Study on the design of unattended SCRS full-condition adaptive bypass flow systems

Abstract

The Supercritical CO₂ Reactor System (SCRS), a member of the fourth-generation reactor family, offers promising prospects for practical application. The use of S-CO₂ as a working fluid leads to a more compact reactor design and higher energy conversion efficiency, positioning it as a strong contender for next-generation unattended intelligent reactor systems. However, the thermophysical properties of S-CO₂ exhibit dramatic changes near the critical point, placing higher demands on the control system to ensure SCRS stability. To address the challenge that PID control theory struggles to meet the complex control demands of supercritical reactors, this paper proposes an intelligent control-based approach. It utilizes neural networks to adaptively adjust the outputs of fuzzy control and PID control in the bypass flow control system. The optimized controller outperforms the original PI controller during normal operational transients and exhibits robust performance under accident conditions. The work presented in this paper provides valuable insights for the design and optimization of control systems in supercritical reactors for practical engineering applications.