Performance analysis and optimization of water-lubricated herringbone groove journal bearings: A novel GA-ANN approach

Abstract

Water lubrication systems have emerged as a promising pollution-free solution for underwater rotor-bearing solutions. Further, texture water-lubricated bearings such as Herringbone Groove Journal Bearings (HGJBs) enhance the stability and performance of underwater vehicle rotors. The present study presents a comprehensive investigation into optimizing HGJB design parameters using Artificial Neural Networks (ANN) and Genetic Algorithm (GA) based techniques. The influence of bearing parameters such as eccentricity ratio, speed, groove depth, groove angle, and the number of grooves on the dynamic characteristics is analyzed. A key novelty of this research is the integration of ANN-based performance prediction with GA-driven optimization, ensuring an efficient and accurate determination of optimal design parameters. The optimized HGJB configuration is compared with the traditional Plain Journal Bearings (PJBs), demonstrating superior stiffness characteristics and enhanced vibration suppression capabilities. The results reveal that HGJBs provide improved dynamic performance, making them highly suitable for critical applications. Additionally, various weighted optimization strategies are explored to balance stiffness and damping, offering tailored solutions for different operational requirements. The findings contribute significantly to the design and performance enhancement of water-lubricated bearings, paving the way for their broader adoption in advanced engineering applications.