Statistical process monitoring based on improved principal component analysis and its application to chemical processes

Tong, Chudong; Yan, Xuefeng; Ma, Yuxin

doi:10.1631/jzus.a1300003

Cited by 13 publications

(10 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Once the data dimension is reduced using PCA, T 2 and Q (SPE) statistics are calculated to detect a fault …”

Section: Process Monitoring Methodsmentioning

confidence: 99%

“…Historically, statistical process control charts and statistical process monitoring (SPM) methods have been readily applied to large-scale and complex processes where model-based approaches are often challenged. − The workhorse of SPM is the univariate control chart, yet it has limited use in a process environment as it overlooks the multivariate nature of processes where there may exist strong correlations among variables . Hence, it becomes hard to visualize and interpret the process behavior unless a dimension reduction step is executed first using methods such as principal components analysis (PCA) prior to the implementation of FDD methods. ,− …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Least Squares Sparse Principal Component Analysis and Parallel Coordinates for Real-Time Process Monitoring

Gajjar

Külahçı

Palazoğlu

2020

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

The unprecedented growth of machine-readable data throughout modern industrial systems has major repercussions for process monitoring activities. In contrast to model-based process monitoring that requires the physical and mathematical knowledge of the system in advance, the data-driven schemes provide an efficient alternative to extract and analyze process information directly from recorded process data. This paper introduces the least squares sparse principal component analysis to obtain readily interpretable sparse principal components. This is done in the context of parallel coordinates, which facilitate the visualization of high dimensional data. The key contribution is the establishment of control limits on independent sparse principal component and residual spaces to facilitate fault detection, complemented by the use of the Random Forests algorithm to carry out the fault diagnosis step. The proposed method is applied to the Tennessee Eastman process to highlight its merits.

show abstract

“…Once the data dimension is reduced using PCA, T 2 and Q (SPE) statistics are calculated to detect a fault …”

Section: Process Monitoring Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Least Squares Sparse Principal Component Analysis and Parallel Coordinates for Real-Time Process Monitoring

Gajjar

Külahçı

Palazoğlu

2020

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…Again, the contribution of T 2 to each variable is shown as a fraction in the plot. [33,[38][39][40][41] 3 | LASSO SPCA WITH FDR AND FAR…”

Section: Fault Detection and Diagnosis Using Pca And Spcamentioning

confidence: 99%

A new criterion for selection of non‐zero loadings for sparse principal component analysis (SPCA)

2021

View full text Add to dashboard Cite

Sparse principal component analysis (SPCA) has recently emerged as an approach aimed at producing compact principal component loadings by suppressing spurious values and thus overcoming some limitations of the traditional principal component analysis (PCA). This paper proposes a fault detection and diagnosis (FDD) method based on SPCA; in this approach, the number of non-zero loadings (NNZL) of SPCAs is selected based on both the false alarm rate (FAR) and the fault detection rate (FDR). The criterion is to have lower FAR and higher FDR. This new feature makes SPCA better suited for FDD, which is demonstrated by comparing its performance with that of three other methods for finding loadings. The overall FDD performances of both PCA and SPCA-based techniques are illustrated using the benchmark continuous stirred tank heater (CSTH) process. The results show that the PCs derived based on the proposed criterion has a better fault diagnosis ability.

show abstract

“…These techniques provide better performance than the single-variable approach. Among them, PCA [8][9][10][11][12][13][14] and partial least squares 15,16 are the most common techniques.…”

Section: Industrial Process Monitoring Techniquementioning

confidence: 99%

Principal component analysis for process monitoring in distributed system environment

Gao

Tian

Chen

2019

Concurrency and Computation

View full text Add to dashboard Cite

In modern industrial production processes, process monitoring plays a major role in improving process efficiency and quality of product, and it is an important task in industrial process. The best way to implement process monitoring is to develop models that describe different phenomena in physics, chemistry, and processes. However, because the modeling of industrial processes is often very complex and has significant intrinsic nonlinearities, a reasonable theoretical modeling method are often impractical. With the adoption of distributed control systems, the application of multivariate statistical monitoring is becoming more and more popular in the distributed system environment. This paper uses the technology for multivariate monitoring of continuous production processes. Based on principal component analysis (PCA) and making full use of system statistics information, the method can well deal with the nonlinear and multimode distribution of industrial data, which is difficult to be dealt with by traditional methods. This method can be used not only in process detection but also in fault detection. Finally, the method is applied to the actual example and the Tennessee-Eastman model; the simulation results prove the feasibility of the proposed method and achieve better results.

show abstract

Statistical process monitoring based on improved principal component analysis and its application to chemical processes

Cited by 13 publications

References 33 publications

Least Squares Sparse Principal Component Analysis and Parallel Coordinates for Real-Time Process Monitoring

Least Squares Sparse Principal Component Analysis and Parallel Coordinates for Real-Time Process Monitoring

A new criterion for selection of non‐zero loadings for sparse principal component analysis (SPCA)

Principal component analysis for process monitoring in distributed system environment

Contact Info

Product

Resources

About