Power system look-ahead scheduling method based on migration type cost function approximation

The invention discloses an electric power system look-ahead scheduling method based on migration type cost function approximation, and the method comprises the following steps: S1, constructing an electric power system look-ahead scheduling model based on a cost function, taking the minimization of the scheduling cost of an electric power system as a target function, and determining the constraint condition of each parameter of a state variable and a decision variable in the cost function; s2, converting the target function according to a cost function approximation theory and considering the influence of random information, and estimating the influence of the random information on the cost function by introducing approximation parameters to obtain an approximation cost function; s3, establishing a mapping relation between the approximate parameters and the random information, performing fitting by adopting a long short-term memory (LSTM) neural network, and identifying parameters of the LSTM to obtain optima