏移。     

基于相似度网格模型的在线经济性能分级评估方法

针对经济性能评估方法中目标函数难以在线计算问题提出一种基于过程数据的在线经济性能分级评估方法。采用自回归潜结构映射（AR-PLS）算法对输入数据矩阵进行分解,在与输出潜变量相关的子空间上建立不同性能等级的离线模型,从而排除无关变化的干扰。然后采用“先标定分区,再对比邻级相似度”的策略设计一个相似度网格模型,将过程性能分为稳定性能级状态和过渡状态,并对离线模型中未出现过的因素造成的性能变化进行识别,以进一步丰富离线数据库。对于不属于最优性能级的过程数据,能够根据变量贡献度诊断造成性能变差的原因。乙烯裂解过程的现场数据测试实验表明本方法可以及时、准确地检测到经济性能的偏移。     

带变遗忘因子的自适应子空间预测控制器设计

针对实际工业过程具有非线性、时变和多变量的特点,提出一种数据驱动的带有变遗忘因子的自适应子空间预测控制方法。该方法将在线子空间辨识与模型预测控制相结合,同时利用期望输出值与实际输出值的误差实现变遗忘因子的自适应更新,并根据当前变遗忘因子构造了过去与将来的Hankel矩阵,从而实现了预测模型的在线更新,提高了控制器对非线性时变特征的辨识灵敏度和适应能力。最后,利用该控制器对四容水箱对象进行仿真研究,验证了算法的有效性。     

基于KECA的非线性工业过程故障检测与诊断新方法

提出了一种基于核熵成分分析(kernel entropy component analysis,KECA)的非线性过程故障检测与诊断新方法。该方法首先利用KECA获取过程数据的得分向量及非线性特征子空间;然后鉴于KECA可以以角结构的方式揭示数据中潜在的集群结构,设计了基于角度的监测指标VoA。该指标通过各得分向量之间的角度方差来描述变换后数据间的结构差异,并根据角度方差的变化情况实现故障检测;接着,为了在检测到故障后有效地进行故障识别,构建了KECA相似度因子来度量特征子空间的相似程度以识别故障模式;最后,以非线性数值案例及Tennessee Eastman过程进行仿真测试研究,结果验证了所提方法的可行性及有效性。     

基于循环自动编码器的间歇过程故障监测

针对间歇过程数据非线性、动态性特征,提出一种基于循环自动编码器(recurrent autoencoder,RAE)的过程故障监测方法。采用长短时记忆(long short-term memory,LSTM)循环神经网络构建自动编码器建立监控模型,相比传统自动编码器,其能有效挖掘时序样本间的动态关联信息。该方法首先利用批次展开与变量展开相结合的三步展开方法将间歇过程数据展开成二维,并通过滑动窗采样得到模型输入序列;然后使用LSTM构建自动编码器,重构输入序列。进一步,利用重构误差构造平方预测误差(squared prediction error, SPE)统计量实现在线监测。最后将所提方法应用于青霉素发酵仿真和重组大肠杆菌发酵过程监测,结果表明,该方法能及时监测到故障,具有较好的监测性能。     

基于二子空间协同训练算法的半监督软测量建模

在工业过程中,存在着辅助变量与主导变量数据比例严重失衡的问题。协同训练算法是其中一种利用无标签数据中的潜在信息以提升学习性能的模型训练方法。然而目前在协同训练软测量建模过程中,学习器之间存在严重的训练特性交叉重叠的问题,这将导致对主导变量的预测性能衰减。针对这一问题,提出基于二子空间协同训练算法的半监督软测量模型two-subspace co-training KNN(TSCO-KNN)。该模型将二子空间分块算法与协同训练算法相结合,利用辅助变量与主成分子空间PCS和残差子空间RS两个特征子空间的相关性程度,将数据变量拆分为两个具有显著差异性的学习数据集,进而使用KNN回归器进行协同训练,共同用于对主导变量的预测。最后在乙烯精馏塔塔顶乙烷浓度和TE过程产品浓度软测量中进行仿真研究,验证本文所提算法的有效性。     

Online complex nonlinear industrial process operating optimality assessment using modified robust total kernel partial M-regression

Although industrial processes often perform perfectly under design conditions,they may deviate from the optimal operating point owing to parameters drift,environmental disturbances,etc.Thus,it is necessary to develop efficacious strategies or procedure to assess the process performance online.In this paper,we explore the issue of operating optimality assessment for complex industrial processes based on performance-similarity considering nonlinearities and outliers simultaneously,and a general enforced online performance assessment framework is proposed.In the offline part,a new and modified total robust kernel projection to latent structures algorithm,T-KPRM,is proposed and used to evaluate the complex nonlinear industrial process,which can effectively extract the optimal-index-related process variation information from process data and establish assessment models for each performance grades overcoming the effects of outlier.In the online part,the online assessment results can be obtained by calculating the similarity between the online data from a sliding window and each of the performance grades.Furthermore,in order to improve the accuracy of online assessment,we propose an online assessment strategy taking account of the effects of noise and process uncertainties.The Euclidean distance between the sliding data window and the optimal evaluation level is employed to measure the contribution rates of variables,which indicate the possible reason for the non-optimal operating performance.The proposed framework is tested on a real industrial case:dense medium coal preparation process,and the results shows the efficiency of the proposed method comparing to the existing method.     

Novel soft sensor method for detecting completion of transition in industrial polymer processes

Soft sensors are widely used to estimate process variables that are difficult to measure online. In polymer plants that produce various grades of polymers, the quality of products must be estimated using soft sensors in order to reduce the amount of off-grade material. However, during grade transition, the predictive accuracy deteriorates because the state in polymer reactors is unsteady, causing the values of process variables to differ from the steady-state values used to construct regression models. Therefore, we have proposed to construct models that detect the completion of transition to ensure that the polymer quality evaluated after transition conforms to the predicted one. By using these models and regression models constructed for each product grade, the polymer quality can be predicted with high accuracy, selecting a regression model appropriately. The proposed method was applied to industrial plant data and was found to exhibit higher predictive performance than traditional methods.     

Between‐phase calibration modeling and transition analysis for phase‐based quality interpretation and prediction

The quality‐concerned between‐phase transition analysis is performed and an improved calibration modeling strategy is designed for quality prediction and interpretation in multiphase batch processes. From the between‐phase viewpoint, the quality‐related phase behaviors are decomposed and two subspaces are separated. In common subspace, the underlying quality‐relevant variation stays invariable between the neighboring phases, showing the common contribution to quality. The other part changes with the alternation of phases and has the different influences on quality interpretation, termed specific subspace here. Based on subspace separation, between‐phase transition regions are distinguished from steady phases. Different models are developed in steady phases and transition regions respectively for online quality prediction. Offline quality analyses are also conducted in two subspaces to explore the time cumulative effects. The proposed method gives an interesting insight into the phase behaviors and between‐phase transitions for quality prediction. The feasibility and performance of the proposed method are illustrated with a typical multiphase batch process. © 2012 American Institute of Chemical Engineers AIChE J, 59: 108–119, 2013     

基于多尺度核JYMKPLS迁移模型的间歇过程产品质量的在线预测方法

针对过程数据不足,且具有强非线性和多尺度特性的新间歇过程,结合迁移学习方法与多尺度核学习方法的优势,提出了一种基于多尺度核JYMKPLS（Joint-Y multi-scale kernel partial least squares）迁移模型的间歇过程产品质量在线预测方法。该方法首先通过迁移学习利用相似源域的旧过程数据提高新间歇过程建模效率和质量预测的精度。然后,针对间歇过程数据的非线性和多尺度特性问题,引入了多尺度核函数以更好地拟合数据变化的趋势,从而提高模型的预测精度。此外,提出模型在线更新和数据剔除,通过在线持续改善迁移模型对新间歇过程的匹配程度,以消除相似过程间的差异性给迁移学习带来的不利影响,从而不断地提升预测精度。最后,通过仿真验证了所提方法的有效性,结果表明,与传统的数据驱动建模方法相比,本文所提方法能够有效提高建模效率和预测精度。     

基于核函数的加权极限学习机污水处理在线故障诊断

污水生化处理中的运行故障会引起出水水质不达标、运行费用增高和环境二次污染等严重问题,需要及时准确地对运行故障进行诊断。考虑到污水处理过程运行状态数据的不平衡性造成故障诊断准确率下降,提出了一种基于核函数的加权极限学习机污水处理过程实时在线故障诊断方法。该方法以极限学习机为基础,采用加权的方式处理数据的不平衡特性,通过核函数的非线性映射来提高数据线性可分的程度。仿真实验证明,本文建立的污水处理在线故障诊断模型在线测试精度高,泛化性能好,模型在线更新速度快,能够比较好地满足准确性和实时性,实现对污水处理过程的在线故障诊断。     

Optimal eigenvectors of spectral datasets: sequential selection from one set vs a collection from two sets

In this paper, the influence of spectral datasets and the method of selection of the corresponding feature vectors on the compression and reconstruction of data is scrutinised. To fulfil this aim, two different sets of reflectance data with the least spectral similarity are selected from different sets of spectral databases and the most optimal eigenvectors are chosen using different strategies. Six and 12 arrangements of eigenvectors obtained from different individual or combined databases are then used for the compression of reflectance spectra of learning sets, as well as those that have not been used in extraction of eigenvectors. The validity of the desired reduced subspaces is assessed by computing the spectral errors between the actual and the reconstructed spectra of samples of learning sets. Moreover, the efficiencies of designed compressed subspaces are evaluated through the numbers of out‐of‐range reconstructed spectra, as well as the spectral and colorimetric deviations between the actual and compressed‐reconstructed reflectance spectra of samples of datasets that were not employed in learning sequence. The results show that in the restricted subspaces, i.e. six‐dimensional subspace, the most effective results are achieved when the reduced subspace is created from a collection of two separate sets of eigenvectors of two different datasets with the maximum degree of dissimilarity, and the reduced spaces that have been made from six eigenvectors of individual datasets lead to higher errors.     

Dynamic optimization for grade transition processes using orthogonal collocation on molecular weight distribution

To meet the demands of a competitive market, an industrial plant often produces several grades of polymer product through the same process in an economical way. As molecular weight distribution (MWD) is a crucial quality index of polymers, dynamic optimization for grade transition based on MWD is highly important, but challenging. This study considers the development of optimization models for MWD-based grade transition. An MWD reconstruction method using orthogonal collocation in two dimensions is developed to capture the dynamic feature of MWD in time and the distributive feature in chain length. The simultaneous collocation approach is adopted to discretize the model. Two optimization formulations are proposed to describe minimizing the transition time as well as off-spec production. Both formulations inherit the advantages of the simultaneous collocation approach. The numerical results show that the proposed methods can efficiently solve the grade transition problem with MWD specification, and obtain high performance control profiles to reduce the production cost. © 2019 American Institute of Chemical Engineers AIChE J, 65: 1198–1210, 2019