☆ 4.7 Article

Fault monitoring for chemical processes using neighborhood embedding discriminative analysis

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2022)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Engineering, Environmental

Fault monitoring using novel adaptive kernel principal component analysis integrating grey relational analysis

Yongming Han et al.

Summary: The proposed method of adaptive kernel principal component analysis (AKPCA) combined with grey relational analysis (GRA) is used to dynamically monitor fault occurrences in chemical processes. By adaptively extracting kernel principal components using a moving window and threshold method, the variables causing faults can be effectively analyzed.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2022)

添加到收藏夹

Article Engineering, Environmental

An enhanced dynamic artificial immune system based on simulated vaccine for early fault diagnosis with limited data

Yuman Yao et al.

Summary: An enhanced dynamic artificial immune system based on simulated vaccine and correlation coefficient methods (SV-CCDAIS) is proposed to improve the process safety of chemical systems with extreme absence of data. By using simulated vaccine, dynamic time warping, and different dynamic correlation measurement methods, the proposed method achieves higher fault diagnosis accuracy and shorter diagnosis time compared to the comparative method.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2022)

添加到收藏夹

Article Engineering, Environmental

Fault detection of petrochemical process based on space-time compressed matrix and Naive Bayes

Zhenyu Deng et al.

Summary: This paper proposes a fault detection approach based on space-time compressed matrix (STCM) and Naive Bayes (NB) to achieve fast learning and prediction. By extracting slowly varying features and compressing the data matrix in space-time, the proposed approach reduces the learning complexity while ensuring classification performance.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2022)

添加到收藏夹

Article Engineering, Environmental

Statistical method based on dissimilarity of variable correlations for multimode chemical process monitoring with transitions

Cheng Ji et al.

Summary: This article proposes a statistical process monitoring method based on the dissimilarity of process variable correlation (DISS-PVC), which is able to monitor multiple stable modes and transitions between modes simultaneously without prior knowledge of the number of operating modes. The method quantifies variable correlation using mutual information and performs fault detection using cosine similarity as a dissimilarity index.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2022)

添加到收藏夹

Article Automation & Control Systems

Deep Learning of Latent Variable Models for Industrial Process Monitoring

Xiangyin Kong et al.

Summary: This article proposes a novel monitoring framework for latent variable models using hierarchical feature extraction, Bayesian inference, and weighting strategy. The framework includes a deep PCA-ICA model for hierarchical feature extraction, Bayesian inference for transforming the features to posterior probabilities, and a weighting strategy for combining the probabilities into new probabilistic statistics. The proposed model is validated using the Tennessee Eastman process and the effectiveness of the deep hierarchical feature extraction structure is further analyzed.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2022)

添加到收藏夹

Article Engineering, Environmental

A deep learning model for process fault prognosis

Rajeevan Arunthavanathan et al.

Summary: Early fault detection and fault prognosis are crucial for safe process operations, requiring early examination of fault symptoms. New fault prognosis methods help in predicting remaining useful life of systems and enhancing process safety.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

添加到收藏夹

Article Engineering, Environmental

A novel orthogonal self-attentive variational autoencoder method for interpretable chemical process fault detection and identification

Xiaotian Bi et al.

Summary: The proposed orthogonal self-attentive variational autoencoder (OSAVA) model in this paper is capable of simultaneously performing fault detection and identification tasks, and providing interpretable results. Evaluation on the Tennessee Eastman process (TEP) demonstrates promising fault detection rate and low detection delay for the OSAVA model.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

添加到收藏夹

Article Engineering, Environmental

Locality preserving randomized canonical correlation analysis for real-time nonlinear process monitoring

Ping Wu et al.

Summary: The paper presents a novel locality preserving randomized canonical correlation analysis (LPRCCA) method for real-time nonlinear process monitoring which maps original data to a randomized low-dimensional feature space and integrates local geometric structure information to improve data mining performance, reducing computational cost and showing significant advantages over kernel-based methods.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

添加到收藏夹

Article Automation & Control Systems

Variational Bayesian probabilistic modeling framework for data-driven distributed process monitoring

Jiashi Jiang et al.

Summary: This study proposes a probabilistic modeling approach based on variational Bayesian for distributed process monitoring, characterizing the variable relationships within and among local units through latent variable models, and verifies the effectiveness of the method through three case studies.

CONTROL ENGINEERING PRACTICE (2021)

添加到收藏夹

Article Automation & Control Systems

Hierarchical hybrid distributed PCA for plant-wide monitoring of chemical processes

Yue Cao et al.

Summary: A hierarchical hybrid distributed PCA (HDPCA) modeling framework is proposed to address the difficulty in carrying out plant-wide monitoring of a process due to intricately correlated process variables. In HDPCA, process variables are divided twice into subblocks in a hierarchical two-layer manner, and a two-layer Bayesian inference fusion strategy is used to obtain distributed monitoring results. The feasibility of HDPCA is demonstrated to outperform other compared methods on benchmark and industrial processes.

CONTROL ENGINEERING PRACTICE (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Quality Variable Prediction for Nonlinear Dynamic Industrial Processes Based on Temporal Convolutional Networks

Xiaofeng Yuan et al.

Summary: Soft sensors have been developed to estimate difficult-to-measure quality variables for real-time process monitoring and control. Two temporal convolutional network (TCN)-based models were proposed to address process nonlinearities and dynamics, with AR-TCN showing improved correlation capture between quality and process variables for enhanced prediction accuracy.

IEEE SENSORS JOURNAL (2021)

添加到收藏夹

Article Automation & Control Systems

Monitoring multimode processes: A modified PCA algorithm with continual learning ability

Jingxin Zhang et al.

Summary: In this paper, a modified PCA algorithm called PCA-EWC is proposed for monitoring multimode processes, which adopts elastic weight consolidation (EWC) to prevent catastrophic forgetting of the PCA model for successive modes. The optimal parameters are obtained through the difference of convex functions, and potential limitations and relevant solutions are discussed to further understand the algorithm.

JOURNAL OF PROCESS CONTROL (2021)

添加到收藏夹

Article Automation & Control Systems

Multivariate statistical process monitoring based on principal discriminative component analysis

Shanzhi Li et al.

Summary: The novel PDCA algorithm extracts discriminative features to uncover deviation between online and normal data, ensuring high efficiency in Multi-variate Statistical Process Monitoring.

JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS (2021)

添加到收藏夹

Article Engineering, Environmental

A data-driven Bayesian network learning method for process fault diagnosis

Md Tanjin Amin et al.

Summary: This study presents a data-driven methodology that integrates PCA with BN for fault detection and diagnosis, with the use of CD and KLD for automatic selection of principal components and development of data-driven BN learning technique. The method utilizes a combination of vine copula and Bayes' theorem to capture nonlinear dependence in high-dimensional process data, eliminating the need for discretization of continuous data.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

添加到收藏夹

Article Engineering, Environmental

Risk-based fault detection and diagnosis for nonlinear and non-Gaussian process systems using R-vine copula

Md Tanjin Amin et al.

Summary: This paper proposes a risk-based fault detection and diagnosis methodology using R-vine copula and event tree for nonlinear and nonGaussian process systems. The methodology shows better performance in detecting and diagnosing abnormal situations compared to conventional techniques.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

添加到收藏夹

Article Engineering, Environmental

Decentralized PCA modeling based on relevance and redundancy variable selection and its application to large-scale dynamic process monitoring

Bing Xiao et al.

Summary: The new RRVS-DPCA method addresses abnormal condition detection in large-scale industrial processes through variable selection based on relevance and redundancy, as well as a weighted contribution plot method to identify root causes of faults. The feasibility and effectiveness of the proposed monitoring scheme were demonstrated through comparisons with state-of-the-art process monitoring methods on numerical examples and the Tennessee Eastman benchmark process.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Deep Learning for Data Modeling of Multirate Quality Variables in Industrial Processes

Xiaofeng Yuan et al.

Summary: A novel deep learning strategy based on multirate stacked autoencoder (MR-SAE) is proposed for predicting both the 50% boiling point and cetane content of diesel oil. The MR-SAE-based model outperforms SAE and deep belief networks in terms of performance, while also having fewer parameters and shorter training time.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2021)

添加到收藏夹

Article Automation & Control Systems

Robust Monitoring and Fault Isolation of Nonlinear Industrial Processes Using Denoising Autoencoder and Elastic Net

Wanke Yu et al.

IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY (2020)

添加到收藏夹

Review Automation & Control Systems

Bridging systems theory and data science: A unifying review of dynamic latent variable analytics and process monitoring

S. Joe Qin et al.

ANNUAL REVIEWS IN CONTROL (2020)

添加到收藏夹

Article Automation & Control Systems

Slow feature analysis-independent component analysis based integrated monitoring approach for industrial processes incorporating dynamic and static characteristics

Jian Huang et al.

CONTROL ENGINEERING PRACTICE (2020)

添加到收藏夹

Article Automation & Control Systems

Comparative study on monitoring schemes for non-Gaussian distributed processes

Gang Li et al.

JOURNAL OF PROCESS CONTROL (2018)

添加到收藏夹

Article Automation & Control Systems

A novel dynamic PCA algorithm for dynamic data modeling and process monitoring

Yining Dong et al.

JOURNAL OF PROCESS CONTROL (2018)

添加到收藏夹

Article Automation & Control Systems

Process monitoring via enhanced neighborhood preserving embedding

Bing Song et al.

CONTROL ENGINEERING PRACTICE (2016)

添加到收藏夹

Article Automation & Control Systems

Nonlocal structure constrained neighborhood preserving embedding model and its application for fault detection

Aimin Miao et al.

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS (2015)

添加到收藏夹

Article Automation & Control Systems

A New Method of Dynamic Latent-Variable Modeling for Process Monitoring

Gang Li et al.

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS (2014)

添加到收藏夹

Article Engineering, Chemical

Time Neighborhood Preserving Embedding Model and Its Application for Fault Detection

Aimin Miao et al.

INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH (2013)

添加到收藏夹

Article Engineering, Chemical

Fault detection and diagnosis based on modified independent component analysis

Jong-Min Lee et al.

AICHE JOURNAL (2006)

添加到收藏夹

Article Automation & Control Systems

Nonlinear process monitoring using JITL-PCA

C Cheng et al.

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS (2005)

添加到收藏夹

Article Engineering, Chemical

Statistical monitoring of dynamic processes based on dynamic independent component analysis

JM Lee et al.

CHEMICAL ENGINEERING SCIENCE (2004)

添加到收藏夹

Article Engineering, Chemical

Reconstruction-based fault identification using a combined index

HH Yue et al.

INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH (2001)

添加到收藏夹

Article Automation & Control Systems

The application of principal component analysis and kernel density estimation to enhance process monitoring

Q Chen et al.

CONTROL ENGINEERING PRACTICE (2000)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.