Forecasting of Day-Ahead Wind Speed/electric Power by Using a Hybrid Machine Learning Algorithm

• The amount of energy that has to be delivered for the following day is currently predicted by power system operators using day-ahead load forecasts. With the use of this forecast, generation resources can be committed a day in advance, some of them may require several hours’ notice to be ready to produce power the following day. In order to determine how much wind power will be available for each hour of the following day, power systems with large penetrations of wind generation rely on day-ahead predictions. The main objective of this study is to improve the day-ahead forecasting of wind power by improving the forecasting method using machine learning. A hybrid approach, which combines a mode decomposition method, Empirical Mode Decomposition (EMD), with Support Vector Regression (SVR), is used. The results suggest that using Support Vector Regression together with the hybrid method, which includes the Empirical Mode Decomposition to predictions can improve the accuracy of predictions. Higher accuracy forecasting of wind power is expected to improve the planning of dispatchable energy generation and pricing for the day-ahead power market.

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Naturresursteknik -- Energisystem (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Environmental Engineering -- Energy Systems (hsv//eng)

NATURVETENSKAP  -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Computer Sciences (hsv//eng)

TEKNIK OCH TEKNOLOGIER  -- Elektroteknik och elektronik -- Annan elektroteknik och elektronik (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Electrical Engineering, Electronic Engineering, Information Engineering -- Other Electrical Engineering, Electronic Engineering, Information Engineering (hsv//eng)

Nyckelord

grid integration

forecasting

energy market

machine learning

renewable energy

wind turbine

Wind energy

Empirical Mode Decomposition (EMD)

Publikations- och innehållstyp

kon (ämneskategori)

ref (ämneskategori)