Abstract
A risk assessment based adaptive ultra-short-term wind power prediction (USTWPP) method is proposed in this paper. The method first extracts features from the historical data, and split every wind power time series (WPTS) into several subsets defined by their stationary patterns. A WPTS that does not match with any of the stationary patterns is then included into a subset of non-stationary patterns. Every WPTS subset is then related to a USTWPP model which is specially selected and optimized offline based on the proposed risk assessment index. For on-line applications, the pattern of the last short WPTS is first recognized, and the relevant prediction model is applied for USTWPP. Experimental results confirm the efficacy of the proposed method.
Original language | English |
---|---|
Journal | CSEE Journal of Power and Energy Systems |
Volume | 2 |
Issue number | 3 |
Pages (from-to) | 59-64 |
Number of pages | 6 |
ISSN | 2096-0042 |
DOIs | |
Publication status | Published - 2016 |
Keywords
- Wind power prediction
- Time series features
- Offline optimization
- Online matching
- Positive error vs negative error
- Error evaluation
- Risk index