A Performance Impovement Method for Optimal Operation of A Domestic Building Microgrid

Microgrid technology brings opportunities to improve the building energy efficiency. While the uncertainties of renewable power sources and the variability of occupants' energy demand brings challenges to the stable and secure operation of the utility grid. Considering the optimal operation of building microgrid under both supply and demand sides' uncertainties, the following major contributions are made in this paper. First, the building microgrid operation problem is formulated as a Markov decision process (MDP) with the objective minimizing the electricity cost. Second, a policy improvement (PI) method with stochastic simulation is proposed to accelerate the on-line solving process. Third, numerical tests are made on a typical domestic building microgrid and PI method is applied to a rule-based control policy and a greedy control policy. Numerical results demonstrate their performance improvement and the effectiveness of the proposed method is validated.