風力発電のための広域気象データによる風速時系列予測の検討 Study on Forecast of Time Series of Wind Velocity for Wind Power Generation by Using Wide Meteorological Data
This paper describes an application of a neural network for forecasting to time variation of wind velocity. We discuss the forecasted results of wind velocity of 4m/s or more on the day as a case study. The neural network is used to forecast the wind velocity and the pattern matching is used to choose the training data of the neural network. It is found from our investigations that forecasting accuracy of the time series of wind velocity is improved by utilization the pattern matching of the weather map data. From the power generation simulation result, it is confirmed that the error of wind velocity greatly affects the power output. Therefore, the prediction accuracy of wind velocity is important.
- 電気学会論文誌. B, 電力・エネルギー部門誌 = The transactions of the Institute of Electrical Engineers of Japan. B, A publication of Power and Energy Society
電気学会論文誌. B, 電力・エネルギー部門誌 = The transactions of the Institute of Electrical Engineers of Japan. B, A publication of Power and Energy Society 128(2), 416-422, 2008-02-01
The Institute of Electrical Engineers of Japan