Drag reduction in liquid flows is pivotal for energy conservation and sustainable development, as it directly mitigates energy consumption across liquid-driven systems—including marine transportation, oil pipeline networks, and industrial liquid transportation. Against rising energy efficiency demands, machine learning (ML) has emerged as a transformative tool to advance drag reduction efforts. By leveraging ML algorithms to analyze large-scale fluid flow datasets, precise predictive models of fluid dynamics can be developed, enabling accurate forecasting of flow behaviors and drag reduction efficacy under diverse operating conditions. Such capabilities offer novel insights and methodologies to accelerate innovation in fluid drag reduction technologies. This paper presents a systematic review of recent advances and practical applications of ML in fluid drag reduction, with a specific focus on liquid drag reduction. It critically examines key application domains, including the optimization of flow control strategies for drag reduction, the enhancement of ship hull design to minimize hydrodynamic resistance, and the tuning of bubble-induced drag reduction parameters. Additionally, the review provides a comprehensive analysis of prevailing challenges and limitations and outlines promising future research directions, emphasizing the development of hybrid ML-physics models, and the scaling of ML-driven solutions for industrial deployment—with the aim of guiding the advancement of energy-efficient fluid systems.
Li et al. (Thu,) studied this question.