Deadline Date: 31 December 2025
The growing complexity of thermal systems in aerospace, energy conversion, and advanced manufacturing demands a paradigm shift beyond traditional numerical methods. While computational fluid dynamics (CFD) and finite element analysis (FEA) have been cornerstones of heat transfer modeling, challenges persist in handling nonlinearities, multi-phase interactions, and real-time optimizationareas where artificial intelligence (AI) and machine learning (ML) offer transformative potential.
This Special Issue of RIMNI seeks to compile pioneering research at the intersection of AI and heat transfer, addressing critical gaps between theoretical ML advancements and their engineering applications. We highlight computational strategies where AI augments or replaces conventional methods to achieve:
Higher accuracy in turbulent/transitional flow predictions,
Reduced computational cost for multi-scale problems,
Uncertainty quantification in complex boundary conditions,
Autonomous optimization of thermal systems (e.g., heat exchangers, electronic cooling).
Key Focus Areas:
AI-Enhanced Modeling:
Neural networks for conjugate heat transfer,
Hyperparameter optimization in thermal surrogate models.
Hybrid Numerical-AI Frameworks:
Coupling CFD with reinforcement learning for active flow control,
Generative adversarial networks (GANs) for synthetic thermal dataset generation.
Industrial Applications:
Digital twins of power plant cooling systems,
AI-driven topology optimization for additive manufacturing heat sinks.
Fundamental Advances:
Interpretable ML for uncovering hidden heat transfer correlations,
Transfer learning across dissimilar thermal systems.
This issue aims to establish best practices for integrating AI into heat transfer workflows while critically examining limitations (e.g., data scarcity, overfitting risks). Contributions should demonstrate rigorous validation against experimental or high-fidelity numerical benchmarks.
The growing complexity of thermal systems in aerospace, energy conversion, and advanced manufacturing demands a paradigm shift beyond traditional numerical methods. While computational fluid dynamics (CFD) and finite element analysis (FEA) have been cornerstones of heat transfer modeling, challenges persist in handling nonlinearities, multi-phase interactions, and real-time optimizationareas ... show more
Editorial board
