ELM Soft‐Sensor Model for SMB Chromatographic Separation Process Based on Heavy‐Tailed Distribution Adaptive Giant Trevally Optimizer

In the SMB chromatographic separation process, the complex mechanism and the characteristic of being affected by multiple parameters, it is challenging for the process to run steadily at the ideal operating point for an extended period of time, especially given its sensitivity to operating parameters and perturbations. The application's main problem is the real‐time monitoring and measurement of component purity. An extreme learning machine (ELM) soft‐sensor model for the SMB chromatographic separation process optimized with the heavy‐tailed distribution adaptive giant trevally algorithm (GTO) is proposed for the prediction of key economic and technical indicators (purity and yield) in the separation process. The activation function was the sigmoid function. Aiming at the shortcomings of the ELM model that the weights and offsets are randomly assigned, there is low reliability and low forecast accuracy, the CSAGTO based on the heavy‐tailed distribution and adaptive positional parameter operator is used to optimize the parameters of the ELM neural network. The proposed CSAGTO with other algorithms is utilized to do simulation experiments for CEC2017 function optimization to assess the efficacy of the suggested algorithm. The above comparison algorithm is then used to carry out soft‐sensor model prediction experiments, and the simulation results demonstrate that the suggested model can match the real‐time control needs of the SMB chromatographic separation process and has greater generalization and prediction accuracy.

Authors

Institutions

Publication Details

Journal
Journal of Chemometrics
Published
2026-11-18
DOI
https://doi.org/10.1002/cem.70197
Primary Topic
Machine Learning and ELM
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

ELM Soft‐Sensor Model for SMB Chromatographic Separation Process Based on Heavy‐Tailed Distribution Adaptive Giant Trevally Optimizer

Jie‐Sheng Wang, Cheng Xing, Yi‐Peng Shang‐Guan, Yong‐Cheng Sun et al.
Journal of Chemometrics
Machine Learning and ELM
article

ELM Soft‐Sensor Model for SMB Chromatographic Separation Process Based on Heavy‐Tailed Distribution Adaptive Giant Trevally Optimizer

Jie‐Sheng Wang, Cheng Xing, Yi‐Peng Shang‐Guan, Yong‐Cheng Sun, Xiao‐Tian Wang
article en

Abstract

In the SMB chromatographic separation process, the complex mechanism and the characteristic of being affected by multiple parameters, it is challenging for the process to run steadily at the ideal operating point for an extended period of time, especially given its sensitivity to operating parameters and perturbations. The application's main problem is the real‐time monitoring and measurement of component purity. An extreme learning machine (ELM) soft‐sensor model for the SMB chromatographic separation process optimized with the heavy‐tailed distribution adaptive giant trevally algorithm (GTO) is proposed for the prediction of key economic and technical indicators (purity and yield) in the separation process. The activation function was the sigmoid function. Aiming at the shortcomings of the ELM model that the weights and offsets are randomly assigned, there is low reliability and low forecast accuracy, the CSAGTO based on the heavy‐tailed distribution and adaptive positional parameter operator is used to optimize the parameters of the ELM neural network. The proposed CSAGTO with other algorithms is utilized to do simulation experiments for CEC2017 function optimization to assess the efficacy of the suggested algorithm. The above comparison algorithm is then used to carry out soft‐sensor model prediction experiments, and the simulation results demonstrate that the suggested model can match the real‐time control needs of the SMB chromatographic separation process and has greater generalization and prediction accuracy.

Journal of ChemometricsVol. 40(12)
University of Science and Technology Liaoning (CN)
Machine Learning and ELM
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon