Short-term Wind Speed Combined Prediction Model Based on ARIMA-GARCH and Elman

作者：Hanjie Xiao, Honglei Tang, Jianhua Zhou · 年份：2020 · DOI：10.1109/ipec49694.2020.9115166 · 被引用次数：5 · 研究领域：Energy Load and Power Forecasting、Hydrological Forecasting Using AI、Evaluation Methods in Various Fields

In order to effectively improve the prediction accuracy, short-term wind speed combination prediction model based on ARIMA-GARCH and Elman is proposed. It adopts ARIMA model to linearly predict short-term wind speed and GARCH to improve the non-stationary series heteroscedasticity problems. What’s more, Elman neural network is used to nonlinearly predict short-term wind speed. Finally, the minimum weighting method of absolute error is used to determine the optimal linear weight of single model so as to realize short-term wind speed prediction. Case analysis shows that, compared with other prediction models, the combination model not only has significantly improved short-term wind speed prediction accuracy, but provides a certain reference to short-term wind speed prediction in relevant fields.