Review of dynamic performance and control strategy of supercritical CO2 Brayton cycle

Abstract

In recent years, the supercritical carbon dioxide Brayton cycle (SCBC) has been regarded as a promising next generation power conversation system, owing to its high efficiency, compact components, applicability for various kinds of heat sources and so on. This paper makes a detailed review of the dynamic performance and control strategy of SCBC. The dynamic simulation model of SCBC is overviewed in detail including different modeling methods of the main component models and validation of system model. As the most inevitable approach to evaluate the dynamic performance of SCBC in practice, existing SCBC test benches concerning SCBC are well collected and presented. Based on these, the open loop dynamic system performances by changing different manipulated variables are reviewed and then various control methods for essential state parameters by different manipulated variables are summarized. Finally, various control strategies of load following and startup/shutdown are clearly presented. With the rapid development of artificial intelligence, combining the core mechanism model and key parameters identification based on experimental data and machine learning to obtain an accurate model within a wide range of working condition is a popular trend in modeling of SCBC. Moreover, deep reinforcement learning will be a potential method for the control strategy in SCBC.