Robust & nonlinear control of an ultra-supercritical coal fired once-through boiler-turbine unit in order to optimize the uncertain problem

The ultra-supercritical once-through boiler (OTB) unit is an advanced power generation technology with high plant efficiency and low emissions. However, it is difficult to realize a coordinate control for the ultra-supercritical OTB unit to achieve the fast and stable dynamic response during the load tracking and grid frequency disturbances and in the presence of unavoidable uncertainties. In this paper, an accurate grey box multivariable coupled nonlinear model of an ultra-supercritical boiler-turbine unit is considered. Steam pressure at throttle valve, specific enthalpy in separator and active power are adjusted at desired values by manipulation of the fuel rate command, feedwater rate command and throttle valve opening. In addition to the mentioned accurate model, limitations of operational parameters and uncertainty sources are considered in the controller design. This makes this research unique from previous works. According to affine nonlinear control theory, a nonlinear robust sliding mode controller is developed. Then using MATLA B mu - DK iteration syntax, a linear H-infinity (H-infinity) robust control is synthesized with selecting proper weight functions. This method is suitable for all types of power plants with a similar mathematical modeling approach. Simulations in MATLAB Simulink toolbox reveals that both designed controllers achieve appropriate performance in the presence of uncertainties and parameters' limitations, but nonlinear controller has better performance. Finally, deficiencies of a linear controller with respect to the proposed designed robust controllers are denoted.

Automated summary

This paper investigates the control problem of ultra-supercritical once-through boiler (OTB) unit and proposes an accurate grey box multivariable coupled nonlinear model. The steam pressure, specific enthalpy, and active power are controlled by manipulating the fuel rate command, feedwater rate command, and throttle valve opening. The controller design takes into account operational parameter limitations and uncertainty sources. A nonlinear robust sliding mode controller is developed based on affine nonlinear control theory, and a linear H-infinity robust control is synthesized using MATLAB. The simulation results demonstrate that both designed controllers perform well in the presence of uncertainties and limitations, but the nonlinear controller has better performance. The deficiencies of a linear controller compared to the proposed robust controllers are also discussed.