Predicting the Compressive Strength of Concrete Containing Binary Supplementary Cementitious Material Using Machine...

Area of Science:

• Materials Science
• Civil Engineering
• Computational Science

Background:

• Supplementary cementitious materials (SCMs) enhance concrete properties and sustainability.
• Combining different SCMs addresses complex concrete performance demands.
• Understanding SCM interactions is crucial for advanced concrete applications.

Purpose of the Study:

• Develop a Machine Learning (ML) model to predict concrete compressive strength.
• Utilize a robust and time-saving method for SCM-based concrete.
• Investigate binary SCM combinations at various ages.

Main Methods:

• Collected a database of concrete mix designs, SCM properties (fly ash, metakaolin, zeolite), and specimen age from literature.
• Prepared 21 experimental mix designs with binary SCMs to supplement existing data.
• Employed Machine Learning algorithms for predictive modeling of compressive strength.

Main Results:

• The ML model demonstrated high accuracy in predicting concrete compressive strength.
• The model precisely forecasts strength for arbitrary SCM combinations and ages.
• The study successfully filled literature gaps regarding binary SCM performance.

Conclusions:

• Machine learning offers an efficient tool for predicting concrete compressive strength with SCMs.
• The developed model can guide the selection of optimal SCM replacement levels.
• This research facilitates the precise utilization of binary cementitious systems in concrete.