A Novel Model-Based Adaptive Feedforward-Feedback Control Method for Real-Time Hybrid Simulation considering Additive Error Model

Adaptive control methods have been widely adopted to handle the variable time delay in real-time hybrid simulation (RTHS). Nevertheless, the initial parameter settings in adaptive control law, the parameter estimation method, and the testing system nonlinearity will affect RTHS’s accuracy and stability at different levels. To this end, this study proposes a novel model-based adaptive feedforward-feedback control method that considers an additive error model. In the proposed method, the time delay and amplitude discrepancy are roughly compensated by a feedforward controller and then finely reduced by an adaptive controller, and an outer-loop control formed by the feedback controller is introduced to improve the ability and robustness furthermore. What’s more, the testing system, composed of the transfer system and physical specimen, is divided into the nominal and additive error models. The feedforward controller is devised using the inverse nominal model, whose parameters are constant. The adaptive controller is designed to adopt a discrete-time additive error model, in which the parameters are identified online by the Kalman filter. Numerical simulations, parametric studies, and actual experiments were carried out to inspect the feasibility and effectiveness of this method thoroughly. Results indicate that the proposed method can effectively improve the accuracy and stability of RTHS and significantly reduce the dependence on the adaptive control law. Moreover, the proposed method exhibits strong robustness and is, therefore, useful in RTHS.