Experimental evaluation of an adaptive inverse compensation technique for real-timesimulation of a large-scale magneto-rheological fluid damper

Abstract

Magneto-rheological (MR) dampers are semi-active control devices whose characteristicsare varied under different current inputs in accordance with semi-active control laws toachieve optimized vibration control of a structural system. Experimental evaluation of theeffectiveness of MR dampers for seismic hazard mitigation using selected control laws isnecessary to enable performance-based design procedures to be developed and for thesedevices to become accepted by the practical design community. Real-time hybrid simulationprovides an economical and efficient experimental technique which enables both thedamper rate dependence and the damper–structure interaction to be accounted for. Asuccessful real-time hybrid simulation requires accurate actuator control to achieve reliableexperimental results. A time delay and actuator time lag (referred to hereafteras simply the delay) can be introduced into the actuator response due to statedetermination, communication, and servo-hydraulic dynamics. The variable currentinputs and resulting variable forces induced by semi-active control laws poseadditional challenges for actuator control by introducing variable delay in a real-timehybrid simulation. In this paper a newly developed adaptive inverse compensationtechnique is experimentally evaluated for application in real-time hybrid simulationinvolving an MR damper subjected to band-limited white noise-generated randomdisplacements and variable current inputs. Actuator control is assessed using differentevaluation criteria. The adaptive inverse compensation method is demonstratedto achieve good actuator control and therefore shows good potential for use inreal-time hybrid simulation of structural systems with semi-active MR dampers.