Optimal Operation of a Hydrogen-based Building Multi-Energy System under Uncertainties

This paper investigates an optimal operation problem of a hydrogen-based building multi-energy system (HBMES). Specifically, we first formulate an operational cost minimization problem of HBMES. Since there are uncertain parameters, inexplicit building thermal dynamics model, temporally coupled operational constraints, as well as the coupling between electricity and heat, solving the minimization problem is nontrivial. To overcome the challenge, we reformulate the problem as a Markov decision process (MDP). Then, we design an algorithm to solve the MDP based on deep deterministic policy gradients and prioritized experience replay. The designed algorithm supports real-time decision without any process of searching optimal solution. Performance evaluation verifies the effectiveness and robustness of the designed algorithm.