Estimation of Isentropic Compressibility of Biodiesel Using ELM Strategy: Application in Biofuel Production Processes

Area of Science:

• Physical Chemistry
• Chemical Engineering
• Materials Science

Background:

• Isentropic compressibility is a key property for biofuels.
• Experimental determination of isentropic compressibility is complex and time-consuming.

Purpose of the Study:

• To develop a novel Machine Learning (ML) model for predicting isentropic compressibility.
• To offer an efficient alternative to experimental methods in biofuel characterization.

Main Methods:

• Utilized the Extreme Learning Machine (ELM) algorithm, a type of ML.
• Trained and validated the ELM model using a database of 483 real-world biofuel datasets.
• Performed sensitivity analysis to identify key input parameters.

Main Results:

• The ELM model achieved high accuracy, with a mean relative error of 0.19 and R² values of 1.
• Demonstrated excellent performance in predicting isentropic compressibility for biodiesel.
• Sensitivity analysis revealed the normal melting point as the most influential input variable.

Conclusions:

• The proposed ML-based ELM method is a highly effective and efficient tool for calculating isentropic compressibility in biofuels.
• This approach significantly reduces the time and complexity associated with experimental measurements.
• Normal melting point is a critical factor for predicting this property.