Experimental Analysis and Genetic Algorithm Parameter Identification of a Reduced-order Model for a Single Spring-mass System Coupled to a Magnetorheological Tuned Liquid Column Damper

Abstract Purpose The present study aims to investigate the dynamic behavior of a magnetorheological tuned liquid column damper (MR-TLCD) coupled to a single degree of freedom (1 DoF) mechanical system and to validate a reduced-order analytical model through parametric identification. Methods Forced vibration tests are performed using a discrete sine-sweep excitation technique to obtain accelerance frequency response functions of the coupled system, with and without magnetic field application. The proposed reduced-order 2 DoF analytical model is formulated using the equivalent parameters identified for each uncoupled subsystem. Model parameters are identified using a genetic-algorithm-based optimization procedure that minimizes the least-squares error between experimental and analytical frequency response functions. Results The experimental approach investigates the natural frequency and damping ratio variations in a 1 DoF mechanical system (primary system) coupled to an MR-TLCD, considering scenarios with and without applied magnetic field. The experimental results show a significant increase in the equivalent damping ratio of the MR-TLCD when the magnetic field is applied, in agreement with previous studies. In addition, a small but consistent increase in the identified natural frequency of the MR-TLCD is observed under magnetic excitation. The reduced-order model parameters identified by the genetic algorithm implementation lead to simulated responses in good agreement with the experimentally measured frequency response functions. Conclusion The proposed framework demonstrates the effectiveness of genetic algorithm-based parameter identification for a reduced-order model. Moreover, this study experimentally investigates the influence of magnetic field on both damping and stiffness-related properties of the system.