Machine Learning Driven Fluidity and Rheological Properties Prediction of Fresh Cement-Based Materials DOI Open Access
Yi Liu, Zeyad M. A. Mohammed, Jialu Ma

et al.

Materials, Journal Year: 2024, Volume and Issue: 17(22), P. 5400 - 5400

Published: Nov. 5, 2024

Controlling workability during the design stage of cement-based material mix ratios is a highly time-consuming and labor-intensive task. Applying artificial intelligence (AI) methods to predict optimize materials can significantly enhance efficiency design. In this study, experimental testing was conducted create dataset 233 samples, including fluidity, dynamic yield stress, plastic viscosity materials. The proportions cement, fly ash (FA), silica fume (SF), water, superplasticizer (SP), hydroxypropyl methylcellulose (HPMC), sand were selected as inputs. Machine learning (ML) employed establish predictive models for these three early indicators. To improve prediction capability, optimized hybrid models, such Particle Swarm Optimization (PSO)-based CatBoost XGBoost, adopted. Furthermore, influence individual input variables on each indicator examined using Shapley Additive Explanations (SHAP) Partial Dependence Plot (PDP) analyses. This study provides novel reference achieving rapid accurate control workability.

Language: Английский

Machine Learning Driven Fluidity and Rheological Properties Prediction of Fresh Cement-Based Materials DOI Open Access
Yi Liu, Zeyad M. A. Mohammed, Jialu Ma

et al.

Materials, Journal Year: 2024, Volume and Issue: 17(22), P. 5400 - 5400

Published: Nov. 5, 2024

Controlling workability during the design stage of cement-based material mix ratios is a highly time-consuming and labor-intensive task. Applying artificial intelligence (AI) methods to predict optimize materials can significantly enhance efficiency design. In this study, experimental testing was conducted create dataset 233 samples, including fluidity, dynamic yield stress, plastic viscosity materials. The proportions cement, fly ash (FA), silica fume (SF), water, superplasticizer (SP), hydroxypropyl methylcellulose (HPMC), sand were selected as inputs. Machine learning (ML) employed establish predictive models for these three early indicators. To improve prediction capability, optimized hybrid models, such Particle Swarm Optimization (PSO)-based CatBoost XGBoost, adopted. Furthermore, influence individual input variables on each indicator examined using Shapley Additive Explanations (SHAP) Partial Dependence Plot (PDP) analyses. This study provides novel reference achieving rapid accurate control workability.

Language: Английский

Citations

0