System-level modeling, analysis and coordinated control design for the pressurized water reactor nuclear power system

• Published in: Energy (Oxford) 2023-11, Vol.283, p.128472, Article 128472
• Main Authors: Cui, Chengcheng, Zhang, Junli, Shen, Jiong
• Format: Article
• Language: English
• Subjects: Characteristics analysis; Modeling; Nonlinear model predictive control; Nuclear power system; Offset-free design
• ISSN: 0360-5442
• DOI: 10.1016/j.energy.2023.128472

A flexible operation of the nuclear power system (NPS) over a large operating range is emphasized when it is used as marine power. The current study exposes two bottlenecks in this field: (1) lack of the system-level high-quality dynamic model whose structure is simple enough for control design limits the implementation of the model-based controller; (2) nonlinearity problem due to the large load variation poses challenges for NPS to maintain its performance over a wide operating range. For these reasons, a first-hand simplified nonlinear model is developed and validated for the pressurized water reactor nuclear power system (PWRNPS) through mixed modeling methods of mechanism analysis and data identification. An optimal 50% full-power compromise steady-state operation scheme is first proposed for PWRNPS to balance its primary coolant temperature and secondary steam pressure. Using the simplified nonlinear model as the predictive model, a novel nonlinear model predictive controller (NMPC) is designed to realize the wide-range flexible operation of PWRNPS. Especially, an integrating tracking error is innovatively added to the cost function of NMPC to achieve offset-free control. Two groups of simulations on a complex PWRNPS process developed on the MATLAB/SIMULINK platform demonstrate the effectiveness of the proposed methods. •New insights into the characteristics of the pressurized water reactor nuclear power system (PWRNPS) are presented.•A 50% full-power optimal compromise operation scheme is proposed for PWRNPS.•A first-hand system-level simplified nonlinear model is developed for PWRNPS.•A novel offset-free strategy is proposed to eliminate the steady-state error of the controlled variables.