基于子空间辨识和增广模型的烟气脱硝系统预测控制

Predictive control of flue gas denitration system based on subspace identification and augmented model

选择性催化还原(SCR)烟气脱硝系统具有大惯性、大延迟的特点,常用的PID控制器不能很好地适用于该系统。结合某火电机组实际运行数据,建立了基于子空间方法的SCR烟气脱硝系统模型,并将模型计算结果与实测数据进行对比,验证了模型具有较强的鲁棒性。将子空间辨识所得脱硝系统状态空间模型进行变换,得到其增广模型,设计了基于增广模型的预测控制器,并将该控制器仿真结果与系统原有PID控制器输出结果及实际运行数据进行了对比。结果表明,与系统原有PID控制器相比,本文设计的预测控制器能够很好地跟踪脱硝系统期望,具有更好的控制效果。

Selective catalytic reduction（SCR）flue gas denitration system has characteristics of large inertia and large delay,the conventional PID controller is not well suited for the system.To solve this problem,combing with the actual operation data of a thermal power unit,a model for SCR flue gas denitrification system based on subspace method was established.Comparison between the calculated results and the measured data verified this model has strong robustness.Moreover,the state space model of the denitration system identified by the subspace identification was transformed,and the augmented model was obtained.Then,the predictive controller based on the augmented model was designed,and the results simulated by the predictive controller were compared with that by the conventional PID controller were compared.The results show this predictive controller can track the denitration system expectations well,and has better control effect.