Optimizing ternary hybrid nanofluids using neural networks, gene expression programming, and multi-objective particle...

Area of Science:

• Materials Science
• Computational Intelligence
• Nanotechnology

Background:

• Nanofluid performance hinges on thermophysical properties.
• Optimizing these properties is crucial for enhancing heat transfer applications.
• Ternary hybrid nanofluids offer potential for superior performance.

Purpose of the Study:

• To develop a hybrid computational intelligence strategy for optimizing ternary hybrid nanofluids.
• To minimize dynamic viscosity and maximize thermal conductivity.
• To identify optimal operating conditions including volume fraction, temperature, and nanomaterial mixing ratio.

Main Methods:

• Integration of machine learning (GMDH, GEP, combinatorial algorithm), multi-objective optimization (MOPSO), and multi-criteria decision-making (TOPSIS).
• Modeling of dynamic viscosity and thermal conductivity using AI techniques.
• Optimization of nanofluid properties based on validated models.

Main Results:

• The combinatorial approach demonstrated high accuracy in modeling (R² = 0.99964-0.99993).
• Optimal volume fractions varied broadly, while optimal temperatures were consistently near 65°C.
• The mixing ratio remained consistent across scenarios, with volume fraction being the key differentiator.

Conclusions:

• The proposed hybrid strategy effectively optimizes ternary hybrid nanofluids.
• Computational intelligence provides a robust framework for enhancing nanofluid thermophysical properties.
• This research offers insights into tailoring nanofluid composition for specific performance targets.