Power Control System Design for Heat Pipe Cooled Reactor Based on Model Predictive Control

Due to the small size, the high inherent safety, and easy modularization and expansion, heat-pipe cooled reactor (HPR) has broad application prospects in deep space exploration, deep-sea submarine exploration and other scenarios. However, the HPR conducts thermal energy through the evaporation and condensation of the working fluid inside the heat pipe. This feature makes the HPR a large time-delay system. If the power control system adopts the conventional PID algorithm, there will be a long settling time and large overshoot. Therefore, the model predictive control algorithm is proposed for the power control system to improve the control performance. The HPR linear model, which is developed by linearization of its nonlinear model, is chosen as the predictive model. Based on the predictive model and the reactor power feedback value, the optimal control value is obtained by solving the optimization problem. Due to the discrepancy between the predive model and the actual system response, a steady-state error occurs. To solve this problem, an integral controller is added before the model predictive controller to eliminate the error. The appropriate control system parameters are tuned by trial and error method. Finally, taking the nonlinear HPR model as the controlled object, it is verified by typical transient that the control system has satisfactory control performance. The model predictive control can effectively overcome the influence of the large time-delay characteristics.