SCI和EI收录∣中国化工学会会刊

中国化学工程学报 ›› 2020, Vol. 28 ›› Issue (3): 832-845.DOI: 10.1016/j.cjche.2019.07.017

• Energy, Resources and Environmental Technology • 上一篇    下一篇

Dynamic optimization oriented modeling and nonlinear model predictive control of the wet limestone FGD system

Lukuan Yang, Wenqi Zhong, Li Sun, Xi Chen, Yingjuan Shao   

  1. Key Laboratory of Energy Conversion and Process Measurement and Control Ministry of Education, School of Energy and Environment, Southeast University, Nanjing 210096, China
  • 收稿日期:2019-03-25 修回日期:2019-07-12 出版日期:2020-03-28 发布日期:2020-06-11
  • 通讯作者: Wenqi Zhong
  • 作者简介:Lukuan Yang,E-mail addresses:220170471@seu.edu.cn;Li Sun,E-mail addresses:sunli12@seu.edu.cn;Xi Chen,E-mail addresses:chenxiseu@seu.edu.cn;Yingjuan Shao,E-mail addresses:yjshao@seu.edu.cn

Dynamic optimization oriented modeling and nonlinear model predictive control of the wet limestone FGD system

Lukuan Yang, Wenqi Zhong, Li Sun, Xi Chen, Yingjuan Shao   

  1. Key Laboratory of Energy Conversion and Process Measurement and Control Ministry of Education, School of Energy and Environment, Southeast University, Nanjing 210096, China
  • Received:2019-03-25 Revised:2019-07-12 Online:2020-03-28 Published:2020-06-11
  • Contact: Wenqi Zhong

摘要: Nonlinear model predictive control (NMPC) scheme is an effective method of multi-objective optimization control in complex industrial systems. In this paper, a NMPC scheme for the wet limestone flue gas desulphurization (WFGD) system is proposed which provides a more flexible framework of optimal control and decision-making compared with PID scheme. At first, a mathematical model of the FGD process is deduced which is suitable for NMPC structure. To equipoise the model's accuracy and conciseness, the wet limestone FGD system is separated into several modules. Based on the conservation laws, a model with reasonable simplification is developed to describe dynamics of different modules for the purpose of controller design. Then, by addressing economic objectives directly into the NMPC scheme, the NMPC controller can minimize economic cost and track the set-point simultaneously. The accuracy of model is validated by the field data of a 1000 MW thermal power plant in Henan Province, China. The simulation results show that the NMPC strategy improves the economic performance and ensures the emission requirement at the same time. In the meantime, the control scheme satisfies the multiobjective control requirements under complex operation conditions (e.g., boiler load fluctuation and set point variation). The mathematical model and NMPC structure provides the basic work for the future development of advanced optimized control algorithms in the wet limestone FGD systems.

关键词: Wet limestone flue gas desulphurization (WFGD) system, Modeling, Nonlinear model predictive control (NMPC), Multi-objective optimization

Abstract: Nonlinear model predictive control (NMPC) scheme is an effective method of multi-objective optimization control in complex industrial systems. In this paper, a NMPC scheme for the wet limestone flue gas desulphurization (WFGD) system is proposed which provides a more flexible framework of optimal control and decision-making compared with PID scheme. At first, a mathematical model of the FGD process is deduced which is suitable for NMPC structure. To equipoise the model's accuracy and conciseness, the wet limestone FGD system is separated into several modules. Based on the conservation laws, a model with reasonable simplification is developed to describe dynamics of different modules for the purpose of controller design. Then, by addressing economic objectives directly into the NMPC scheme, the NMPC controller can minimize economic cost and track the set-point simultaneously. The accuracy of model is validated by the field data of a 1000 MW thermal power plant in Henan Province, China. The simulation results show that the NMPC strategy improves the economic performance and ensures the emission requirement at the same time. In the meantime, the control scheme satisfies the multiobjective control requirements under complex operation conditions (e.g., boiler load fluctuation and set point variation). The mathematical model and NMPC structure provides the basic work for the future development of advanced optimized control algorithms in the wet limestone FGD systems.

Key words: Wet limestone flue gas desulphurization (WFGD) system, Modeling, Nonlinear model predictive control (NMPC), Multi-objective optimization