Power Prediction Research of Wind Farm Based on LS-SVM Multi-model Modeling
Available Online February 2017.
- https://doi.org/10.2991/icmeim-17.2017.105How to use a DOI?
- LS-SVM, Multi-model, Affinity Propagation Clustering, Wind Power Prediction
- Characteristics of multiple working conditions are existed in the medium-term power prediction model of wind farm. To solve this problem, amulti-model modeling method for power prediction based on LS-SVM algorithm is presented. In the method, affinity propagation clustering algorithm is used to cluster the training samples. Then, the sub-models are trained by LS-SVM. The predicted values of the testing samples are forecasted by the sub-models after being classified by the similarity measurement. Finally, experiments of modeling and prediction are arranged by using the wind farm field data. The experiments show that the proposed method has high prediction accuracy and good generalization ability.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Bei Chen PY - 2017/02 DA - 2017/02 TI - Power Prediction Research of Wind Farm Based on LS-SVM Multi-model Modeling BT - 2017 International Conference on Manufacturing Engineering and Intelligent Materials (ICMEIM 2017) PB - Atlantis Press SP - 618 EP - 623 SN - 2352-5401 UR - https://doi.org/10.2991/icmeim-17.2017.105 DO - https://doi.org/10.2991/icmeim-17.2017.105 ID - Chen2017/02 ER -