Definition of model-based control strategies for the Molten Salt Fast Reactor nuclear power plant

Abstract

In this paper, different control strategies for the full-power mode operation (i.e. from 50% to 100% of reactor rated power) of the MSFR are identified, developed, and their performances tested, starting from the results of a MSFR power plant simulator. In order to support the qualitative results gained from the free-dynamics outcomes, a technique known as Relative Gain Array is employed to obtain a quantitative measure of the degree of interaction between the various input and output variables, and to identify the most favorable couplings. In the development of control strategies, particular attention is placed on several specific aspects, in particular the load-following capabilities of the power plant, the need to keep the molten salts temperatures within a rather narrow window to avoid boiling or freezing risks during operational transients, and the need to maintain the operating conditions of the energy conversion system as constant as possible. Four different decentralized-feedback control strategies are implemented – with different number of controlled outputs – by employing conventional PID controllers, which are designed and tuned with appropriate phase and gain margins to account for modeling uncertainties. The different control strategies are then tested on the power plant simulator in two typical controlled operational transients, highlighting the very promising behavior of the MSFR in terms of load-following capabilities. The results of the present work will also provide valuable insights in support to the optimization and finalization of the MSFR power plant design.