This study investigates the rapid identification of dynamic characteristics for squeeze film dampers under static eccentricity conditions. Fluid domain models of squeeze film dampers with varying static eccentricities and oil film widths were established to analyze their influence on the dynamic coefficients. Four machine learning models were developed and compared for their accuracy in predicting the dampers' dynamic characteristics. The results indicate that an increase in static eccentricity significantly enhances damping and stiffness, while a larger oil film width concurrently increases both damping and stiffness. The PSO-BP neural network demonstrated optimal performance in predicting dynamic characteristics under static eccentricity conditions. The findings provide a foundation for the rapid design and intelligent control of dampers.
 

Squeeze film damper,Static eccentricity,Machine learning,Dynamic characteristics