Energy storage frequency modulation instruction prediction method and system for instruction decomposition

Abstract

The invention discloses an energy storage frequency modulation instruction prediction method and system based on instruction decomposition, and belongs to the technical field of energy storage system frequency modulation instruction prediction.The method comprises the steps that firstly, an original frequency modulation instruction sequence is obtained, and a corresponding trend change angle is calculated; then, the original frequency modulation instruction sequence is decomposed into a longitudinal trend sequence and a transverse trend sequence based on the trend change angle. Then, respectively calculating the fluctuation conditions of the two sequences; and correcting the original frequency modulation instruction sequence according to the difference between the transverse trend fluctuation and the longitudinal trend fluctuation. And finally, inputting the corrected sequence into the GRU network for prediction, thereby obtaining a predicted frequency modulation instruction sequence. According to the method, the prediction accuracy and reliability are improved by decomposing and correcting the frequency modulation instruction sequence.

Images