摘要
针对锅炉-汽轮机组易受扰动、非线性等特点,提出一种基于扩展状态观测器的模糊模型预测控制方法。首先,基于实际被控对象的输入输出数据建立1 000 MW锅炉-汽轮机组的模糊模型;然后利用输出代价函数在模糊模型上设计一个稳定的模糊模型预测控制器,以实现节流蒸汽压力、分离器蒸汽焓和负荷的稳定、快速和准确跟踪。将一种改进的线性扩展状态观测器与预测控制器相结合,以增强其抗干扰性能,该观测器可以在线估计系统的行为变化和未知干扰,通过设计合适的李雅普诺夫函数证明了整个控制系统的闭环稳定性。在对某1 000 MW锅炉-汽轮机组模型的节流阀压力、蒸汽分离焓和有效功率(负荷)参数的仿真中,具有扩展状态观测器的模糊模型预测控制超调量和调节时间等指标均优于积分模型预测控制、稳定模型预测跟踪控制器(stable model predictive tracking controller,SMPTC)和广义扩展状态观测器的控制(generalized extended state observer control,GESOBC),证明了所提出方法的有效性。
A fuzzy model predictive control method based on extended state observer is proposed for boiler turbine units,which are prone to disturbances and nonlinearity.Firstly,a fuzzy model of a 1 000 MW boiler turbine unit was established based on the input and output data of the actual controlled object;Then,using the output cost function,a stable fuzzy model predictive controller was designed on the fuzzy model to achieve stable,fast,and accurate tracking of throttling steam pressure,separator steam enthalpy and load.Combining an improved linear extended state observer with a predictive controller to enhance its anti-interference performance,the observer could estimate the behavior changes and unknown disturbances of the system online.The closed-loop stability of the entire control system was demonstrated by designing an appropriate Lyapunov function.In the simulation of the throttle valve pressure,steam separation enthalpy,and effective power(load) parameters of a 1 000 MW boiler turbine unit model,the extended state observer fuzzy model predictive control(ESOFMPC) with extended state observer outperforms the integral model predictive control in terms of overshoot and regulation time.The stability model predictive tracking controller(SMPTC) and generalized extended state observer control(GESOBC) demonstrate the effectiveness of the proposed method.
作者
夏建国
张桂林
吕先平
王炳奎
XIA Jianguo;ZHANG Guilin;LYU Xianping;WANG Bingkui(Baowu Group Yicheng Steel Co.,Ltd.,Ezhou 436000,China)
出处
《湖北大学学报（自然科学版）》
CAS
2024年第3期384-391,共8页
Journal of Hubei University：Natural Science
基金
安徽高校自然科学研究重大项目(KJ2021ZD0042)
安徽省重点研究与开发计划项目(2022f04020004)资助。
关键词
汽轮机组
稳定模型预测控制
模糊控制
扩张状态观测器
steam turbine unit
stable model predictive control
Fuzzy control
extended state observer
