Optimizing base fluid composition for PEMFC cooling: A machine learning approach to balance thermal and rheological...

Area of Science:

• Materials Science and Engineering
• Nanotechnology
• Sustainable Energy Systems

Background:

• Proton Exchange Membrane Fuel Cells (PEMFCs) are crucial for sustainable energy, requiring effective thermal management.
• Nanofluids offer superior cooling compared to conventional fluids, with reduced graphene oxide (rGO) hybrid nanofluids showing promise.
• Limited research exists on rGO-based hybrid nanofluids for PEMFC applications, highlighting the need for this study.

Purpose of the Study:

• To experimentally investigate the thermal and rheological properties of Al₂O₃ and rGO hybrid nanofluids.
• To explore the influence of base fluid composition (ethylene glycol and water) and nanoparticle concentration on fluid properties.
• To develop and validate predictive models for thermal conductivity and viscosity using machine learning techniques.

Main Methods:

• Preparation of hybrid nanofluids with varying concentrations of Al₂O₃ and rGO in different ethylene glycol (EG) and water (W) mixtures.
• Experimental evaluation of dispersion stability, viscosity, and thermal conductivity at various temperatures and concentrations.
• Application of machine learning models (Linear Regression, Decision Tree, eXtreme Gradient Boosting) for predicting thermal and rheological properties.

Main Results:

• Increased EG proportion decreased thermal conductivity but increased viscosity.
• Maximum thermal conductivity enhancement (1.23 ratio) at 80:20 W:EG (1 vol%, 60°C); maximum viscosity enhancement (1.48 ratio) at 20:80 W:EG (30°C).
• XGBoost models demonstrated superior predictive accuracy for both thermal conductivity (Test R² = 0.9941) and viscosity (Test R² = 0.9944).

Conclusions:

• rGO-based hybrid nanofluids exhibit significant potential for enhancing PEMFC thermal management.
• Machine learning models, particularly XGBoost, accurately predict nanofluid properties, aiding in efficient cooling system design.
• Understanding the impact of base fluid ratio, temperature, and concentration is crucial for optimizing nanofluid performance in PEMFCs.