Dynamic latent variable regression for inferential sensor modeling and monitoring

Zhu, Qinqin; Qin, S. Joe; Dong, Yining

doi:10.1016/j.compchemeng.2020.106809

Cited by 41 publications

(32 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reader is referred to ref. 51. applications, [56][57][58] a topic we will cover in more detail later in this manuscript.…”

Section: Reaction Chemistry and Engineering Reviewmentioning

confidence: 99%

“…This way, the variability and contribution in noisy inputs will only appear if a higher number of nodes is used (similar to having a higher number of principal components). Reducing the number of redundant sensors to look at while capturing the system dynamics is a necessary step for realistic industrial data applications, 56–58 a topic we will cover in more detail later in this manuscript.…”

Section: Industrial Applications In Manufacturingmentioning

confidence: 99%

See 1 more Smart Citation

Industrial data science – a review of machine learning applications for chemical and process industries

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The reader is referred to ref. 51. applications, [56][57][58] a topic we will cover in more detail later in this manuscript.…”

Section: Reaction Chemistry and Engineering Reviewmentioning

confidence: 99%

Section: Industrial Applications In Manufacturingmentioning

confidence: 99%

Industrial data science – a review of machine learning applications for chemical and process industries

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Although the LVR model has a consistent inner model and outer model and has a stronger prediction power compared to the PLS method, it may suffer ill-conditioning caused by the collinearity in process data. A regularization term is incorporated into the LVR algorithm to overcome the potential ill-conditioning, which performs similar to the LVR model where pseudo-inverse is utilized as a substitution, 14 where objective function 1 is modified as 18…”

Section: Preliminariesmentioning

confidence: 99%

Kernel-Regularized Latent-Variable Regression Models for Dynamic Processes

Sun

Çınar

Xia

et al. 2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Establishing relations between variables and realtime prediction of quality variables or other key indicators are critical for dynamic processes including industrial and biological processes. In this study, a novel multivariate statistical modeling method named "kernel-regularized latent variable regression (KrLVR) approach" is proposed for capturing the dynamics of a process by building KrLVR models with process and quality data. First, a regularization term based on a kernel matrix is incorporated into the objective of the latent variable regression model. Consequently, the proposed KrLVR method has the ability to overcome potential ill-conditioning resulting from collinearity in process data and has a stronger prediction power. Besides, the inner model is consistent with the outer model, which enables the proposed method to predict quality data with fewer latent variables. Second, the prior knowledge of dynamic processes such as exponential stability and smoothness can be integrated into the modeling process by using an appropriate kernel matrix. In addition, to meet the requirement of exponential stability of the model, the weights of the model should decay exponentially, that is, coincident with determining the number of historical observations for data augmentation (identification of the model structure). Therefore, the problem of tuning model complexity is eluded, and it becomes finding appropriate hyper-parameters of the kernel matrix. Moreover, the empirical Bayesian method is utilized for estimating hyper-parameters of the kernel matrix from augmented process and quality data. Three case studies illustrate the performance of the proposed KrLVR method by comparing with several other relevant methods.

show abstract

“…Modern industrial processes are often characterized by high dimensions, strong multicollinearity, nonlinearity, and high noise. , The question of how to successfully adapt the data-driven approach to the industrial process with the above characteristics has become the focus of the industry. Machine learning and deep learning have gained significant interest and have been the dominant approaches in industrial processes. − Due to their excellent predictive accuracy, they are successfully applied to soft sensor and process monitoring. However, most machine learning models are black-box models, and as such it is difficult to interpret their behavior in relation to the process variables.…”

Section: Introductionmentioning

confidence: 99%

Causal Discovery Based on Observational Data and Process Knowledge in Industrial Processes

Cao

et al. 2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Causal discovery approaches are gaining popularity in industrial processes. Existing causal discovery algorithms can indeed find some important causal relationships from industrial data, but, at the same time, the algorithms may also give some incorrect causal relationships. In order to deal with this problem, we give four kinds of process knowledge definitions according to the special characteristics of complex industrial processes. Causal discovery algorithms will yield more accurate results and deeper insights if the process knowledge is properly addressed. Based on commercial-scale fluid catalytic cracker unit data, we validate the effectiveness of the proposed methods with some state-of-the-art causal discovery algorithms.

show abstract

Dynamic latent variable regression for inferential sensor modeling and monitoring

Cited by 41 publications

References 28 publications

Industrial data science – a review of machine learning applications for chemical and process industries

Industrial data science – a review of machine learning applications for chemical and process industries

Kernel-Regularized Latent-Variable Regression Models for Dynamic Processes

Causal Discovery Based on Observational Data and Process Knowledge in Industrial Processes

Contact Info

Product

Resources

About