In order to solve the problem of low accuracy of wind power prediction caused by wind speed uncertainty and volatility

this paper proposes a VMD-ISSA-GRU combination model based on variational mode decomposition(VMD)

improved sparrow search algorithm(ISSA) and gated recurrent neural network(GRU).Firstly

the center frequency method is used to determine the number of modal components after VMD decomposition

which can effectively avoid over-decomposition or insufficient decomposition. Then

chaotic mapping

nonlinear decreasing weights and a mutation strategy are introduced to improve the sparrow search algorithm to optimize the gated recurrent neural network

and then an ISSA-GRU prediction model is established for each decomposed subsequence. Finally

the predicted value of each subseries is superimposed and the final predicted value is obtained. The experimental results show that

the mean absolute error

mean absolute percentage error and root mean square error of the VMD-ISSA-GRU model are 1.211 8

1.890 0 and 1.591 6 MW

respectively. Compared with the conventional GRU

long short-term memory(LSTM) neural network

Bi-directional LSTM(BiLSTM) neural network model and other combination models

the prediction accuracy has been significantly improved

which can solve the problem of low prediction accuracy of wind power.