Air Entrainment in Hydraulic Oil: A Comprehensive Study of Influential Parameters Using Computational Fluid Dynamics and Artificial Neural Network

Abstract

Hydraulic power units (HPUs) are vital in industrial applications, but they encounter challenges due to air entrainment in hydraulic oil, which impacts flow stability, reservoir performance, and the environmental footprint. This study aims to quantify the influence of key design parameters—including initial air content, baffle configuration, volumetric flowrate, and oil temperature—on air entrainment behavior in hydraulic reservoirs and to identify the dominant contributors to entrained air volume using computational fluid dynamics (CFD)-based multiphase flow modeling and machine learning analysis. This study combines CFD multiphase flow modeling with machine learning algorithms to enhance understanding of HPU behavior, specifically regarding air entrainment and material efficiency. Through 300 simulations, the analysis systematically examines critical parameters, including initial air content, baffle configuration, volume flowrate, and oil temperature on air distribution and circulation. Results show that increasing baffle numbers can elevate air entrainment up to 13-fold for specific initial air levels while maintaining a Circulation Ratio (Cr) above 2 min effectively reduces air content without expanding reservoir size. A key finding is the potential reduction in reservoir volume by up to 50%, resulting in a 37% decrease in carbon emissions and substantial material savings, with only a 3% increase in air content. This study presents a framework for HPU optimization that addresses both operational stability and environmental sustainability, thereby supporting advancements in hydraulic engineering by striking a balance between efficiency and reduced resource use.