07-04-2023 дата публикации
Номер: CN115935819A
Принадлежит:
The invention discloses a short-term wind power prediction method and system. The method comprises the following steps: acquiring wind power original sequence data of a wind power plant; the method comprises the following steps of: according to an improved empirical mode decomposition (EMD) algorithm based on wavelet transform, adding mean white noise into a wind power original sequence, and carrying out EMD decomposition and multi-frequency reconstruction for multiple times so as to eliminate abnormal data in original data; and dividing the processed wind power sequence data into a training set and a test set, training and verifying a pre-constructed CNN-GRU combined neural network model, and outputting a wind power predicted value according to the obtained CNN-GRU combined neural network model. According to the method, data in wind power prediction can be effectively processed, the precision of model training is improved, the model structure is optimized, a more accurate wind power prediction ...
Подробнее