Please ensure to cite the paper when utilizing the dataset in a research study. Refer to the paper link or BibTeX provided below.
Nanofluids have emerged as transformative materials in advanced thermal management, combining base fluids with nanoparticles to achieve superior thermophysical characteristics such as enhanced thermal conductivity and heat transfer efficiency. These advanced fluids are indispensable for a variety of industrial applications, including heat exchangers, cooling systems, and renewable energy technologies, where efficient thermal management is critical.
Accurately predicting the thermal conductivity of nanofluids is essential for optimizing their performance and designing intelligent thermal systems. However, the complex, nonlinear relationships between particle properties, temperature, and concentration pose significant challenges for traditional predictive models. This dataset facilitates the development of robust machine learning and deep learning frameworks, capable of addressing data scarcity and providing high-fidelity predictions for both experimental and theoretical thermal conductivity scenarios.
This dataset is designed for predicting the thermal conductivity of nanofluids using a robust set of experimental and theoretical parameters. The dataset comprises 278 samples across twelve nanofluid combinations, involving four types of nanoparticles and three base fluids. Each sample is characterized by:
Input Parameters:
- Nanoparticle Material: Copper (Cu), Copper(II) Oxide (CuO), Aluminum (Al), and Aluminum Oxide (Al2O3)
- Base Fluid: Water, ethylene glycol, and transformer oil
- Physical Properties: Temperature (°C), particle size (nm), and particle volume fraction (%)
- Conductivity Properties: Thermal conductivity of the particle (kp) and thermal conductivity of the liquid medium (km)
Output Parameter:
- Exp-TC: Thermal conductivity based on experimentally reported values.
- KKL-TC: Effective thermal conductivity calculated using the Koo-Kleinstreuer-Li (KKL) model
Learning heat: High-fidelity experimental and Koo–Kleinstreuer–Li thermal conductivity predictions in nanofluids via advanced data augmentation and metaheuristic search
The dataset is also available to be downloaded from following sites, given the paper is referenced in the study:
If you are using the dataset, please cite using this BibTeX:
@article{sheththermal,
title = {Learning heat: High-fidelity experimental and Koo–Kleinstreuer–Li thermal conductivity predictions in nanofluids via advanced data augmentation and metaheuristic search},
journal = {International Communications in Heat and Mass Transfer},
volume = {171},
pages = {110022},
year = {2026},
issn = {0735-1933},
doi = {https://doi.org/10.1016/j.icheatmasstransfer.2025.110022},
url = {https://www.sciencedirect.com/science/article/pii/S0735193325014484},
author = {Farhan Sheth and Priya Mathur and Hammad Shaikh and Dheeraj Kumar and Shweta Mishra and Amit Kumar Gupta},
keywords = {Heat transfer, Thermal conductivity, Material science modeling, Machine learning, Deep learning, Optimization, Augmentation},
}