REAL-TIME ECONOMIC DISPATCH METHOD OF POWER SYSTEM

摘要

The present invention relates to a real-time economic dispatch method of a power system comprising the following steps: determining system initial data, and describing the power system in forms of nodes, branches, and parameters; determining an optimization objective and a constraint condition according to a problem, to construct a real-time economic dispatch model; setting a topology structure between the nodes; constructing a neural network for each node, and setting a parameter; setting an initial variable of the neural network; performing optimization by means of the neural network; determining whether a termination condition is satisfied, and outputting a result when the termination condition is satisfied, or repeating the neural network optimization when the termination condition is not satisfied. Compared with an existing discrete distributed algorithm, the method provided by the present invention significantly reduces a calculation time; and compared with an existing continuous distributed algorithm, the method can consider both a global equation constraint and a global inequation constraint in the economic dispatch, thereby improving feasibility of a calculated real-time economic dispatch result.