Power structure optimization for grid-connected photovoltaic system based on Markov decision processes

Power structure optimization is to determine the planning time, the type and the capacity of power supply, thus the best way to meet the load demand for development. With the help of Markov decision-making process (MDP), it has been established based on the grid-connected PV system generation. Using the optimal strategy of MDP, one step transition probability matrix and the expected total reward criterion have been obtained. According to weather conditions, PV generation was distributed to load, grid power and batteries. The example result shows that the expected total reward criterion of MDP applying to grid-connected photovoltaic system is more powerful.