Intelligent Control Method for SCO2 Brayton Cycle Reactors

In recent years, with advancements in turbomachinery and compact heat exchanger technologies, gas-cooled fast reactors (GCFR) using supercritical carbon dioxide (SCO2) as the working fluid have become a research focus. Compared to traditional helium Brayton and steam Rankine cycles, the SCO2 Brayton cycle offers higher efficiency and greater potential for miniaturization. However, the high-pressure and high-temperature operation of SCO2 reactors significantly increases control complexity, and conventional PID controllers show limitations under wide-ranging load changes or operational condition switches.This study proposes a PID controller optimization method based on Deep Deterministic Policy Gradient (DDPG) to dynamically adjust PID parameters for improved control performance. Simulations of step load changes and loss-of-flow accidents were conducted in a 15 MWth SCO2 gas-cooled microreactor. Results demonstrate that the method can quickly respond to load changes, maintain core power stability, and effectively limit core temperature rise during loss-of-flow accidents, ensuring key parameters remain within safety limits. The DDPG-PID controller outperforms conventional PID controllers in dynamic performance and safety, providing a reliable control solution for SCO2 reactors.