관류형 보일러의 비선형 저자모델 개발에 관한 연구

A Study on Development of Nonlinear Low Order Models for an Once-through Type Boiler

By use of the real recorded data obtained from a power plant, nonlinear low-order state space models are developed for an once-through type power plant boiler. In order to understand the boiler dynamics and to use for the back-up controller design, the unknown model parameters have been estimated using the nonlinear estimation technique, i.e. Extended Kalman Filter method. It is shown that the simulation results coincide with the measurement data within 5% relative error range, which are acceptable from a back-up controller design point of view.