Experimental and Modelling of Alkali-Activated Mortar Compressive Strength Using Hybrid Support Vector Regression and...

Area of Science:

• Materials Science
• Civil Engineering
• Computational Intelligence

Background:

• Alkali-activated limestone powder and natural pozzolan mortar (AALNM) offer a sustainable alternative to traditional concrete.
• Accurate prediction of compressive strength (CS) is crucial for structural integrity and material development.
• Existing modeling techniques may not fully capture the complex relationships influencing AALNM's CS.

Purpose of the Study:

• To develop novel hybrid models for predicting the compressive strength (CS) of alkali-activated limestone powder and natural pozzolan mortar (AALNM).
• To evaluate the accuracy and performance of these models compared to existing methods.
• To promote the use of environmentally friendly concrete with enhanced strength properties.

Main Methods:

• Development of hybrid models integrating genetic algorithm (GA) and support vector regression (SVR).
• Training and testing of models (GA-SVR-CS1, GA-SVR-CS3, GA-SVR-CS14, GA-SVR-CS28E) to predict CS at various ages (1, 3, 14, and 28 days).
• Performance evaluation using correlation coefficient and root mean square error (RMSE) on unseen data.

Main Results:

• Hybrid GA-SVR models achieved high accuracy in predicting AALNM compressive strength: up to 96.64% (1-day), 90.84% (3-day), and 93.40% (14-day) based on correlation coefficient.
• The GA-SVR-CS28E model, predicting 28-day CS from 14-day strength, significantly outperformed other models predicting 28-day CS from earlier strengths (1, 3, or 7 days) or all descriptors, showing substantial RMSE improvements.
• The developed models demonstrate superior predictive capabilities for AALNM compressive strength.

Conclusions:

• Hybrid GA-SVR models provide a highly accurate and efficient method for predicting the compressive strength of alkali-activated limestone powder and natural pozzolan mortar.
• These models can facilitate the development and application of sustainable, high-strength concrete.
• The study offers effective computational tools for material scientists and engineers in the field of concrete technology.