Power system dynamic element modeling method and device based on differential neural network

The invention provides a power system dynamic element modeling method and device based on a differential neural network. The method comprises the following steps: determining a state variable of a to-be-modeled power system; acquiring a system data set; wherein a system data set is obtained through a measurement or simulation experiment, and the system data set comprises a state variable, an input variable and an output variable which are used for model training; training a differential neural network model according to the system data set to obtain a dynamic element model of the power system; the power system dynamic element model comprises differential equation expression and algebraic equation expression; the differential equation expression is used for describing state information of the power system, and the algebraic equation expression is used for describing output information of the power system. The differential neural network model is trained through the system data set, derivation operation is easy