A load following reactivity control system for nuclear microreactors

Abstract

Microreactors are scaled-down versions of small modular nuclear reactors designed to simplify the deployment and reduce the cost of nuclear energy. One of the key requirements for achieving these goals is the development of reliable load-following control systems. In this study, we designed and analyzed three load-following control systems for a microreactor design. The microreactor core is modeled using a nonlinear point kinetics model coupled with thermal-hydraulic, fission product poison, and reactivity models. The first control method implemented was a nonlinear second-order super-twisting sliding mode control system (STC), enhanced with a moving average filter to mitigate chattering—a high-frequency phenomenon that can damage actuators. The second approach utilized a nonlinear model predictive control (NMPC) scheme with an extended Kalman filter for improved state estimation. The final design was a PID controller, optimized with anti-windup compensation for the integral term and a filter for the derivative component to handle noise. We evaluated these control systems through simulation experiments, focusing on their stability, tracking performance, and control effort. The results show that all three control systems achieved the desired load-following capability. While the PID controller required the least control effort, error analysis metrics such as integral absolute error (IAE) and integral time absolute error (ITAE) revealed that its performance deteriorates over time due to its linear nature. In contrast, the second-order sliding mode controller (STC) and NMPC demonstrated superior error handling and robustness than PID concerning accuracy and stability. Nonetheless, unlike the nonlinear model predictive controller, the second-order sliding mode control system still suffers from chattering.