Actuator dynamics compensation for real-time hybrid simulation: an adaptive approach by means of a nonlinear estimator

Abstract

In real-time hybrid simulation, hydraulic actuators, equipped with suitable controllers, are typically used to impose displacements to experimental substructures. Interaction between actuators and physical substructures can result in a nonlinear behaviour of the overall experimental testing system (ETS), making the controller design very challenging. The accuracy of the hydraulic actuation system (HAS) is very crucial because actuator displacement errors lead to incorrect simulation results. For this purpose, several methods have been developed by researchers in order to compensate tracking error of HASs. This paper presents a novel adaptive compensator that takes into account the actual ETS dynamics by adopting an extend Kalman filter for the real-time estimation of the ETS model parameters. The adaptive approach improves the actuator control accuracy and avoids ad hoc system identification procedures. The novel compensator has been verified experimentally on a test rig for seismic isolator shear tests. The feasibility of the proposed compensation method has been also demonstrated through real-time hybrid simulation of a building with a base isolation system. Both numerical and experimental results confirmed that the proposed compensation strategy provides good results even in the case of inevitable nonlinearities of the ETS. Furthermore, the method has also demonstrated good performance in terms of stability and robustness with respect to variations of the operating conditions.