Generalized Discrete-time nonlinear disturbance observer based fuzzy model predictive control for boiler–turbine systems

• Published in: ISA transactions Vol. 90; pp. 89 - 106
• Main Authors: Kong, Lei, Yuan, Jingqi
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.07.2019
• Subjects: Disturbance observer based control; Fuzzy model predictive control; Generalized discrete-time nonlinear disturbance observer; High-order disturbance; Input constraint; Disturbance observer based control; Generalized discrete-time nonlinear disturbance observer; High-order disturbance; Input constraint; Fuzzy model predictive control
• ISSN: 0019-0578; 1879-2022; 1879-2022
• DOI: 10.1016/j.isatra.2019.01.003

The boiler–turbine system (BTS) is usually subject to the tight input constraint, the strong nonlinearity and the complex disturbance, which makes the control a challenging task To this end, a disturbance observer based fuzzy model predictive control (DOBFMPC) scheme is proposed for the BTS in this paper. The generalized discrete-time nonlinear disturbance observer (GDNDO) is first developed to estimate the higher-order disturbance by systematically extending the conventional nonlinear disturbance observer. The GDNDO exhibits a series structure of the internal states, and can precisely estimate the disturbance if its order is equal to or greater than that of the disturbance In addition, a baseline fuzzy model predictive control (FMPC) law is synthesized on the fuzzy model. With FMPC, the asymptotic stability is guaranteed, and meanwhile the input constraints are satisfied by both the free control variables and the future control inputs in the form of the state feedback law. At last, the disturbance estimate and the FMPC are applied to constitute the DOBFMPC law. With the proper design of the disturbance compensation gain, the disturbance influence is removed from the output channels by the composite DOBFMPC law at the steady state. Simulations for a 300 MW subcritical BTS well demonstrate the effectiveness of the proposed control scheme. [Display omitted] •A disturbance-observer-based fuzzy model predictive control scheme is proposed.•A novel generalized discrete-time nonlinear disturbance observer is designed.•Input constraints are satisfied by both the free and future control inputs in fuzzy MPC.•Disturbance compensation gain design removes the disturbance effect at steady state.