Medium-voltage distribution network closed-loop economic operation optimization method and system based on deep learning

The invention provides a medium-voltage power distribution network closed-loop economic operation optimization method and system based on deep learning, and the method comprises the steps: constructing an active and voltage mathematical model of each node through employing a deep belief network (DBN) by employing the historical data of each node and a power injection point in a closed-loop operation network; meanwhile, a deep belief network DBN is optimized by adopting a CSO algorithm; and changing the reactive power injection condition of each power injection point by taking the operation economy of the closed-loop operation network as an optimization index until the operation of the closed-loop operation network is most economical. According to the method, the node power flow distribution model constructed by the neural network is combined with the economic operation optimization method in the closed-loop operation state, the power flow distribution model of each node of the network is obtained under the co