Previous approaches to dispatching nuclear integrated energy systems (NIES) have focused on the profitability and flexibility of these systems to operate on energy grids with highly variable pricing. Because of the complexity of these systems to model and design, less focus has been given to ensuring these dispatch strategies are physically achievable. It is imperative to find approaches that allow the system to within the desired NIES operating conditions with limited future foresight. This research looks to employ deep reinforcement learning (DRL) and a dynamic system model written in Modelica to find a safe and profitable dispatch strategy for a solar nuclear hybrid design. The DRL agent is shown to find a novel dispatch strategy for both power ramping and power levels that respects operational limits. This DRL dispatch is compared to other dispatching strategies including an optimal design solution found from mixed integer linear programming. It is found that incorporating the physics of such a tightly coupled IES limits the profitability of this dispatch strategy. As a result, the MILP solution overestimates the design’s generated revenue. DRL on the other hand can significantly reduce the number of breaches of operational conditions during energy arbitrage while maintaining profitability. The work paves the way for a more detailed assessment of IES profitability and could be used to aid operator decisions on future IES projects.
