Thermodynamic analysis and performance prediction on dynamic response characteristic of PCHE in 1000 MW S-CO2 coal fired power plant

Highlights: • The CFD method is adopted to establish the dynamic model of the S-CO₂ PCHE. • The dynamic response trends of key parameters on the PCHE are analyzed. • The equilibration time of the PCHE with different conditions are compared. • A neural network is trained to predict the performance of the PCHE. -- Abstract: The studies of supercritical carbon dioxide (S-CO₂) power generation cycles have attracted wide attention from different energy industries in recent years. At present, there is a lack of studies on the dynamic characteristics of S-CO₂ power systems, especially in S-CO₂ coal-fired power plants. In order to further provide the theoretical and technical support for the integration of energy internet and the deep peak-load regulation, it is important to establish the dynamic model of the S-CO₂ power system. Heat exchanger is a key component of the system. Therefore, the dynamic response characteristics of the heat exchanger need to be studied because the related research will lay the foundation of building the dynamic model of the entire S-CO₂ power system. In this paper, the effects of dynamic responses characteristics of thermodynamic parameters (fluid outlet temperature, total surface heat flux and surface heat transfer coefficient of fluid channels) on the printed circuit heat exchanger (PCHE) of 1000 MW S-CO₂ coal-fired power plants are studied in detail. Afterward, a neural network is trained to predict the performance of the PCHE. First, the computational fluid dynamics (CFD) method is adopted to establish the dynamic model of the S-CO₂/S-CO₂ PCHE. Second, when the fluid inlet temperature or the fluid mass flow rate is changed, the dynamic response trends of thermodynamic parameters on the PCHE of 1000 MW S-CO₂ coal-fired power plants are analyzed. Third, the comparisons of the equilibration time of the PCHE with different boundary conditions (the fluid mass flow rate and the fluid inlet temperature) are investigated. Finally, a neural network is trained to predict the performance of the PCHE.