Shahriar Osfouri

Passive thermal management has gained significant attention due to its simple structure, low energy consumption, and suitability for a wide range of applications. However, the design of passive thermal management systems remains challenging due to numerous design parameters, experimental uncertainties, high computational costs, and complex geometries. These factors hinder the development of efficient and accurate design methodologies. One promising approach to addressing these challenges is the application of artificial intelligence (AI) for performance prediction and design optimization. This review examines recent advances in the application of AI in passive thermal management, with a particular focus on heat pipes, thermosyphons, heat sinks, vapor chambers, and phase change materials. Both data-driven and physics-based modeling approaches, as well as hybrid methods that combine multiple algorithms, are discussed focusing on their ability to improve thermal performance, particularly by reducing thermal resistance. The reviewed studies demonstrate that AI-based models can predict the thermal behavior of passive thermal systems with high accuracy and significantly improve their thermal performance. Nevertheless, the reliability of these predictions is strongly dependent on data quality, dataset size, and appropriate model selection, highlighting key challenges and future directions.