A Causal Model-Inspired Automatic Feature-Selection Method for Developing Data-Driven Soft Sensors in Complex Industrial Processes

Sun, Yan-Ning; Qin, Wei; Hu, Jinhua; Xu, Hongwei; Sun, Poly Z. H.

doi:10.1016/j.eng.2022.06.019

Cited by 26 publications

(8 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The work ow of the CART algorithm based on feature selection is shown in Fig. 3 (Sun et al, 2023). The initial values for classi cation and regression are decision trees.…”

Section: Cart Algorithmmentioning

confidence: 99%

Transformative Analysis of Mental Health in Rural China: The CART Algorithm Approach

Fu,

He,

2024

Preprint

View full text Add to dashboard Cite

Mental health remains a critical concern in China, particularly for patients with severe psychiatric disorders in rural areas. This study aimed to analyze the health status of Chinese patients with severe psychiatric disorders using the CART algorithm. Mental health is a critical facet of overall well-being, yet understanding and addressing the complexities of severe psychiatric disorders in rural China presents unique challenges. This comprehensive study employs advanced analytical techniques to explore and illuminate multifaceted aspects of mental health, with a specific focus on patients with severe psychiatric disorders, healthcare professionals, and rural residents in China. This study's novelty is in its creative utilization of the CART algorithm to assess the well-being of Chinese individuals grappling with severe psychiatric conditions. This cutting-edge data analysis method opens up a potential path for enhancing mental health strategies and optimizing resource allocation. In essence, this study offers a holistic examination of mental health in rural China, encompassing various dimensions, from predictive elements to the challenges faced by healthcare professionals. Its findings aim to inform the development of effective mental health strategies and resource allocation, enhancing the overall well-being of individuals grappling with severe psychiatric disorders in this region. The government and relevant authorities are recommended to ensure their physical and mental health. The lack of mental health information in rural China also negatively impacts patients' behavior in seeking and using medical services. Thus, measures to promote different forms of mental health education are proposed. In conclusion, the treatment of patients with severe mental illness is crucial to the physical and mental health of millions of people.

show abstract

“…The work ow of the CART algorithm based on feature selection is shown in Fig. 3 (Sun et al, 2023). The initial values for classi cation and regression are decision trees.…”

Section: Cart Algorithmmentioning

confidence: 99%

Transformative Analysis of Mental Health in Rural China: The CART Algorithm Approach

Fu,

He,

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…According to the above Equations (16)(17)(18)(19)(20), one can obtain an input matrix with 2001 samples and four variables, and an output matrix with 2001 samples and one variable for the numerical case. To simulate the dynamic features of the process data, the time delay d is set to 2, and the dimensions of the argument variable are 12.…”

Section: Data Pre-processingmentioning

confidence: 99%

“…The amount of data collected from industrial processes is often large, slowly time-varying, and nonlinear, among other characteristics. [19] To find the valuable insights hidden in the data, some latent variable methods have been proposed and widely applied, such as canonical correlation analysis (CCA), principal component analysis (PCA), partial least squares (PLS), and various methods expanded on the basis of these methods. [20][21][22] The PCA method is employed to extract features that contribute more to the independent and dependent variables.…”

Section: Introductionmentioning

confidence: 99%

Dynamic nonlinear soft sensor modelling method using linear slow feature analysis and least squares support vector regression for batch processes

Zhang,

Liu,

Zhang

et al. 2023

Can J Chem Eng

View full text Add to dashboard Cite

Data‐driven soft sensor models have been extensively utilized in industrial processes. Batch processes are usually employed to manufacture low‐volume and high value‐added products in chemical, materials, and pharmaceutical industries. The most distinctive features of batch process lie in nonlinear, repetition, and slow time varying characteristics. In this paper, a data‐driven soft sensor modelling method based on linear slow feature analysis (LSFA) and least squares support vector regression (LSSVR) is proposed. In this method, LSFA was used to effectively capture the driving force behind the data features that change as slowly as possible. Then, a LSSVR model was constructed between the extracted slow feature variables and quality variables. Finally, a numerical example, industrial penicillin fermentation processes, and cobalt oxalate synthesis process were utilized to confirm the prediction accuracy and model reliability of the proposed approach.

show abstract

“…This method has distinct physical significance, intuitively reflects changes in internal variables of the chemical processes, and has strong interpretability. However, establishing a model using this method requires an ideal environment and meeting a large number of assumptions [6]. At present, the chemical processes are too complex and parameters are difficult to obtain, so the model established under this method has poor practicality and low accuracy.…”

Section: Introductionmentioning

confidence: 99%

Research on soft sensor modeling method for complex chemical processes based on local semi-supervised selective ensemble learning

Liu,

Li,

Zhang

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

To address the adverse effects of nonlinearity and dynamic time-varying in complex chemical processes on the accuracy of the soft sensor model, a local-semi-supervised ensemble learning for soft sensor modeling (LS-SEL-LSTM) method is proposed in this article. Firstly, a hierarchical clustering method incorporating spatiotemporal criteria is proposed to reduce the influence of nonlinearity in global model prediction accuracy. The method considers the dynamic time-varying characteristics of temporal data and generates multiple local datasets. Then, to address the issue of multi-rate between auxiliary variables and dominant variables, a semi-supervised weight fusion mechanism based on temporal correlation is proposed, which effectively utilizes auxiliary variables to reconstruct local semi-supervised datasets and establishes local soft sensing models using LSTM. Concurrently, the parameters of the established model were optimized using the Flower Pollination Algorithm (FPA). Subsequently, a selective ensemble learning method based on sub-model prediction accuracy and an adaptive combination weight calculation method for sub-models were proposed to improve the prediction accuracy. Finally, the effectiveness of the proposed method was verified through the actual dataset of the sulfur recovery process. The results indicate that LS-SEL-LSTM performs well in handling complex chemical processes with nonlinear and dynamic time-varying characteristics.

show abstract

A Causal Model-Inspired Automatic Feature-Selection Method for Developing Data-Driven Soft Sensors in Complex Industrial Processes

Cited by 26 publications

References 46 publications

Transformative Analysis of Mental Health in Rural China: The CART Algorithm Approach

Transformative Analysis of Mental Health in Rural China: The CART Algorithm Approach

Dynamic nonlinear soft sensor modelling method using linear slow feature analysis and least squares support vector regression for batch processes

Research on soft sensor modeling method for complex chemical processes based on local semi-supervised selective ensemble learning

Contact Info

Product

Resources

About