Optimal Scheduling of Rural Microgrid Based on Approximate Dynamic Programming

Rural power grid is an important part of smart grid construction. This paper proposes a rural microgrid optimal scheduling method based on approximate dynamic programming (ADP). Firstly, a rural microgrid optimization model with electric vehicles (EV) is established, and the multi-period optimization problem is modeled as a Markov decision process (MDP). Then, the piecewise linear function based on EV energy storage is used to approximate the value function to obtain the approximate global optimal solution, so as to avoid the dimensionality curse and optimize the system online. Finally, the operating costs obtained by the simulation are compared with the operating costs without considering the access of EV to the power grid. It can be obtained that the operating cost of microgrid is smaller than that without considering electric vehicles. At the same time, the ADP online optimization results are better than the myopic strategy results.