Advancing Construction 3D Printing with Predictive Interlayer Bonding Strength: A Stacking Model Paradigm.

Area of Science:

• Materials Science
• Civil Engineering
• Computer Science

Background:

• 3D printing concrete requires enhanced quality stability for practical applications.
• Accurate prediction of interlayer bonding strength (IBS) is crucial for 3D printed concrete structures.

Purpose of the Study:

• To develop a novel machine learning (ML) model for predicting the interlayer bonding strength (IBS) of 3D printing concrete.
• To improve the quality stability and production scalability of 3D printed concrete.

Main Methods:

• Implementation of a stacking machine learning strategy.
• Utilizing Support Vector Regression (SVR), K-Nearest Neighbors (KNN), and Gaussian Process Regression (GPR) as base models.
• Employing 10-fold cross-validation and statistical performance evaluations.

Main Results:

• The developed stacking model significantly outperformed individual base models in IBS prediction.
• The coefficient of determination (R²) improved from 0.91 to 0.96 with the stacking model.
• Demonstrated superior prediction accuracy for interlayer bonding strength.

Conclusions:

• The novel stacking ML model effectively enhances the prediction accuracy of interlayer bonding strength in 3D printing concrete.
• This advancement facilitates improved quality stability and supports the scaled-up production of 3D printed concrete structures.