A combined forecasting system based on statistical method, artificial neural networks, and deep learning methods for short-term wind speed forecasting. (15th February 2021)

• Record Type: Journal Article
• Authors: Jiang, Ping; Liu, Zhenkun; Niu, Xinsong; Zhang, Lifang
• Is Part Of: Energy. Volume 217(2021)
• Publication Date: 2021-02-15
• Subjects: Artificial intelligence -- Combined forecasting system -- Data preprocessing -- Sub-model selection strategy -- Wind speed forecasting; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2020.119361
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Wind speed forecasting is gaining importance as the share of wind energy in electricity systems increases. Numerous forecasting approaches have been used to predict wind speeds. However, considering the differences in wind speed time-series, there is no universal approach that has proven to be accurate under all circumstances. In our study, a combined prediction system is proposed, which consists of four parts: optimal sub-model selection, point prediction based on a modified multi-objective optimization algorithm, interval forecasting based on distribution fitting, and forecasting system evaluation. The developed combined system integrates the merits of the sub-models and provides accurate point and interval forecasting performance. The experimental results reveal that the proposed combined forecasting system can provide effective wind speed point and interval forecasts. The absolute percentage error values of the proposed system for point forecasting are 2.9220%, 3.1696%, and 4.8358% at Site 1 and 2.2719%, 2.5882%, and 3.4799% at Site 2 for one-, two-, and three-step forecasts, respectively. Therefore, the proposed system is deemed more useful for the scheduling and management of electric power systems than other benchmark models. Highlights: A combined forecasting system is developed for short-term wind speed forecasting. A comprehensive indicator is proposed to determine the best sub-models. Point and interval forecasting of wind speed are executed for wind farm management. A modified multi-objective optimization is used to determine the optimal weight. Distribution functions is adopted to measure wind speed characteristic.