基于VAE-DBN的故障分类方法在化工过程中的应用

针对化工过程高维数据的故障特征难以提取的难题,提出变分自动编码器(VAE)结合深度置信网络(DBN)的混合故障诊断方法. 在VAE的编码部分对隐变量空间Z添加约束,通过重参数化方法进行反向传播训练,可无监督地学习不同故障对应的隐变量特征,其作为DBN分类模型的输入特征训练网络,输入测试集进行故障诊断. 田纳西伊斯曼流程(TE)应用结果表明,VAE能提取原始数据更加抽象有效的特征,VAE?DBN分类准确.     

Research on TE process fault diagnosis method based on DBN and dropout

In recent years, deep learning has shown outstanding performance and potential in pattern recognition and feature extraction, which has attracted an increasing amount of attention from engineering researchers and academics. Fault diagnosis methods based on deep learning have also become the focus of a significant amount of research. In this paper, a nonlinear process fault diagnosis and identification method based on DBN-dropout is proposed. The deep belief network (DBN) has significant advantages in dealing with nonlinear processes, and it can extract the abstract representation of nonlinear process data to build a deep network to achieve the real-time monitoring of process operations. Dropout technology can reduce overfitting and improve the generalization ability of the model. Afterwards, the Tennessee Eastman (TE) process is employed to analyze the performance of the proposed approach.     

基于数据特征提取的AANN-ELM研究及化工应用

针对极限学习机不能有效解决化工过程中高维数据建模的问题,本文将其与自联想神经网络结合,通过自联想神经网络过滤输入数据中存在的冗余信息、提取特征分量,并对所提取的特征分量采用极限学习机进行训练,由此形成了一种基于数据特征提取的AANN-ELM(auto-associative neural network-extreme learning machine)神经网络。同时,以UCI标准数据集进行测试,以精对苯二甲酸(PTA)溶剂系统进行验证,结果表明,AANN-ELM在处理高维数据时具有学习速度快、网络稳定性强、建模精度高的特点,为神经网络在复杂化工生产中的应用提供了新思路。     

基于稀疏过滤特征学习的化工过程故障检测方法

过程安全一直以来是化学工业中尤为重要的问题之一,故障检测与诊断（FDD）作为化工异常工况管理最有力的工具之一,给过程安全提供了保障。随着深度学习的发展,很多智能学习算法已经被提出,然而这些算法却很少被应用到FDD中来。提出了一种基于稀疏过滤和逻辑回归（SFLR）算法的化工过程故障检测新方法。采用TE过程和环己烷无催化氧化制环己酮过程对提出的方法进行了验证,结果表明,所提出的方法均具有较高的诊断精度,案例研究表明提出的方法可以及时有效地诊断出故障。     

基于TA-ConvBiLSTM的化工过程关键工艺参数预测

化工过程中，掌握关键工艺参数的变化趋势对于消除潜在波动、维持工况稳定作用巨大。然而，传统的浅层静态模型很难对非线性和动态性显著的复杂序列数据进行精准预测。针对上述难题，提出一种深度预测模型TA-ConvBiLSTM，将卷积神经网络(convolutional neural networks, CNN)和双向长短时记忆网络(bi-directional long short term memory, BiLSTM)集成到统一框架内，使其不仅能在每个时间步上自动挖掘高维变量间的隐含关联，更能横跨所有时间步自适应提取有用的深层时序特征。此外，引入时间注意力(temporal attention, TA)机制，为反映目标变化规律的重要信息增加权重，避免其因输入序列过长、深层特征太多而被掩盖。所提出方法的有效性在国内某延迟焦化装置炉管温度预测的案例中得到验证。     

基于极深因子分解机的化工过程故障诊断方法

近年来,随着化学工艺、设备的复杂化和大型化程度不断深入,如何为化工企业及时、准确地诊断故障、排除事故,成为一个极具挑战性的问题。目前,以深度学习为代表的化工过程故障检测与诊断技术成为业界解决问题的主要思路之一,但现有深度方法在构建诊断模型时只关注了变量的非线性高阶特征,不能充分、全面地挖掘多源交互信息,难以有效地融合各类异构数据。基于此,提出一种基于极深因子分解机的化工过程故障诊断方法,通过并行融合三类不同网络模型(分解机模型、深度神经网络模型、压缩交互网络模型),实现对高阶、低阶及线性特征的自动提取和高效整合。为了评估模型性能,从单故障诊断和多故障混合诊断的角度出发,在田纳西-伊斯曼过程(TE)仿真数据集上进行了广泛的对比实验,结果表明,所提方法较以往故障诊断方法在精确率和召回率等指标上具有明显优势。     

基于DBN-ELM的聚丙烯熔融指数的软测量

针对聚丙烯熔融指数软测量中预测精度不高的缺点,将基于深度置信网络-极限学习机（DBN-ELM）的软测量方法应用到熔融指数的软测量中。与传统深度置信网络（DBN）不同的是,该方法将极限学习机（ELM）算法运用到深度置信网络的训练中。首先用深度置信网络对原始数据进行数值分析来提取特征,然后将提取的特征输入到极限学习机中进行训练,得到软测量模型。实验验证表明,与支持向量机和单纯的深度置信网络模型相比,该方法具有更高的测量精度。     

A nonlinear kernel Gaussian mixture model based inferential monitoring approach for fault detection and diagnosis of chemical processes

A nonlinear kernel Gaussian mixture model (NKGMM) based inferential monitoring method is proposed in this article for chemical process fault detection and diagnosis. Aimed at the multimode non-Gaussian process with within-mode nonlinearity, the developed NKGMM approach projects the operating data from the raw measurement space into the high-dimensional kernel feature space. Thus the Gaussian mixture model can be estimated in the feature space with each component satisfying multivariate Gaussianity. As a comparison, the conventional independent component analysis (ICA) searches for the non-Gaussian subspace with maximized negentropy, which is not equivalent to the multi-Gaussianity in multimode process. The regular Gaussian mixture model (GMM) method, on the other hand, assumes the Gaussianity of each cluster in the original data space and thus cannot effectively handle the within-mode nonlinearity. With the extracted kernel Gaussian components, the geometric distance driven inferential index is further derived to monitor the process operation and detect the faulty events. Moreover, the kernel Gaussian mixture based inferential index is decomposed into variable contributions for fault diagnosis. For the simulated multimode wastewater treatment process, the proposed NKGMM approach outperforms the ICA and GMM methods in early detection of process faults, minimization of false alarms, and isolation of faulty variables of nonlinear and non-Gaussian multimode processes.     

A novel dynamic bayesian network‐based networked process monitoring approach for fault detection,propagation identification,and root cause diagnosis

A novel networked process monitoring, fault propagation identification, and root cause diagnosis approach is developed in this study. First, process network structure is determined from prior process knowledge and analysis. The network model parameters including the conditional probability density functions of different nodes are then estimated from process operating data to characterize the causal relationships among the monitored variables. Subsequently, the Bayesian inference‐based abnormality likelihood index is proposed to detect abnormal events in chemical processes. After the process fault is detected, the novel dynamic Bayesian probability and contribution indices are further developed from the transitional probabilities of monitored variables to identify the major faulty effect variables with significant upsets. With the dynamic Bayesian contribution index, the statistical inference rules are, thus, designed to search for the fault propagation pathways from the downstream backwards to the upstream process. In this way, the ending nodes in the identified propagation pathways can be captured as the root cause variables of process faults. Meanwhile, the identified fault propagation sequence provides an in‐depth understanding as to the interactive effects of faults throughout the processes. The proposed approach is demonstrated using the illustrative continuous stirred tank reactor system and the Tennessee Eastman chemical process with the fault propagation identification results compared against those of the transfer entropy‐based monitoring method. The results show that the novel networked process monitoring and diagnosis approach can accurately detect abnormal events, identify the fault propagation pathways, and diagnose the root cause variables. © 2013 American Institute of Chemical Engineers AIChE J, 59: 2348–2365, 2013     

Chemical process fault diagnosis based on a combined deep learning method

The study on fault detection and diagnosis (FDD) of chemical processes has always been the top priority of the chemical process safety. In this paper, a fault diagnosis method combining the deep convolutional with the recurrent neural network (DCRNN) is proposed. In this method, the data from chemical processes are input to the deep convolutional neural network (DCNN) to extract features in spatial domains, and then, the features are fused into the bidirectional recurrent neural network (BRNN). Due to the powerful capabilities of DCNN to extract features in spatial domains and the sensitivity to time series of RNN, the combined method can adaptively learn the dynamic information of the raw data in both spatial and temporal domains and has unique advantages in multivariate chemical processes. The application of the DCRNN model in the Tennessee Eastman (TE) process demonstrates the high accuracy of this proposal in identifying abnormal conditions for the chemical process, compared with the traditional fault identification algorithms of deep learning.     

Deep learning technique for process fault detection and diagnosis in the presence of incomplete data

In modern industrial processes, timely detection and diagnosis of process abnormalities are critical for monitoring process operations. Various fault detection and diagnosis (FDD) methods have been proposed and implemented, the performance of which, however, could be drastically influenced by the common presence of incomplete or missing data in real industrial scenarios. This paper presents a new FDD approach based on an incomplete data imputation technique for process fault recognition. It employs the modified stacked autoencoder, a deep learning structure, in the phase of incomplete data treatment, and classifies data representations rather than the imputed complete data in the phase of fault identification. A benchmark process, the Tennessee Eastman process, is employed to illustrate the effectiveness and applicability of the proposed method.     

基于深度学习的化工过程故障检测与诊断研究综述

化工过程的故障检测与诊断对于现代化工系统的可靠性和安全性具有重要意义.深度学习作为一项新兴的技术,引起了学术界和工业界的广泛关注.从方法的角度出发,将基于深度学习的化工过程故障检测与诊断技术分为:基于自动编码器的方法、基于深度置信网络的方法、基于卷积神经网络的方法和基于循环神经网络的方法,并分别对4种方法的最新研究进展...     

A Kernel Time Structure Independent Component Analysis Method for Nonlinear Process Monitoring

Kernel independent component analysis (KICA) is a newly emerging nonlinear process monitoring method, which can extract mutually independent latent variables called independent components (ICs) from process variables. However, when more than one IC have Gaussian distribution, it cannot extract the IC feature effectively and thus its monitoring performance will be degraded drastically. To solve such a problem, a kernel time structure independent component analysis (KTSICA) method is proposed for monitoring nonlinear process in this paper. The original process data are mapped into a feature space nonlinearly and then the whitened data are calculated in the feature space by the kernel trick. Subsequently, a time structure independent component analysis algorithm, which has no requirement for the distribution of ICs, is proposed to extract the IC feature. Finally, two monitoring statistics are built to detect process faults. When some fault is detected, a nonlinear fault identification method is developed to identify fault variables based on sensitivity analysis. The proposed monitoring method is applied in the Tennessee Eastman benchmark process. Applications demonstrate the superiority of KTSICA over KICA.     

Deep learning technique for process fault detection and diagnosis in the presence of incomplete data

In modern industrial processes, timely detection and diagnosis of process abnormalities are critical for monitoring process operations. Various fault detection and diagnosis (FDD) methods have been proposed and implemented, the performance of which, however, could be drastically influenced by the common presence of incomplete or missing data in real industrial scenarios. This paper presents a new FDD approach based on an incomplete data imputation technique for process fault recognition. It employs the modified stacked autoencoder, a deep learning structure, in the phase of incomplete data treatment, and classifies data representations rather than the imputed complete data in the phase of fault identification. A benchmark process, the Tennessee Eastman process, is employed to illustrate the effectiveness and applicability of the proposed method.     

基于约翰逊转换的鲁棒化工过程监控方法

化工厂中一个小故障可能导致大事故,从而造成生命财产损失和环境破坏。为了防止小故障演变成大事故,化学工业需要有效的过程监控来及时检测故障和诊断故障原因。传统化工过程监控方法主元分析法(Principal Component Analysis, PCA)假设数据服从高斯分布,实践中有时并不满足该条件。此外,其使用方差、协方差捕捉数据非线性变化时,鲁棒性较差。本工作提出一种改进的主元分析法—基于约翰逊转换的鲁棒过程监控方法。首先引入约翰逊正态转换(Johnson Transformation)使过程数据服从高斯分布;其次使用鲁棒性强的斯皮尔曼相关系数(Spearman Correlation Coefficient)矩阵代替传统主元分析法的协方差矩阵提取特征向量,构造特征空间;最后将过程数据投影到特征空间,使用T2和SPE统计量实施过程监控。将此方法应用于TE过程故障案例,并与PCA和核主元分析法(Kernel Principal Component Analysis, KPCA)对比,验证了此方法的有效性。     

人工神经网络在化工过程中的应用进展

人工神经网络作为人工智能的重要组成部分,以其超强的鲁棒性、容错性、可充分逼近任何复杂的非线性关系、并行处理、可学习和自适应等优点在改善化工过程传统生产技术诊断滞后、难以优化控制、物性估算误差较大以及不能处理非线性复杂情况等问题上有着广阔的发展空间。本文概述了人工神经网络的原理和发展历程,综述并分析了人工神经网络在故障诊断、过程控制和优化、质量控制、定量结构-活性/性质相关性分析、物性估算、专家系统以及聚类分析等化工过程中的应用原理以及研究进展和现状。最后指出卷积神经网络等深度学习算法的性能高、速度快,在化工过程中发展和应用深度学习算法将成为其发展方向和研究热点。     

Fault diagnosis of chemical processes based on partitioning PCA and variable reasoning strategy

Fault detection and identification are challenging tasks in chemical processes, the aimof which is to decide out of control samples and find fault sensors timely and effectively. This paper develops a partitioning principal component analysis (PPCA) method for process monitoring. A variable reasoning strategy is proposed and applied to recognize multiple fault variables. Compared with traditional process monitoring methods, the PPCA strategy not only reflects the local behavior of process variation in each model (each direction of principal components), but also improves the monitoring performance through the combination of local monitoring results. Then, a variable reasoning strategy is introduced to locate fault variables. Unlike the contribution plot, this method locates normal and fault variables effectively, and gives initiatory judgment for ambiguous variables. Finally, the effectiveness of the proposed process monitoring and fault variable identification schemes is verified through a numerical example and TE chemical process.     

改进CCM算法检测外部扰动下系统变量间的时滞和因果关系

在化工过程中,可以通过分析变量间的相互作用和时滞关系,推理故障的传播路径和网络,指出故障的根原因。这对提高过程安全性,增强经济效益具有重要意义,是研究的热点和难点。常见的互相关函数和传递熵等方法,由于只适用于线性或弱非线性系统,或计算量较大,往往无法准确地获得变量间的时滞信息和作用强度,在实际应用中存在不足。近年来,在生态领域研究中提出的交叉收敛映射（CCM）算法,被认为是一种适用于非线性耦合过程因果分析的方法,可适用于耦合变量间时滞关系的检测。但对于带有外部扰动的化工过程,CCM无法根据随时受到扰动的过程数据构造出稳定的嵌入流形,导致了时滞和因果分析失败。而基于CCM进行改进的扰动过滤交叉映射（DFCM）方法,通过分析外部扰动对系统的影响,预先筛选扰动数据,再将过滤后的数据代入交叉映射的计算中。算例表明,这种处理能有效地避免扰动下嵌入流形不稳定的问题,适用于处于扰动下的化工过程,并得到良好的时滞和因果关系分析效果。     

Accelerated Recursive Feature Elimination Based on Support Vector Machine for Key Variable Identification

Key variable identification for classifications is related to many trouble-shooting problems in process industries. Recursive feature elimination based on support vector machine （SVM-RFE） has been proposed recently in application for feature selection in cancer diagnosis. In this paper, SVM-RFE is used to the key variable selection in fault diagnosis, and an accelerated SVM-RFE procedure based on heuristic criterion is proposed. The data from Tennessee Eastman process （TEP） simulator is used to evaluate the effectiveness of the key variable selection using accelerated SVM-RFE （A-SVM-RFE）. A-SVM-RFE integrates computational rate and algorithm effectiveness into a consistent framework. It not only can correctly identify the key variables, but also has very good computational rate. In comparison with contribution charts combined with principal component aralysis （PCA） and other two SVM-RFE algorithms, A-SVM-RFE performs better. It is more fitting for industrial application.