Machine learning-based estimation and optimization of phoenix Dactylifera Seed Powder reinforced vinyl ester...

Area of Science:

• Materials Science
• Computational Materials Science
• Polymer Science

Background:

• Sustainable composites are crucial for reducing environmental impact.
• Phoenix Dactylifera Seed Powder (PDSP) offers a novel reinforcement for vinyl ester (VE) composites.
• Predicting composite properties is essential for material design and application.

Purpose of the Study:

• To develop a machine learning (ML) framework for predicting mechanical and thermal properties of PDSP/VE composites.
• To evaluate the performance of various ML algorithms in this prediction task.
• To identify key parameters influencing composite properties and enable data-driven design.

Main Methods:

• Experimental generation of mechanical and thermal data for PDSP/VE composites.
• Literature data compilation for comprehensive model training.
• Evaluation of Linear Regression, Support Vector Machine (SVM), Random Forest, and Decision Tree algorithms.
• Validation of predictive accuracy using R-squared values.

Main Results:

• The Support Vector Machine (SVM) model exhibited superior predictive accuracy.
• High R-squared values were achieved for tensile strength (0.91), flexural strength (0.83), hardness (0.86), and heat deflection temperature (0.85).
• Filler weight percentage was identified as the most influential parameter.

Conclusions:

• ML, particularly SVM, can reliably predict the properties of PDSP/VE composites.
• This approach reduces the need for extensive experimental testing.
• Optimized design of sustainable composites for automotive and civil engineering applications is facilitated.