关键词: 深度学习, 时序预测, 风力发电, 长短期记忆神经网络

Abstract:

Wind power generation process has strong randomness, which leads to low accuracy of wind power prediction. In view of the above phenomenon, a wind power generation power prediction method based on deep learning algorithm was proposed. Taking the historical wind power data as input, a wind power prediction model was established to realize the wind power prediction on a time scale in the future. The results of the example show that compared with the traditional time-series prediction method, the average absolute error of the wind power prediction results based on long short-term memory neural network is smaller in each index, which verifies the feasibility and effectiveness of the above method in wind power generation prediction, and improves the accuracy of wind power generation prediction.

Key words: deep learning, time-series prediction, wind power generation, long short-term memory (LSTM) neural network