Numerical Simulation and Bayesian Optimization CatBoost Prediction Method for Characteristic Parameters of Veneer Roller Pressing and Defibering DOI Open Access
Qi Wang, Chenglin Yan, Yahui Zhang

et al.

Forests, Journal Year: 2024, Volume and Issue: 15(12), P. 2173 - 2173

Published: Dec. 10, 2024

Defibering equipment is employed in the production of scrimber for purpose wood veneer rolling, cutting, and directional fiber separation. However, current defibering exhibits a notable degree automation deficiency, relying more on manual operation empirical methods process control, which impedes stability quality. This study presented an in-depth finite element analysis roller-pressing equipment, prediction method incorporating numerical simulation ensemble learning was proposed through data collection feature selection. The objective to integrate this into intelligent decision-making system with aim improving productivity effectively stabilizing product results ABAQUS 2020 revealed that roller gap velocity as well geometrical parameters veneer, have significant influence effect. Combining these factors, 702 experiments were devised executed, database constructed based model-building outcomes. strain stress observed served represent force deformation. CatBoost algorithm used establish models key effect, Bayesian Optimization 5-fold cross-validation techniques enabled achieve coefficients determination 0.98 0.97 training test datasets, respectively. Shapley Additive Explanation provide insight contribution each feature, thereby guiding selection simplifying model. show scheme can determine core then practical control strategy online control.

Language: Английский

Numerical Simulation and Bayesian Optimization CatBoost Prediction Method for Characteristic Parameters of Veneer Roller Pressing and Defibering DOI Open Access
Qi Wang, Chenglin Yan, Yahui Zhang

et al.

Forests, Journal Year: 2024, Volume and Issue: 15(12), P. 2173 - 2173

Published: Dec. 10, 2024

Defibering equipment is employed in the production of scrimber for purpose wood veneer rolling, cutting, and directional fiber separation. However, current defibering exhibits a notable degree automation deficiency, relying more on manual operation empirical methods process control, which impedes stability quality. This study presented an in-depth finite element analysis roller-pressing equipment, prediction method incorporating numerical simulation ensemble learning was proposed through data collection feature selection. The objective to integrate this into intelligent decision-making system with aim improving productivity effectively stabilizing product results ABAQUS 2020 revealed that roller gap velocity as well geometrical parameters veneer, have significant influence effect. Combining these factors, 702 experiments were devised executed, database constructed based model-building outcomes. strain stress observed served represent force deformation. CatBoost algorithm used establish models key effect, Bayesian Optimization 5-fold cross-validation techniques enabled achieve coefficients determination 0.98 0.97 training test datasets, respectively. Shapley Additive Explanation provide insight contribution each feature, thereby guiding selection simplifying model. show scheme can determine core then practical control strategy online control.

Language: Английский

Citations

0