电厂锅炉除氧器系统的解耦广义预测控制研究

Study on Decoupling Generalized Predictive Control of Boiler Deaerator System in Power Plant

由于除氧器具有非线性、大迟延、强耦合和模型时变的特性，传统的PID调节器难以使除氧控制系统达到理想的控制效果。为了使除氧器水位和冷凝泵出口压力的控制取得满意的效果，采用解耦广义预测控制方法。该算法首先利用预测模型得到系统未来时刻的输出，然后将设定输出值和预测值间的预测误差变化率作为自适应控制器的输入，控制器利用最小二乘算法推理得到控制输出，并在系统中增加一个相互耦合项，来调整原有的自适应控制规律，获取未知参数信息并有效地抑制强耦合造成的控制量波动。通过仿真表明，该算法计算简单、鲁棒性较强、控制品质较高，运用此算法的该除氧系统比常规系统具有更优越的性能。

Because deaerator is provided with the characteristics of nonlinearity, large delay, strong coupling and model time- varying, the traditional PID regulator of deoxidizing control system is difficult to achieve the desired control effect. In order to make the control for the deaerator water level and the condensate pump outlet pressure to achieve the satisfactory results, the decoupling generalized predictive control methods are adopted. The algorithm firstly gets the system output of the next moment by making use of the predictive models, and then sets the rate of change of prediction error between the output value and the predictive value as the input of the adaptive controller, the controller obtains the control output by using the reasoning of reeursive least square algorithm and a intercoupling term is added in the system to adjust the original adaptive control law to obtain the unknown parameters information and effectively restrain the fluctuation of controlled quantity caused by the strong coupling. The simulation result shows that the algorithm is characterized with the advantages of simple computation, stronger robustness and higher control quality, the deoxidizing system with the algorithm is of the preferable performances than the conventional system.