Abstract
The potential benefits of having wind speed predictability are obvious in wind power generation (knowing wind speed ahead of time is useful in automatic power dispatch, load scheduling, and wind turbine control). A new approach for wind speed forecasting is presented in this paper. This method makes use of memorized wind speed to predict the next step wind speed. It is shown that as long as the wind speed variation is not extremely brisk, the proposed algorithm is able to predict the wind speed with a fairly good precision. The proposed predicting scheme does not assume the precise knowledge of the wind model. The result is verified via computer simulation.
| Original language | English |
|---|---|
| Pages (from-to) | 35-47 |
| Number of pages | 13 |
| Journal | Wind Engineering |
| Volume | 24 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - Jan 2000 |
| Externally published | Yes |
Keywords
- Convergence
- Memory-based predictor
- Simulation
- Wind power generation
- Wind speed forecast