电站汽温控制系统结构复杂,机组运行过程中影响主汽温稳定的因素众多。主蒸汽流量变化、锅炉燃烧状况改变等因素均可能干扰主汽温稳定,影响机组运行的经济性和安全性。基于模型的预测控制算法在应对诸如煤质变化引起的烟气侧不可测扰动时,由于缺乏对扰动的建模,控制效果不理想。针对该问题,通过对过热器进行机理建模,并在额定工况点进行局部线性化获得状态空间模型,提出基于多模型集的主汽温扰动抑制预测控制方法。当不考虑不可测扰动时,其等价于普通的预测控制算法。仿真结果表明,该方法能够有效抑制进入汽温系统的不可测扰动,进一步提高汽温的调节品质。
There are many factors affecting the stability of the main steam temperature(MST) in thermal power plant operation due to its complex structure.Variations of the main steam flow and boiler combustion condition are all possible disturbances affecting the stability of the MST,and further threatening the safety and economy of the entire power plant.Conventional model predictive control(MPC),because of its lack of disturbances modeling,often fails to reject unknown disturbances caused by uncertain disturbances such as coal quality variations in the fuel gas side.To solve this problem,this paper proposed a multiple disturbance model predictive control(MDMPC) for the MST system.An analytical superheater model was first established,and then a local state space model was derived at the rated load and was used in simulations.It proves that MDMPC is equivalent to the conventional MPC when no active disturbances exist.Simulation results also demonstrate the MDMPC's successful rejection of unknown disturbances entering the MST system,which further improves the control quality of the MST system.