부산대학교 도서관

In this paper, a parameter estimation approach is developed for speed governor and nonlinear hydraulic turbine. Performance of the method is compared in several aspects, including terms of convergence rate, time calculation, robustness to various perturbations and inputs/outputs. Nine different types of perturbations are investigated in this study to evaluate the behavior of the identification problem. Also, precise models are utilized for the PID controller, servomotor and hydraulic turbine. Quasi steady state (QSS) method is also used to increase computational efficiency. The perturbations analysis is carried out in detail and applied to virtual and real tests which are obtained from a hardware-in-the-loop (HIL) test bench. The results show a comparison between estimation errors after various perturbation concurrences to obtain the best and the worst perturbations for the estimation process. It also compares between several estimation methods.