Developing a Novel Method for Estimating the Speed of Sound in Biodiesel Known as Grey Wolf Optimizer Support Vector...

Area of Science:

• * Thermodynamics and Fluid Dynamics
• * Chemical Engineering and Materials Science

Background:

• * Accurate prediction of physical properties like the speed of sound is crucial for biodiesel process design and optimization.
• * Existing models may lack the robustness and accuracy required for diverse biodiesel compositions and operating conditions.

Purpose of the Study:

• * To develop a robust and accurate predictive model for the speed of sound in biodiesel.
• * To identify key physical properties influencing the speed of sound in biodiesel.
• * To establish a benchmark for machine learning applications in predicting thermophysical properties of biofuels.

Main Methods:

• * Compilation of an extensive dataset on the speed of sound in biodiesel from existing literature.
• * Optimization of a Support Vector Machine (SVM) model using the Grey Wolf Optimization (GWO) algorithm.
• * Correlation analysis between the speed of sound and properties such as pressure, temperature, molecular weight, and normal melting point.

Main Results:

• * Achieved highly satisfactory model performance with R-squared (R²) of 1 and Root Mean Square Error (RMSE) of 1.4024.
• * Sensitivity analysis revealed pressure as the dominant factor, with a relevancy factor of 87.92%.
• * The proposed GWO-SVM model demonstrated superior accuracy compared to other machine learning methods in the literature.

Conclusions:

• * The GWO-SVM model provides a highly accurate and reliable method for predicting the speed of sound in biodiesel.
• * Pressure is identified as the most critical parameter affecting the speed of sound in biodiesel.
• * This study highlights the potential of advanced machine learning techniques for modeling complex thermophysical properties in biofuels.