WIND ENERGY FORECASTING: BY USING ARTIFICIAL NEURAL NETWORK - GENETIC ALGORITHM

As wind is an intermittent generation resource and weather changes can cause large and rapid changes in output, system operators will need accurate and robust wind energy forecasting systems in the future. Rapid changes of wind generation relative to load require rapid dispatching of generation and transmission resources to balance generation versus load, regulate voltage and frequency, and maintain system performance within the limits established by National Grid. Wind energy forecasts can help the energy network operator to anticipate rapid changes of wind energy generation versus load and to make the decisions. The study has been done for ANN and also with the combination of ANN and GA for short term wind power forecasting of wind power plants. The performance of these developed forecasting models have been tested and analyzed with wind power data available from the operational records of wind power plants. The results show that the combination of ANN and GA model does wind power output forecasting very well except during the gust. This forecasting model can also be implemented in different time-scales, which will help wind energy trading in the open electricity markets.

Dr Mohan Kolhe is with School of Engineering, Physics and Mathematics of the University of Dundee, UK. E-mail: M.L.Kolhe@dundee.ac.uk Tzu Chao Lin is with the Taiwan Power Company, Taiwan. E-mail: u772020@taipower.com.tw Dr Jussi Maunuksela is with the University of Jyvaskyla, Finland. E-mail: jussi.o.maunuksela@jyu.fi